High-penetration new energy grid dispatching method and system considering deep peak regulation

Abstract

The present invention proposes a high-penetration new energy grid dispatching method and system that takes deep peak regulation into account, including: constructing a rolling dispatching model that takes into account wind power participation in the deep peak regulation market, the model including a day-ahead dispatching model and an intraday dispatching model; the objective function of the day-ahead dispatching model is to minimize the sum of the unit start-up and shutdown costs, operating costs, and deep peak regulation resource dispatching costs, taking into account various day-ahead related constraints; the objective function of the intraday dispatching model is to minimize the total operating cost, taking into account various intraday constraints; the rolling dispatching model is used to enable wind power to participate in dispatching at all levels, and coordinate day-ahead and intraday real-time dispatching. The rolling dispatching model of the present invention can make full use of intraday and real-time wind power and load forecast results to improve the accuracy of decision-making. Compared with the day-ahead dispatching model, the rolling dispatching model can replenish spare capacity from the market in a timely manner, thereby reducing the total operating cost.

Description

Technical Field

[0001] The present invention belongs to the technical field of power grid dispatching, and in particular relates to a high-penetration new energy power grid dispatching method and system considering deep peak regulation. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] Under the current on-grid electricity price formation mechanism, peak shaving is a critical task shared by all power generation companies, tracking load and achieving a balanced generation and consumption. With the increasing penetration of renewable energy, peak shaving must not only accommodate cyclical load fluctuations but also the volatility of renewable energy output. However, the costs associated with this incremental demand for peak shaving are currently not fully compensated in most regions of China.

[0004] Mature electricity markets abroad rely on spot electricity markets to match supply and demand, determine generator output plans, and solve peak load regulation problems through market competition to form sequential electricity prices. To better adapt to the changing trends in sequential electricity prices in this market environment, researchers are experimenting with combining more controllable hydropower, energy storage devices, and even demand-side resources with renewable energy generation. This approach, by changing the temporal distribution of overall power generation output, can yield better economic benefits.

[0005] Because China has yet to establish a comprehensive electricity spot market system, peak-shaving responsibilities are currently primarily distributed among conventional power sources through planned means. Unlike regions with established electricity spot markets, peak-shaving in China focuses more on reducing power generation during periods of low load. Due to a lack of fast-scaling resources like gas-fired units in my country's power resources, and the inability of coal-fired power plants to frequently start and shut down during day-ahead market clearing, deep peak-shaving remains a much-needed ancillary service. With the development of the spot market, deep peak-shaving is being integrated with the day-ahead electricity market, with thermal power plants submitting and quoting deep peak-shaving gears on a day-ahead basis. When renewable energy generation is high, the system will utilize the reduced peak-shaving capacity of thermal power plants. However, renewable energy currently does not participate in deep peak-shaving. To accommodate this, excessive deep peak-shaving resources may be deployed, increasing the total electricity purchase cost. Therefore, introducing renewable energy into the paid peak-shaving market is an inevitable development trend in the peak-shaving ancillary service market.

[0006] Peak shaving has clearer responsibilities and beneficiaries, and with the diversification of power generation energy sources, peak shaving increasingly reflects the system's need for increased flexibility. Current research on peak shaving compensation mechanisms primarily relies on cost analysis and statistics to arrive at a relatively reasonable compensation standard, primarily based on the economic characteristics of hydropower and thermal power units. In addition to cost-based approaches, some research explores peak shaving rights trading mechanisms based on the universal responsibility of peak shaving, which to some extent reflects the system's demand for effective peak shaving. Furthermore, based on the analysis of peak shaving responsibilities, the demarcation between uncompensated and paid peak shaving is also a question worth considering.

[0007] In some cases, renewable energy can also become a provider of peak load regulation, by actively reducing output and other methods to avoid traditional energy units such as thermal power being forced to lower their load rates or even shut down during short-term load troughs, causing additional economic losses and safety risks.

[0008] Due to the uncertainty of power generation output, renewable energy represented by wind power is usually considered as peak-shaving consumers. When the peak-shaving capacity of conventional units is exceeded, they can only passively accept dispatching instructions to abandon wind and solar power, and insufficient consideration is given to the economic efficiency of their operation. Summary of the Invention

[0009] To overcome the deficiencies of the above-mentioned prior art, the present invention provides a high-penetration new energy grid dispatching method that takes into account deep peak regulation, realizes the coordination of day-ahead, intraday and real-time dispatching, and improves the economy of wind power and the overall system.

[0010] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:

[0011] First, a high-penetration new energy grid dispatching method considering deep peak regulation is disclosed, including:

[0012] Constructing a rolling dispatch model that takes into account wind power participation in the deep peak-shaving market, including a day-ahead dispatch model and an intraday dispatch model;

[0013] The objective function of the day-ahead scheduling model is to minimize the sum of the unit start-up and shutdown costs, operating costs, and deep peak-shaving resource scheduling costs, taking into account various day-ahead related constraints;

[0014] The objective function of the intraday scheduling model is to minimize the total operating cost, taking into account various constraints within the day;

[0015] The rolling dispatch model is used to enable wind power to participate in dispatch at all levels, and coordinate day-ahead and intra-day real-time dispatch.

[0016] As a further technical solution, the day-ahead scheduling model is executed in the day-ahead market, the intraday scheduling model of the present invention is executed in the intraday market, and the rolling scheduling model is executed in the real-time scheduling process.

[0017] As a further technical solution, the operation arrangement of slow-machine units and the purchase of deep peak-shaving resources on the day before obtained by the day-ahead dispatching model will be used as the input of the intraday dispatching model; the operation arrangement of fast-machine units and the purchase of deep peak-shaving resources on the day before obtained by the intraday dispatching model will be used as the input of the rolling dispatching model.

[0018] As a further technical solution, the constraints of the day-ahead scheduling model include power balance constraints, line flow constraints, the system's need to meet spinning reserve demand constraints and ramping requirements, output constraints after wind turbines participate in peak regulation, upper and lower limit constraints on thermal power unit output, units' need to meet start-stop constraints, conventional unit output ramping constraints considering deep peak regulation, unit benchmark output limits participating in the peak regulation market, constraints between the upward / downward ramping provided by the unit at the previous moment and the downward / upward ramping provided by the unit at the next moment, constraints considering the deep peak regulation unit's participation in start-stop peak regulation, and constraints between start-stop variables and state variables.

