Time sequence random production simulation method considering flexibility demand in electricity market environment

Abstract

The invention belongs to the technical field of electric power system dispatching optimization and electric power and electric quantity balance analysis, and relates to a time sequence random production simulation method considering a flexibility demand in an electric power market environment, which comprises the following steps: 1, constructing an electric power system time sequence-random production simulation framework considering the flexibility demand in the market environment; 2, the upper layer model obtains a start-stop scheduling scheme by taking the minimization of the total operation cost of the system as a target, calculates the time sequence flexibility supply capability of each type of unit, and provides an input basis for the lower layer; 3, constructing a multi-state flexibility supply model of the unit under different time scales in a lower layer model, carrying out multi-time scale system flexibility supply and demand probabilistic matching by adopting an equivalent electric quantity function method, and realizing flexibility index evaluation under a short time scale; 4, converting and solving the model; according to the invention, support is provided for coordinating the operation economy and flexibility of the system under security constraints.

Description

Technical Field

[0001] The present invention belongs to the technical field of power system dispatch optimization and power and electricity balance analysis, and relates to a time-series random production simulation method considering flexibility requirements in a power market environment. Background Art

[0002] With the gradual establishment and improvement of the power market mechanism. As an important tool for power system dispatch optimization and power balance analysis, production simulation technology has been widely used in key issues such as system operation economic analysis, resource allocation optimization, and reliability assessment in the market environment. With the accelerated promotion of large-scale grid connection of renewable energy power generation, the operating characteristics of the power system are undergoing profound changes. The structural transformation of the system's operating characteristics has put forward higher requirements on the modeling accuracy and complexity of production simulation methods. Traditional methods are difficult to fully represent the real dynamic operating status of the system, and the accuracy of the assessment of the system's flexibility level is facing challenges.

[0003] Traditional production simulation methods primarily fall into two categories: sequential production simulation and stochastic production simulation. Sequential production simulation uses load time series as input and, incorporating the start / stop constraints, ramping capabilities, and fuel characteristics of various units, dynamically simulates their operation throughout the dispatch cycle. This can be used to generate power market clearing curves, providing a reference for market participants to formulate reasonable bidding plans. Stochastic production simulation focuses on the impact of random factors on system operation and is based on continuous load curves. It is suitable for rapidly analyzing system risks and assessing the rationality of reserve capacity in medium- and long-term market transactions. To address the system uncertainty challenges brought about by the increasing penetration of renewable energy, researchers have recently introduced a variety of uncertainty optimization techniques into the sequential production simulation framework, such as multi-scenario stochastic programming, chance-constrained methods, and robust optimization models. Stochastic programming, for example, characterizes the stochasticity of source and load output by constructing a set of scenarios, but suffers from large solution sizes and high computational overhead. Chance-constrained models can relax dispatch constraints at a certain confidence level, but improperly setting the confidence parameter can lead to safety or economic losses. Robust optimization is suitable for robust decision-making in extreme situations, but its conservative nature often limits the economic benefits of the system. In general, although time-series production simulation can accurately describe the time-series-related characteristics of the random process of source and load output, the inherent contradiction between large-scale simulation and solution efficiency makes this method difficult to apply to complex system analysis over long time scales. In contrast, stochastic production simulation methods have higher computational efficiency and can complete system state analysis without relying on a large number of time-series paths. Typical methods include the equivalent power function method, the semi-invariant method, and the universal generating function method. However, due to its inherent static statistical model, stochastic production simulation methods cannot reflect the time-series coupling effects in system operation, and the accuracy of the simulation needs to be improved.

[0004] In addition, as the uncertainty level of new power systems continues to rise, the contradiction between system flexibility supply and demand is becoming increasingly prominent. Evaluating operating costs while considering the supply and demand of system flexibility is of great significance for improving the accuracy and feasibility of system operation plans. Existing production simulation methods have gradually focused on the adequacy of system flexibility. The current main ideas include two categories: one is based on random production simulation methods, which directly constructs the probability distribution of flexibility supply and demand, combines probability convolution calculation to obtain the distribution characteristics of system flexibility margin, and uses statistical parameters (such as expected value) as measurement indicators; the other method is based on time series production simulation, which is usually carried out on a shorter time scale such as day, analyzes the evolution process of the system's up and down adjustment capabilities under different operating conditions, and extracts flexibility evaluation results with time series information.