[0019] As a further technical solution, the intraday scheduling in the intraday scheduling model is rolling scheduled according to the scheduling cycle, and in each scheduling cycle, the scheduling step size and the number of scheduling periods are set.

[0020] As a further technical solution, the constraints of the intraday scheduling model include: intraday power balance constraints, intraday traditional slow machine operation constraints, intraday fast machine operation constraints, day-ahead clearing unit standby output adjustment constraints, intraday wind turbine standby output constraints, line flow constraints, upper and lower limit constraints on thermal power unit output, and rotating reserve constraints that the system needs to meet.

[0021] As a further technical solution, during day-ahead dispatch, the deep peak-shaving resource call price and predicted power generation are calculated and submitted, as well as the output price, deep peak-shaving price, power generation capacity, deep peak-shaving segment capacity, ramp rate, minimum start-up and shutdown time, and minimum power generation submitted by other units;

[0022] The dispatcher arranges the day-ahead output according to the objective function formula and constraint condition formula of the intraday dispatch model based on the reported call prices, call capacities, unit parameters, and load forecast values ​​of all units. The calculated result is the dispatcher sending the following information to the units 24 hours in advance for each 24-hour period: the start and stop and output of each slow unit, the start and stop and output of each fast unit, the grid-connected power generation of each wind turbine, the standby arrangement of each fast and slow unit, and the deep peak load arrangement. Based on the calculation results;

[0023] The unit sends call information to the unit in the form of scheduling.

[0024] As a further technical solution, in intraday scheduling, the unit call volume scheduled on the previous day is regarded as a resource that can be used within the day;

[0025] When the total daily power consumption deviation is positive and small, the power generation instruction dispatched to the unit is the superposition of the output call amount scheduled on the previous day and the reserve call amount scheduled on the previous day.

[0026] When the deviation is positive and large, a new unit needs to be started;

[0027] When the total electricity consumption deviation is negative, the deep peak-shaving capacity scheduled the day before will be used. If the deep peak-shaving capacity scheduled the day before is insufficient, new deep peak-shaving resources will need to be added within the day.

[0028] The dispatch output value is the following information transmitted to the unit 2 hours in advance with a dispatch step of 15 minutes and 8 dispatch periods: slow unit output call, fast unit start and stop schedule, fast unit output call, intraday deep peak regulation supplementary schedule, and standby schedule.

[0029] Secondly, a high-penetration new energy grid dispatching system considering deep peak regulation is disclosed, including:

[0030] A rolling dispatch model construction module is configured to: construct a rolling dispatch model that takes into account wind power participation in the deep peak-shaving market, the model including a day-ahead dispatch model and an intraday dispatch model;

[0031] The objective function of the day-ahead scheduling model is to minimize the sum of the unit start-up and shutdown costs, operating costs, and deep peak-shaving resource scheduling costs, taking into account various day-ahead related constraints;

[0032] The objective function of the intraday scheduling model is to minimize the total operating cost, taking into account various constraints within the day;

[0033] The dispatching module is configured to: utilize the rolling dispatching model to enable wind power to participate in dispatching at all levels, and coordinate day-ahead and day-intraday real-time dispatching.

[0034] One or more of the above technical solutions have the following beneficial effects:

[0035] The rolling dispatch model proposed in the present invention, which takes into account the participation of wind power in the deep peak-shaving market, allows wind power to participate in market quotations, so that the ability of wind power to provide peak-shaving resources can be fully utilized and participate in dispatch at all levels, thus realizing the coordination of day-ahead, intraday and real-time dispatch and improving the economic efficiency of wind power and the system as a whole.

[0036] By adopting a day-ahead, intraday, and real-time rolling scheduling model, the present invention increases the supply of peak-shaving resources with wind power resources, fully replacing the peak-shaving resources provided by traditional units with higher quotations. The market no longer needs to call on expensive peak-shaving resources to absorb more wind power, but can abandon wind power at a lower price, thereby reducing the total scheduling cost of the system; and wind power can obtain higher profits from paid wind abandonment.

[0037] The rolling dispatch model of the present invention can fully utilize the intraday, real-time wind power and load forecast results to improve the accuracy of decision-making. Compared with the day-ahead dispatch model, the rolling dispatch model can replenish spare capacity from the market in a timely manner, thereby reducing the total operating cost.

[0038] Advantages of additional aspects of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0040] Figure 1 A structural diagram of a PJM5-bus system according to an embodiment of the present invention;

[0041] Figure 2(a)-Figure 2(b) A schematic diagram of load and wind power prediction curves according to an embodiment of the present invention;

[0042] Figure 3 This is a schematic diagram of the market clearing situation when wind power does not participate in the ancillary service market according to an embodiment of the present invention;

[0043] Figure 4 Schematic diagram of market clearing when wind power participates in the peak load regulation market according to an embodiment of the present invention;

[0044] Figure 5(a)-Figure 5(b) This is a schematic diagram of the peak-shaving resource purchase situation according to an embodiment of the present invention;

[0045] Figure 6(a)-Figure 6(b) Schematic diagram of marginal prices of market resources during clearing in an embodiment of the present invention;

[0046] Figure 7 This is a schematic diagram of the real-time call of market resources for 24-hour peak load regulation by wind turbines according to an embodiment of the present invention;

[0047] Figure 8 Schematic diagram of adjustment amounts of thermal power units and wind power in intraday scheduling and real-time scheduling according to an embodiment of the present invention;

[0048] Figure 9This is a schematic diagram of the call volume of thermal power units and wind power peak load regulation services in real-time scheduling according to an embodiment of the present invention;

[0049] Figure 10 This is a schematic diagram of 24-hour resource scheduling results of a traditional scheduling model according to an embodiment of the present invention;

[0050] Figure 11 Schematic diagram of the method flow of an embodiment of the present invention. DETAILED DESCRIPTION

[0051] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0052] It should be noted that the terms used herein are for describing particular embodiments only and are not intended to limit the exemplary embodiments according to the present invention.