[0005] However, current methods still have certain limitations in terms of collaborative modeling of time series and random factors, and the design of flexible supply and demand matching mechanisms, making it difficult to meet the analysis needs of system operation flexibility. Specifically,

[0006] 1) Existing production simulation methods are unable to take into account both the system timing operation and the uncertainty characteristics of source and load output: using only the timing production simulation method is difficult to efficiently characterize the system randomness and is not suitable for complex system analysis over a long time scale; using only the random production simulation method ignores the coupling relationship between timings, and the simulation accuracy needs to be improved.

[0007] 2) Existing methods generally fail to establish a systematic matching mechanism to describe the probabilistic characteristics of flexibility supply and demand. The modeling of the flexibility supply and demand balance relationship is imperfect, making it difficult to characterize the flexibility level of the system under different operating scenarios.

[0008] 3) Existing flexibility indicators are mostly a posteriori static evaluation results, which cannot achieve dynamic feedback adjustment. There is a lack of flexibility early warning feedback for system operation to support dynamic optimization and updating of resource scheduling plans.

[0009] Therefore, a method that integrates sequential production simulation with stochastic production simulation is needed to solve the above technical problems. Summary of the Invention

[0010] The technical solution adopted by the present invention to solve the technical problem is: a time-series random production simulation method considering flexibility demand in the power market environment, comprising the following steps:

[0011] Step 1: Construct a power system sequential-stochastic production simulation framework considering flexibility requirements in a market environment. The simulation framework is a two-layer simulation framework, including an upper-layer model and a lower-layer model.

[0012] Step 2: The upper-level model aims to minimize the total system operating cost and obtains a start-stop scheduling plan that satisfies the system power balance and unit technical characteristics. On this basis, it calculates the timing flexibility supply capacity of each type of unit to provide input basis for the lower-level model.

[0013] Step 3: The lower-level model describes the system's random flexibility demand based on typical probabilistic scenarios, constructs a multi-state flexibility supply model for units at different time scales, and uses the equivalent power function method to perform probabilistic matching of system flexibility supply and demand at multiple time scales, thus achieving flexibility indicator evaluation at short time scales.

[0014] Step 4: Model transformation and solution.

[0015] Preferably, in step 2, the total operating cost of the system is minimized as follows:

[0016] (1)

[0017] In formula (1): is the total operating cost of the system; 、 are the unit fuel cost and startup cost respectively; The system load loss cost, Renewable energy curtailment costs, of which:

[0018] (2)

[0019] (6)

[0020] (11)

[0021] (12)

[0022] In formulas (2), (6), (11), and (12), 、 、 represent the fuel costs of conventional thermal power units, combined cycle units, and cogeneration units respectively; 、 、 、 Represents conventional thermal power units , combined heat and power units , hydropower station , combined cycle units from Mode switched to The single startup cost of the model, 、 、 It is a 0-1 integer variable, representing a conventional thermal power unit , combined heat and power units , hydropower station exist Whether the time period changes from shutdown to operation, It is a 0-1 integer variable, representing the combined cycle unit of Mode in The operating status of the time period, 、 、 、 、 They are thermal power units, combined cycle units, Combined cycle unit, Model: Number of combined cycle units and cogeneration units; For nodes exist The amount of power loss load during the period; is the unit power loss cost; For the system Heat loss load during the period; is the unit heat loss cost; 、 For wind farms , photovoltaic power stations exist The amount of power wasted during the time period; is the penalty coefficient for unit amount of abandoned renewable energy; For reservoirs exist Discarded water flow rate during the period; is the unit water abandonment penalty coefficient; It is the duration of a single period and the minimum time granularity that can be used when quantifying the supply and demand of system flexibility; 、 、 are the number of wind farms, photovoltaic power stations, and hydropower stations respectively.