[0053] In the absence of conflict, the embodiments of the present invention and the features thereof may be combined with each other.

[0054] Example 1

[0055] This embodiment discloses a high-penetration new energy grid scheduling method considering deep peak regulation, including:

[0056] Constructing a rolling dispatch model that takes into account wind power participation in the deep peak-shaving market, including a day-ahead dispatch model and an intraday dispatch model;

[0057] The objective function of the day-ahead scheduling model is to minimize the sum of the unit start-up and shutdown costs, operating costs, and deep peak-shaving resource scheduling costs, taking into account various day-ahead related constraints;

[0058] The objective function of the intraday scheduling model is to minimize the total operating cost, taking into account various constraints within the day;

[0059] The rolling dispatch model is used to enable wind power to participate in dispatch at all levels, and coordinate day-ahead and intra-day real-time dispatch.

[0060] At present, the bidding rules for my country's new energy participating in peak-shaving auxiliary services are the same as those for thermal power units. Renewable energy sources such as wind power and photovoltaics submit their bids for peak-shaving auxiliary services. The bids can be step-like, but in order to ensure that the market clearing model is a convex problem, the bids must be increasing.

[0061] Currently, there are two viable models. One is similar to thermal power, where renewable energy sources submit bids for reduced output. The economic significance of this is that renewable energy sources are willing to pay for wind curtailment and proactively reduce output to reduce the cost of deep peak regulation. For example, if a wind farm (50-100MW) offers a bid for reduced peak regulation of 0.2 yuan / kWh, this means that when unpaid peak regulation resources have been used and paid peak regulation is needed, the wind farm is willing to voluntarily curtail wind power at 0.2 yuan / kW, at which point the output of other units will be equal to the load.

[0062] The second method is the Shanxi peak-shaving market, where renewable energy sources submit capacity-price bids to avoid curtailment of wind and solar power. For example, if a wind farm's (50-100MW) curtailment bid is 0.2 yuan / kWh, and curtailment reaches the 50-100MW range, the wind farm is willing to reduce the price by 0.2 yuan / kWh for this portion of generated power. If the market energy clearing price is 0.5 yuan / kWh and the settlement price for curtailment avoidance is 0.2 yuan / kWh, the settlement price for this portion of the winning bid is 0.3 yuan / kWh. (If the wind farm does not participate in the market, it will be directly curtailed, and the settlement price will be 0.) When renewable energy curtailment occurs in the system, the higher the curtailment avoidance price bid, the cheaper the electricity and the higher the priority for grid access. The difference between the actual curtailment of renewable energy and the amount that should be curtailed is the avoided curtailment of renewable energy. The deep peak-shaving market price is the average of the marginal price of thermal power units and the marginal price of renewable energy units. At this point, the total output of the wind farm and other units is equal to the load.

[0063] The quotation structure of renewable energy output reduction and peak load regulation service is as follows:

[0064] Fixed costs: The fixed costs of peak load shaving services refer to the additional equipment costs required to provide peak load shaving, such as wind turbines. This cost is calculated based on the investment cost and the annual depreciation factor:

[0065]

[0066] Where ρ dps A is the unit fixed cost in the peak-shaving cost; y is the annual depreciation of the equipment; A mt and S avet The annual maintenance cost and personnel cost increased due to the peak load shaving task; P ps is the unit's expected annual peak-shaving service bid volume; κ is the unit's historical peak-shaving service winning probability.

[0067] Opportunity cost: According to the dispatch, when the generator set is running at reduced output for peak load regulation, the installed capacity cannot fully participate in the electricity market due to peak load regulation, and the opportunity to profit from the electricity market is lost. For example, at a certain moment, the generator set is running at reduced output for peak load regulation, and the actual power generation output is P i,t, and the rated generating capacity of the unit is P i,N Therefore, the power generation loss of this unit during this period is ΔP=(P i,N -P i,t )·Δt, if the grid-connected electricity price of this type of unit is ρ clear,t , the opportunity cost of unsold electricity generation is:

[0068]

[0069] Clearing Rules: Currently, some in the industry believe that the introduction of deep peak shaving will cause the unit bid curve in the day-ahead market to first decline and then rise, thereby rendering the market-clearing model non-convex. Therefore, addressing this issue through mechanism design and model construction, and ensuring the practicality of day-ahead market clearing, is key to integrating deep peak shaving with the day-ahead market. Some also believe that after spot trading is implemented in the future, day-ahead market trading could be initiated first, followed by deep peak shaving trading. If the day-ahead market and deep peak shaving are cleared separately, unit output must be deep peaked based on the day-ahead market clearing results. As units ramp up from minimum output, some units that could previously perform deep peak shaving during off-peak periods will lose this ability due to unit ramping constraints. Consider a 600MW unit with a ramping constraint of 150MW / h. When clearing the day-ahead energy market and the deep peak-shaving market separately, the day-ahead market clears 300 MW in period 1 and 450 MW in period 2. During deep peak-shaving transactions, assuming a low unit bid, 50 MW of deep peak-shaving capacity can be achieved. However, due to ramping constraints, the unit loses its deep peak-shaving capacity in period 1. When clearing the day-ahead and deep peak-shaving markets together, the clearing result is 300 MW cleared from the day-ahead market in period 1, 50 MW cleared from the deep peak-shaving market, and 400 MW cleared from the energy market in period 2, satisfying the ramping constraints. Therefore, the current sequential scheduling approach cannot achieve optimal resource allocation. Combining the day-ahead market with deep peak-shaving can expand the scope for optimal resource allocation. For example, the PJM market in the United States jointly dispatches power and reserve.

[0070] Day-ahead scheduling model:

[0071] In order to generate electricity, the unit needs to obtain dispatch instructions from the dispatch center. If these parameters are not provided, the dispatch center will not include the unit in the dispatchable resources.

[0072] Objective function

[0073] min f=C G,str +C G,ope +C G,dps (3)

[0074] Where: C G,str 、CG,ope 、C G,adj The startup and shutdown costs, operating costs, and deep peak-shaving dispatching costs are calculated by adding up the startup and shutdown costs, operating costs, and deep peak-shaving dispatching costs for each unit. The startup costs, operating costs, and deep peak-shaving ancillary service costs for a single unit within a 24-hour operating cycle are calculated as follows.

[0075]

[0076]

[0077]

[0078] Where: a i 、b i 、c i are the quadratic coefficient, linear coefficient and constant coefficient of the power generation cost curve of unit i, c i is the unit startup cost, u i is the unit startup cost, is the k-th deep peak load quotation of unit i. The above four constant parameters are derived from the declared information of each dispatchable unit and are known quantities of the dispatch model. T The value is 24. C G,str,i is the startup cost of unit i in one operating cycle, is a 0-1 variable indicating whether unit i is started at time t, for A 0-1 variable indicating whether unit i has a shutdown action at time t is the output variable of unit i at time t; C G,ope,i is the operating cost of unit i in one operating cycle; C G,dps,i is the deep peak load regulation auxiliary service cost of unit i in one operation cycle, and is the purchase price (constant, unit declaration information) and purchase quantity (variable) of the deep peak-shaving service of unit i at time t, It is a 0-1 variable indicating whether the service of the k-th segment of deep peak regulation of unit i at time t is successful, and K is the number of quotation segments.

[0079] Constraints:

[0080] ① Power balance constraint. The unit output arrangement must meet the predicted load demand, as shown in Equation (7). The predicted load demand is predicted by the dispatch center based on historical electricity consumption data, weather information, and other information, and is a known quantity in the model.

[0081]

[0082] Where: is the base output arrangement of unit i at time t, is the predicted load output of node m at time t. N G is the total number of units (including thermal power and wind power). M is the total number of loads.

[0083] ② Line power flow constraint. Assuming that the system has sufficient reactive power, only active power flow is considered and the line power flow is calculated based on the DC power flow, see formula (8).

[0084]

[0085] Where: The maximum active power allowed to flow through line L after considering the margin. This value is determined by the line model, operating age, temperature, etc. and is a known quantity in the model. G is the power transfer distribution factor matrix, which is determined by the grid structure and is a known quantity. m,L With G i,L are the power transfer factors corresponding to line L and node i and node m in the matrix respectively.

[0086] ③ The system needs to meet the spinning reserve demand constraints and ramping requirements, see (9)-(13).

[0087]

[0088]

[0089]

[0090]

[0091]

[0092] This constraint is for conventional units such as thermal power plants, whose output can be controlled, and reserves need to be arranged to cope with the uncertainty of wind power and load. is the number of thermal power units, including fast units that start and stop quickly and slow units that start and stop slowly. The system's positive and negative spare capacity requirements are set by the dispatch center based on historical experience and are known quantities in the model; is the operating state of unit i at time t, a 0-1 variable, 0 for shutdown and 1 for operation; To adjust the system's spinning reserve requirements upward or downward at time t, the dispatch center sets it based on the forecast load and wind power output, which is a known quantity in the model. is the ramp-up and ramp-down rate variable that unit i can provide at time t, which is related to the unit's own characteristics and the amount of money it clears in the deep peak-shaving market. Equation (13) shows that deep peak-shaving requires that the bids of different segments of the operating units be cleared in sequence, which is the current bidding rule in the deep peak-shaving market.

[0093] ④ The output constraint of the wind turbine after participating in peak load regulation is:

[0094]

[0095] in, is the bid amount of wind power in the energy market, Contribute to wind power forecasting. is a 0-1 variable indicating whether the bid is successful in the wind power deep peak regulation market. The winning bid amount for wind power in the deep peak-shaving market cannot exceed the amount of wind power connected to the grid, as deep peak-shaving is achieved through paid wind curtailment.

[0096] ⑤ Upper and lower limits of thermal power unit output, see formula (16)

[0097]

[0098] Where: are the maximum and minimum technical outputs of unit i respectively, and the output of the unit should be within this range.

[0099] ⑥ The unit needs to meet the start-stop constraints, see equations (16) and (17);

[0100]

[0101]

[0102] Formula (16) indicates that the unit startup time must not be less than the minimum startup time T ON , formula (17) indicates that the unit downtime must not be less than the minimum downtime T OFF .

[0103] ⑦ Considering the output ramping constraints of conventional units for deep peak regulation

[0104]

[0105]

[0106]

[0107] Where, and is the maximum up and down ramp rate that unit i can provide at time t. It is a fixed parameter related to unit performance and is a known quantity. Formulas (18) and (19) indicate that the unit's up and down ramp rate must not exceed the maximum up and down ramp rate capability. Formula (20) indicates that the unit's current output and the ramp rate it provides are limited by the unit's maximum output.

[0108] ⑧ Baseline output limit of units participating in the peak load regulation market. Related constraints are shown in equations (21)-(22).

[0109]

[0110]

[0111] in, It is the free part of the peak load shaving service. is the set of units participating in the deep peak-shaving market. Formula (21) indicates that the ramp-down of units participating in deep peak-shaving cannot exceed the available capacity for deep peak-shaving. Formula (22) indicates that the output provided by deep peak-shaving units after ramp-down must be greater than the minimum output allowed by the units.

[0112] ⑨ Consider the constraints between the upward / downward ramp provided by the unit at the previous moment and the downward / upward ramp provided by the unit at the next moment, see formula (23)

[0113]

[0114] Where: is the maximum ramp rate of unit i in 1 min, which is a known constant. 60 Indicates that the interval between time periods is 60 minutes.

[0115] ⑩ Considering the constraints of deep peak-shaving units participating in start-stop peak-shaving, see formula (24).

[0116]

[0117] Where, It is a 0-1 variable indicating whether unit i has a shutdown action at time t.

[0118] The constraints between the start and stop variables and the state variables are (25)(26):

[0119]

[0120]

[0121] Intraday scheduling model:

[0122] Intraday scheduling refers to a rolling schedule with a scheduling cycle of 2 hours, which is executed 12 times in 24 hours a day. In each scheduling cycle, the scheduling step is 15 minutes, the number of scheduling periods is 8, and the scheduling objective function minimizes the total operating cost, that is:

[0123] Objective function:

[0124] min f=C ope,Δ +C fast,st +C fast,ope +C adj,Δ (26)

[0125] Where: C ope,Δ is the cost of calling the unit for clearing the day ahead, C fast,st 、C fast,ope are the startup and operation costs of the daily quick start unit, C adj,Δ It is the intraday standby cost.