[0023] More preferably, in step 2, in calculating the timing flexibility supply capacity of each type of unit, the flexibility supply capacity of conventional thermal power units, combined cycle gas units, and cogeneration units is:

[0024] (56)

[0025] (57)

[0026] (58)

[0027] (59)

[0028] (60)

[0029] (61)

[0030] In formulas (56) to (61): Provide flexibility to assess the time granularity of capacity; 、 For conventional thermal power units exist Upward and downward flexibility supply in time periods; 、 Combined cycle unit exist Upward and downward flexibility supply in time periods; 、 For combined heat and power units exist Upward and downward flexibility supply in time periods; For the crew exist Maximum thermal power variation on time scale, 、 、 It is a 0-1 integer variable, representing a thermal power unit , combined cycle units The operating status of mode m and cogeneration unit c in period t, 、 、 、 、 Conventional thermal power units Maximum power, minimum power, Power, ramp-up rate, ramp-down rate during the time period, 、 、 、 、 are the maximum power, minimum power, Power, ramp-up rate, ramp-down rate during the time period, 、 、 、 、 are the maximum power, minimum power, Power, ramp-up rate, ramp-down rate during the time period, is the thermal power output of unit c during period t, It is the reduction in power generation caused by unit heat supply increment under the condition of constant total steam inlet volume of unit c.

[0031] More preferably, in step 2, in calculating the timing flexibility supply capacity of each type of unit, the flexibility supply capacity of hydropower is:

[0032] (62)

[0033] (63)

[0034] In formulas (62) and (63): 、 For hydropower stations exist Upward and downward flexibility supply of time period, is a 0-1 integer variable, representing a hydropower station The running status in period t, 、 、 Hydropower Station Maximum power, minimum power, The power of the period, is the output coefficient, 、 Hydropower Station The maximum and minimum power generation flow, is the effective water head, 、 、 The maximum storage capacity, minimum storage capacity, Storage capacity during the period, Provides flexibility to assess the time granularity of capability, is the discharge delay from the upper reservoir k to reservoir h, For reservoir h in The natural water flow during the period, is the hydropower unit k in the period Power generation flow, is the maximum water rejection flow of unit h.

[0035] Preferably, in step 3, the equivalent electrical quantity function method (EEFM) is used to perform a probabilistic matching analysis on the system flexibility supply capacity and flexibility demand, and then calculate the system flexibility shortage and the probability of insufficient flexibility, and return them to the upper layer as the lower-layer flexibility verification result.

[0036] More optimally, the equivalent energy function method (EEFM) maps each system flexibility demand level to the energy demand at that level into a discrete function relationship. When considering generator failures, the discrete energy functions of each flexibility demand level are directly subjected to probability correction and convolution calculation. The original energy function is:

[0037] (75)

[0038] (76)

[0039] In formulas (75) and (76): is the step length of the flexibility demand level interval, which is the common divisor of the maximum flexibility supply of all units; The level of flexibility required; is the original continuous flexibility demand curve, is the original power function, exist arrive The area under this curve is the amount of electricity corresponding to this level of flexibility demand; is the maximum discretization interval index corresponding to the flexibility demand level, The maximum value of the original flexibility requirement, is the interval step of the flexibility demand level, Indicates that the value is not greater than An integer;.

[0040] Preferably, the step 4 specifically includes: the upper-level sequential production simulation model contains nonlinear expressions, which are linearized using a linearization method; the sequential production simulation model is converted into a mixed integer linear programming model, and then solved using a commercial solver.

[0041] The beneficial effects of the present invention are:

[0042] 1. The present invention defines the flexibility demand of the power system as the random fluctuation of the net load caused by the superposition of new energy output and load changes. A flexible system must have the ability to respond promptly to the up and down ramping demand of the net load. The flexibility of the power system is highly dependent on the operating state of the system. With the continuous change of the power balance point, the flexibility level of the system is significantly affected by the output state of various units and the real-time evolution of the spatiotemporal distribution of the net load. The time-series production simulation method can describe in detail the dynamic response process of power supply and load within the scheduling cycle at a fine time resolution, reveal the flexibility supply level of the system under different operating conditions, and provide support for coordinating the economy and flexibility of system operation under safety constraints.