[0126]

[0127] Where, The unit call cost corresponding to the difference between the call output of unit i within the day and the scheduled output on the day before; It is the difference between the unit’s intraday dispatch quantity and the day-ahead dispatch quantity. T The number of time periods for intraday scheduling, the value is 8.

[0128]

[0129]

[0130] In order to cope with the difference between the day-ahead forecast and the intraday forecast, it is necessary to rearrange the start and stop of the fast start unit. fast,st is the start-up and shutdown cost of the daily express machine, C fast,ope is the daily operating cost of the express machine. k 、 a k 、b k 、c k They are respectively the startup cost, shutdown cost, quadratic term coefficient, linear term coefficient and constant coefficient of the operating cost of fast machine k, all of which are known quantities. is a 0-1 variable indicating whether unit k is started at time t, N is the output of unit k scheduled for the day at time t, which is a variable. FG is the number of units that can be quickly adjusted, which is a known quantity.

[0131]

[0132] Where, d iThe unit's intraday deep peak-shaving quotation is a known quantity; The target value (variable) of intraday deep peak load regulation arranged by scheduling; d ju d jd The daily upward and downward peak-shaving quotations submitted by wind power based on new forecast data are known quantities. It is the bid amount (variable) for the wind turbine’s daily output increase and downward peak regulation.

[0133] Constraints:

[0134] Similar to day-ahead dispatch constraints, intraday dispatch needs to meet operational constraints, including: intraday power balance constraints, line flow constraints, upper and lower output limits of thermal power units, and spinning reserve constraints that the system needs to meet.

[0135] ① Intraday power balance constraint.

[0136]

[0137]

[0138]

[0139]

[0140] The intraday power balance constraint requires that the output of the units scheduled during the day be equal to the load forecast value. The output after the unit is adjusted. It is the output arranged the day before, which, when added to the output adjusted during the day, becomes the dispatch instruction issued by the system dispatcher to the unit during the day. is the daily net load forecast value, They are daily load and wind power forecast output respectively. t , ΔW t It is the load shedding and wind curtailment variable.

[0141] ②Constraints on traditional slow-speed operation during the day:

[0142]

[0143]

[0144] Where: are the maximum and minimum technical outputs of unit i, The start and stop schedule of the slow-moving units obtained by solving the day-ahead dispatch model (the value solved in the day-ahead model) is a known quantity in the intraday dispatch model. is the output of the adjacent stages of the unit (each 15 minutes is a stage, T 15 ) is the ramp rate of unit i.

[0145] ③Intraday express operation constraints:

[0146]

[0147]

[0148]

[0149]

[0150] Where: are the maximum and minimum technical outputs of fast machine k respectively. is the 0-1 variable of the start and stop status of the fast machine k at time t.

[0151] ④ Constraints on standby output adjustment of units cleared on the previous day:

[0152]

[0153]

[0154]

[0155] Formula (39) indicates that the deviation between the unit's daily output and the day-ahead dispatch arrangement cannot exceed the maximum value of the upward and downward adjustment amount, where the downward adjustment amount is formula (40), indicating that it cannot exceed the sum of the day-ahead and intraday deep peak-shaving arrangement capacity. Formula (41) is the dispatch price for this deviation. When the deviation is positive, the dispatch price is the intraday power generation energy quotation. (known quantity); when the deviation is between the deep peak-shaving amount scheduled the day before and 0, no additional payment is required and the unit cost of the call is 0; when the absolute value of the negative deviation is larger than the deep peak-shaving amount scheduled the day before, it is necessary to supplement deep peak-shaving resources within the day, and the price is the quoted price of deep peak-shaving resources within the day. (known quantity).

[0156] The intraday traditional slow machine operation constraints, intraday fast machine operation constraints, and intraday fast and slow machine ramp rate requirement constraints are the same as the day before, except that the time scale is shortened to a step size of 15 minutes and the scheduling cycle is 2 hours.

[0157] ⑤ In addition, it is necessary to add the following daily wind turbine standby output constraints:

[0158]

[0159]

[0160] Where: The wind turbine adjusts its capacity bid upward / downward within the day according to the positive and negative deviations between the new output forecast and the day-ahead market clearing volume.

[0161] The calling model of the real-time scheduling stage is the same as the existing scheduling model.

[0162] The process of day-ahead, intraday, and real-time rolling scheduling is as follows:

[0163] In the day-ahead dispatch, wind turbines calculate and submit the deep peak-shaving resource call price and predicted power generation according to formulas (1) and (2). The output (power generation) price, deep peak-shaving price, power generation capacity, deep peak-shaving segment capacity, ramp rate, minimum start-up and shutdown time, minimum power generation, and other information submitted by other units are the same as the current dispatch method used by the power grid company. The dispatcher arranges the day-ahead output according to the objective function formula (1) and constraint formulas (7)-(26) based on the call price, call capacity, unit parameters, and load forecast values ​​reported by all units. The calculated result is that the dispatcher sends the following information to the units for each 24-hour period 24 hours in advance: the start and stop and output of each slow unit, the start and stop and output of each fast unit, the online power generation of each wind turbine, the standby arrangement quantity of each fast and slow unit, and the deep peak-shaving arrangement quantity. According to the calculation results, the unit sends the call information to the unit in the form of dispatch. Since it needs to pay the corresponding price to it, it can also be regarded as purchasing the above resources from the unit.

[0164] In intraday dispatch, the day-ahead scheduled capacity is considered available for intraday use. However, due to the uncertainty of load and wind power output, actual wind turbine output and load demand may deviate from forecasts. Generators must purchase additional resources from generators to account for these discrepancies. Day-ahead schedules are disrupted. The start and stop schedules for slow-generating units cannot be changed due to unit characteristics. Other capacity is adjusted based on the bids submitted by the units within the day (which can differ from the day-ahead values). When the total power demand deviation is positive and small, the dispatcher's generation order is the sum of the day-ahead scheduled capacity and the day-ahead scheduled reserve capacity. When the deviation is positive and large, new units (fast units) must be activated, requiring rescheduling of the start and stop schedules of fast units and the generation of both fast and slow units. When the total power demand deviation is negative, the day-ahead scheduled deep peak-shaving capacity is used. If the day-ahead scheduled deep peak-shaving capacity is insufficient, new deep peak-shaving resources (typically more expensive) must be added within the day. The output forecast value of the wind turbine during the day is much more accurate than the day-ahead forecast value. The deviation between the day-ahead forecast value and the day-ahead forecast value can be used as a resource to be called. If the load demand is greater than the day-ahead estimate and the wind power deviation is also positive (more power generation), then the dispatcher can dispatch this part of the wind power resource; if the load demand is smaller than the day-ahead estimate and the wind power deviation is negative, the deep peak-shaving resource will be dispatched after deducting this part of the wind power deviation. The above process still uses an optimization model to solve the dispatch scheme. The objective function is formula (26) and the constraints are formulas (31)-(43). The output value is the following information transmitted to the unit 2 hours in advance with a dispatch step of 15 minutes and 8 dispatch periods: slow machine output call (day-ahead output arrangement plus day-ahead output arrangement), fast machine unit start and stop arrangement, fast machine unit output call (day-ahead output arrangement plus day-ahead output arrangement), day-ahead deep peak-shaving supplementary arrangement, and standby arrangement.

[0165] The real-time stage is every 5 minutes, and economic scheduling is performed according to the previously arranged 15-minute dispatchable resources (output arrangement, reserve, deep peak regulation and unit startup status). Similar to the existing scheduling method, the output arrangement of each unit is obtained on a 5-minute scale.

[0166] When implementing it specifically, Figure 11For the application of the present invention in the dispatching center, according to the dispatching process of the dispatching center, the dispatching center obtains the information of the trading entities in the day-ahead market (power generation cost, start-up and shutdown cost, etc.) and implements day-ahead dispatching; implements intraday dispatching 2 hours in advance according to the latest short-term load forecast results and the trading entity information (power generation cost, quotation) in the intraday market; and performs real-time dispatching in 15 minutes in real time according to the ultra-short-term load forecast results of 15 minutes with a step size of 5 minutes. The present invention jointly dispatches deep peak shaving and energy in day-ahead dispatching, intraday dispatching and real-time dispatching, executes the day-ahead dispatching model of the present invention in the day-ahead market, executes the intraday dispatching model of the present invention in the intraday market, and executes the real-time dispatching model of the present invention in the real-time dispatching process. The operation arrangement of the slow-machine unit and the purchase amount of the day-ahead deep peak shaving resources obtained by the day-ahead dispatching model will be used as the input of the intraday dispatching model; the operation arrangement of the fast-machine unit and the purchase amount of the intraday deep peak shaving resources obtained by the intraday dispatching model will be used as the input of the implementation dispatching model.

[0167] Case analysis:

[0168] This paper verifies the proposed scheduling model based on the PJM5-bus system design example. The structure of the PJM5-bus system is as follows: Figure 1 The model consists of five generators and three load nodes. The generator parameters are provided in the literature. It is assumed that bus_5 is connected to a wind turbine. The model is solved using CPLEX software.

[0169] Market Clearing Cost Analysis:

[0170] Figure 2(a)-Figure 2(b) Shown are the 48-hour forecast curves for total load and wind power. The forecast errors for wind power and load curves decrease in order during the day-ahead, intraday, and real-time scheduling phases. Assuming a 20% day-ahead forecast error for wind power, the day-ahead wind power forecast curve is the actual wind power curve plus white noise with a mean of 0 and a standard deviation of 0.2 (assuming the forecast error follows a normal distribution). Similarly, assuming the intraday and real-time forecast errors for wind power are 5% and 2%, respectively, and the day-ahead, intraday, and real-time forecast errors for load are 3%, 1%, and 0.5%, respectively, the forecast curves are the actual curves plus the corresponding white noise. Wind power has a high degree of uncertainty. To verify the effectiveness of wind turbines in peak and frequency regulation, the wind power curve for the northern region, which has strong anti-peaking characteristics, was selected.

[0171] In order to study the scheduling of all resources within a day, the following example first studies the impact of day-ahead, intraday, and real-time rolling scheduling on day-ahead market clearing. Figure 3 and Figure 4 The market clearing situation when wind power does not participate in the ancillary service market and the market clearing situation when wind power only participates in the peak load regulation market are given. It can be seen that

[0172] Figure 5(a)-Figure 5(b) The peak-shaving resource purchase situation when wind turbines participate in the peak-shaving market is given. Figure 5(a) shows the case when wind turbines do not participate, and Figure 5(b) shows the case when wind turbines participate. It can be seen that when wind turbines provide peak-shaving resources, the supply of peak-shaving resources is more sufficient, and thus replaces a considerable part of the resources provided by slow turbines.

[0173] The electricity clearing price in the power market is shown in Figure 6(a). It can be seen that the participation of wind turbines in the peak load regulation market effectively reduces the electricity price. This is because Figure 3 and Figure 4 By comparison, the entry of wind power into the peak-shaving market effectively reduces the demand for high-priced flexible regulation resources (typically, the G4 unit is withdrawn from the market between 10:00 and 12:00). The peak-shaving market clearing price is shown in Figure 6(b), which shows that the increased supply of resources reduces the peak-shaving market clearing price.

[0174] The system dispatching costs when wind turbines participate in the peak-shaving market and the revenue from the sale of electricity and ancillary service products by wind turbines are shown in Table 1. It can be seen that the participation of wind power in the peak-shaving market can effectively reduce the total dispatching costs of the market. This is because the market no longer needs to call on expensive peak-shaving resources to absorb more wind power, but can abandon wind power at a lower price, thereby reducing the total system cost; and wind power is equivalent to paid abandonment at this time, so it can obtain higher profits.

[0175] Table 1 Total system dispatch cost (48 hours) when wind turbines participate in the ancillary service market

[0176]

[0177] Table 2 Revenue of wind turbines participating in the ancillary service market (48 hours)

[0178]

[0179] Figure 7 The paper gives the configuration of electricity, standby and frequency regulation of thermal power units and wind turbines for 24 hours a day when wind turbines participate in peak regulation, as well as the purchase of peak regulation resources before and during the day.

[0180] Figure 8 The adjustment of thermal power and wind turbines during the day-ahead and intraday dispatch phases when wind turbines participate in peak load regulation is given. It can be seen that wind turbines bear a portion of the power adjustment and play an important role in maintaining system power balance.

[0181] Figure 9The data shows the peak-shaving service requests for thermal power units and wind power in real-time scheduling. It shows that 1) timely adjustments to wind power replace some of the adjustments required for thermal power. 2) Wind turbines provide a certain amount of peak-shaving capacity. 3) Due to the uncertainty of day-ahead forecasts, some peak-shaving products are awarded in the intraday market, supplementing day-ahead resources to ensure stable system operation. However, in real-time scheduling, due to the low marginal cost of wind power, wind turbine peak-shaving requests are given a lower priority to save costs.

[0182] In order to compare the impact of wind turbines participating in the ancillary service market on resource scheduling, Figure 10 24-hour resource scheduling results of the traditional scheduling model.

[0183] Comparing the dispatch results of the existing Shandong dispatch model with those of the rolling dispatch model proposed in this paper, it can be seen that the existing dispatch model utilizes far more frequency regulation and reserve capacity than the rolling dispatch model. Furthermore, due to the large total spinning reserve capacity required by the system, all power adjustments must be handled by thermal power units. Under the existing dispatch model, all system resources must be purchased in the day-ahead market, requiring a total reserve capacity of 1963.4 MW. However, under the rolling dispatch model, the total capacity required for day-ahead purchases is only 1076.8 MW. Peak-shaving resources are purchased based on the upper limit of wind power and the lower limit of load forecasts, totaling 3303.28 MW. The peak-shaving resources purchased under the rolling dispatch model are 2182.7 MW, both of which are lower than the ancillary service capacity required by the day-ahead dispatch model alone. This is because the intraday dispatch phase of the rolling dispatch model significantly improves the forecast accuracy of wind power and load compared to the day-ahead phase. Therefore, reserve capacity can be replenished at a lower price in the intraday market, significantly reducing the high compensation costs of resources dispatched in the real-time market.

[0184] Example 2

[0185] The purpose of this embodiment is to provide a computing device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the program.

[0186] Example 3

[0187] The purpose of this embodiment is to provide a computer-readable storage medium.

[0188] A computer-readable storage medium stores a computer program, which, when executed by a processor, performs the steps of the above method.

[0189] Example 4

[0190] The purpose of this embodiment is to provide a high-penetration new energy grid dispatching system that takes into account deep peak regulation, including:

[0191] A rolling dispatch model construction module is configured to: construct a rolling dispatch model that takes into account wind power participation in the deep peak-shaving market, the model including a day-ahead dispatch model and an intraday dispatch model;

[0192] The objective function of the day-ahead scheduling model is to minimize the sum of the unit start-up and shutdown costs, operating costs, and deep peak-shaving resource scheduling costs, taking into account various day-ahead related constraints;

[0193] The objective function of the intraday scheduling model is to minimize the total operating cost, taking into account various constraints within the day;

[0194] The dispatching module is configured to: utilize the rolling dispatching model to enable wind power to participate in dispatching at all levels, and coordinate day-ahead and day-intraday real-time dispatching.

[0195] This paper addresses the long-standing problem of renewable energy being considered a peak-shaving resource for consumers, preventing them from actively participating in peak-shaving. Using the ideal peak-shaving curve as a benchmark for peak-shaving contributions and the power adjustment within a unit time period as the transaction target, a peak-shaving model that considers renewable energy participation is designed. The goal is to incentivize various resources to provide peak-shaving through peak-shaving transactions and promote renewable energy consumption. A case study demonstrates the model's effectiveness and feasibility.

[0196] The steps involved in the apparatuses of Examples 2, 3, and 4 above correspond to those of Method Example 1. For detailed implementations, please refer to the relevant description of Example 1. The term "computer-readable storage medium" should be understood to mean a single medium or multiple media containing one or more instruction sets; it should also be understood to include any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and causing the processor to perform any method of the present invention.

[0197] Those skilled in the art will appreciate that the modules or steps of the present invention described above can be implemented using a general-purpose computer device. Alternatively, they can be implemented using program code executable by a computing device, which can then be stored in a storage device and executed by the computing device. Alternatively, they can be fabricated into separate integrated circuit modules, or multiple modules or steps can be fabricated into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

[0198] Although the above describes the specific embodiments of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without any creative work are still within the scope of protection of the present invention.

Claims

1. A high-penetration new energy grid dispatching method considering deep peak regulation is characterized by: include: Constructing a rolling dispatch model that takes into account wind power participation in the deep peak-shaving market, including a day-ahead dispatch model and an intraday dispatch model; The objective function of the day-ahead scheduling model is to minimize the sum of the unit start-up and shutdown costs, operating costs, and deep peak-shaving resource scheduling costs, taking into account various day-ahead related constraints; The objective function of the day-ahead scheduling model is: Where: 、 、 The cost of starting and stopping the unit, the cost of operation, and the cost of deep peak-shaving scheduling are respectively calculated by adding up the cost of starting and stopping the unit, the cost of operation, and the cost of deep peak-shaving scheduling for each unit. The deep peak-shaving auxiliary service cost of a single unit in a 24-hour operation cycle is calculated as follows: ； Where: For the crew The cost of deep peak load regulation auxiliary services in one operation cycle, and For the moment unit No. The purchase price of the deep peak-shaving service in the segment is a known quantity, and the purchase quantity is a variable; For the moment unit Deep peak shaving A 0-1 variable indicating whether the service of the segment quotation was successful. The number of quotation segments; is the number of time periods for day-ahead scheduling; The objective function of the intraday scheduling model is to minimize the total operating cost, taking into account various constraints within the day; The objective function of the intraday scheduling model is: ; Where: is the cost of calling the unit for clearing the day before, 、 are the startup and operation costs of the daily quick start unit, is the intraday standby cost; Where, The unit's intraday deep peak-shaving quotation is a known quantity; The bid amount of intraday deep peak load regulation arranged by the dispatcher is a variable; 、 The daily upward and downward peak-shaving quotations submitted by wind power based on new forecast data are known quantities. 、 The bid amount for the wind turbine's daily upward and downward peak-shaving output is a variable; The number of time periods for intraday scheduling; The rolling dispatch model is used to enable wind power to participate in dispatch at all levels, and coordinate day-ahead and intra-day real-time dispatch.

2. The high-penetration new energy grid dispatching method considering deep peak regulation according to claim 1 is characterized in that, The day-ahead scheduling model is implemented in the day-ahead market, the intraday scheduling model is implemented in the intraday market, and the rolling scheduling model is implemented in the real-time scheduling process.

3. The high-penetration new energy grid dispatching method considering deep peak regulation according to claim 1 is characterized in that: The operation schedule of slow-machine units and the purchase of deep peak-shaving resources obtained by the day-ahead dispatching model will serve as the input of the intraday dispatching model; the operation schedule of fast-machine units and the purchase of deep peak-shaving resources obtained by the intraday dispatching model will serve as the input of the rolling dispatching model.

4. The high-penetration new energy grid dispatching method considering deep peak regulation according to claim 1 is characterized in that: The constraints of the day-ahead dispatch model include power balance constraints, line flow constraints, the need for the system to meet spinning reserve demand constraints and ramping requirements, output constraints after wind turbines participate in peak regulation, upper and lower output limits of thermal power units, units that need to meet start and stop constraints, output ramping constraints of conventional units considering deep peak regulation, base output limits of units participating in the peak regulation market, constraints between the upward / downward ramping provided by the unit at the previous moment and the downward / upward ramping provided by the unit at the next moment, constraints considering the participation of deep peak regulation units in start and stop peak regulation, and constraints between start and stop variables and state variables; In the intraday scheduling model, intraday scheduling is performed in a rolling manner according to a scheduling cycle. In each scheduling cycle, a scheduling step size and a number of scheduling periods are set.

5. The high-penetration new energy grid dispatching method considering deep peak regulation according to claim 1 is characterized in that: The constraints of the intraday scheduling model include: intraday power balance constraints, intraday traditional slow machine operation constraints, intraday fast machine operation constraints, day-ahead clearing unit standby output adjustment constraints, intraday wind turbine standby output constraints, line flow constraints, thermal power unit output upper and lower limit constraints, and rotating reserve constraints that the system needs to meet.

6. The high-penetration new energy grid dispatching method considering deep peak regulation according to claim 1 is characterized in that: During day-ahead dispatch, the deep peak-shaving resource call price and predicted power generation are calculated and submitted, as well as the output price, deep peak-shaving price, power generation capacity, deep peak-shaving segment capacity, ramp rate, minimum start-up and shutdown time, and minimum power generation submitted by other units; The dispatcher arranges the output of the day ahead according to the objective function formula and constraint condition formula of the intraday dispatch model based on the call price, call capacity, unit parameters and load forecast value information reported by all units. The calculated result is that the dispatcher sends the following information to the units 24 hours in advance for each 24-hour period: the start and stop and output of each slow unit, the start and stop and output of each fast unit, the grid-connected power generation of each wind turbine, the standby arrangement quantity of each fast and slow unit, and the deep peak regulation arrangement quantity; based on the calculated result, the dispatcher sends the call information to the units in the form of dispatch.

7. The high-penetration new energy grid dispatching method considering deep peak regulation according to claim 1 is characterized in that: In intraday scheduling, the unit call volume scheduled on the previous day is regarded as the resources that can be used within the day; When the total daily power consumption deviation is positive and small, the power generation instruction dispatched to the unit is the superposition of the output call amount scheduled on the previous day and the reserve call amount scheduled on the previous day. When the deviation is positive and large, a new unit needs to be started; When the total electricity consumption deviation is negative, the deep peak-shaving capacity scheduled the day before will be used. If the deep peak-shaving capacity scheduled the day before is insufficient, new deep peak-shaving resources will need to be added within the day. The dispatch output value is the following information transmitted to the unit 2 hours in advance with a dispatch step of 15 minutes and 8 dispatch periods: slow unit output call, fast unit start and stop schedule, fast unit output call, intraday deep peak regulation supplementary schedule, and standby schedule.

8. A high-penetration new energy grid dispatching system considering deep peak regulation is characterized by: include: A rolling dispatch model construction module is configured to: construct a rolling dispatch model that takes into account wind power participation in the deep peak-shaving market, the model including a day-ahead dispatch model and an intraday dispatch model; The objective function of the day-ahead scheduling model is to minimize the sum of the unit start-up and shutdown costs, operating costs, and deep peak-shaving resource scheduling costs, taking into account various day-ahead related constraints; The objective function of the day-ahead scheduling model is: Where: 、 、 The cost of starting and stopping the unit, the cost of operation, and the cost of deep peak-shaving scheduling are respectively calculated by adding up the cost of starting and stopping the unit, the cost of operation, and the cost of deep peak-shaving scheduling for each unit. The deep peak-shaving auxiliary service cost of a single unit in a 24-hour operation cycle is calculated as follows: ； In the formula For the crew The cost of deep peak load regulation auxiliary services in one operation cycle, and For the moment unit No. The purchase price of the deep peak-shaving service in the segment is a known quantity, and the purchase quantity is a variable; For the moment unit Deep peak shaving A 0-1 variable indicating whether the service of the segment quotation was successful. The number of quotation segments; is the number of time periods for day-ahead scheduling; The objective function of the intraday scheduling model is to minimize the total operating cost, taking into account various constraints within the day; The objective function of the intraday scheduling model is: ; Where: is the cost of calling the unit for clearing the day before, 、 are the startup and operation costs of the daily quick start unit, is the intraday standby cost; Where, The unit's intraday deep peak-shaving quotation is a known quantity; The bid amount of intraday deep peak load regulation arranged by the dispatcher is a variable; 、 The daily upward and downward peak-shaving quotations submitted by wind power based on new forecast data are known quantities. 、 The bid amount for the wind turbine's daily upward and downward peak-shaving output is a variable; The number of time periods for intraday scheduling; The dispatching module is configured to: utilize the rolling dispatching model to enable wind power to participate in dispatching at all levels, and coordinate day-ahead and day-intraday real-time dispatching.

9. A computing device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method described in any one of claims 1 to 7 are executed.

Images