[0043] 2. The stochastic production simulation method of the present invention can effectively characterize random factors such as source-load fluctuations and equipment failures, significantly reducing computational complexity while ensuring simulation accuracy. It is suitable for analyzing the stochastic supply and demand balance of systems over medium and long timescales. Therefore, the present invention combines the complementary advantages of sequential analysis of sequential production simulation and probabilistic modeling of stochastic production simulation to construct a flexible, dual-layer sequential-stochastic production simulation framework for power systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a simulation framework diagram of the time-series random production simulation method considering flexibility demand in the power market environment of the present invention;

[0045] Figure 2 This is a thermoelectric characteristic curve diagram of the extraction steam CHP unit of the present invention;

[0046] Figure 3 is a schematic diagram of the unit flexibility supply of the present invention;

[0047] Figure 4 is a schematic diagram of an equivalent continuous flexibility demand curve of the present invention;

[0048] Figure 5 It is a model solving flow chart of the present invention;

[0049] Figure 6 is a modified IEEE 30-node system topology diagram of the present invention;

[0050] Figure 7 This is a typical net load pattern diagram of each season for the simulation system of the present invention;

[0051] Figure 8 It is a graph of the annual flexibility index results of each probability scenario of the system of the present invention;

[0052] Figure 9 This is a graph showing the flexibility index results for each system in each month according to the present invention. DETAILED DESCRIPTION

[0053] The following will provide a clear and complete description of the relevant technologies in the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0054] The double-layer simulation framework of this embodiment is as follows Figure 1 As shown in the figure, the upper-layer model is a long-term sequential production simulation model for the power system. With the goal of minimizing the total operating cost of the system, it obtains a start-stop scheduling plan that satisfies constraints such as system power balance and unit technical characteristics. Based on this plan, it calculates the sequential flexibility supply capacity of various types of units, providing input for the lower-layer model. The lower-layer model describes the system's stochastic flexibility requirements based on typical probabilistic scenarios, constructs a multi-state flexibility supply model for units at different time scales, and uses the equivalent power function method to perform probabilistic matching of system flexibility supply and demand at multiple time scales, thus achieving flexibility indicator evaluation at short time scales.

[0055] The upper-level model models the operating modes of various types of generator sets and the system's refined scheduling process, and achieves economical and efficient allocation of power generation resources by rationally arranging the start and stop plans and output strategies of the units.

[0056] 1) Upper-level model objective function: Minimize the total operating cost of the system.

[0057] (1)

[0058] Where: is the total operating cost of the system; 、 are the unit fuel cost and startup cost respectively; The system load loss cost, The cost of abandoning renewable energy.

[0059] (a) Unit fuel cost

[0060] (2)

[0061] (3)

[0062] (4)

[0063] (5)

[0064] Where: 、 、 are the fuel costs of conventional thermal power units, combined cycle units, and combined heat and power (CHP) units; 、 、 Conventional thermal power units , combined cycle units model , CHP units exist Power output during the time period; For CHP units exist Thermal power output during the time period; For the crew Under the condition of constant total steam inlet, the reduction in power generation caused by unit heat supply increase; 、 、 For conventional thermal power units Cost coefficient; 、 、 Combined cycle unit In mode Cost coefficient under 、 、 For CHP units Cost coefficient; is the total simulation time; 、 、 are the number of conventional thermal power units, combined cycle units, and CHP units respectively; is the mode number of the combined cycle unit.

[0065] (b) Unit startup costs

[0066] (6)

[0067] (7)

[0068] (8)

[0069] (9)

[0070] (10)

[0071] Where: 、 、 、 For conventional thermal power units , CHP units , hydropower station , combined cycle units from Mode switched to The single startup cost of the model; 、 、 It is a 0-1 integer variable, representing a conventional thermal power unit , CHP units , hydropower station exist Whether the time period changes from shutdown to operation; It is a 0-1 integer variable, representing the combined cycle unit exist Is the time period from Mode switched to model; 、 、 、 It is a 0-1 integer variable, representing a conventional thermal power unit , CHP units , hydropower station , combined cycle units of Mode in The operation status of the time period, 0 represents outage, 1 represents operation; 、 、 、 For conventional thermal power units , CHP units , hydropower station , combined cycle units of Mode in -1 period of operation status.

[0072] (c) Loss of load costs

[0073] (11)

[0074] Where: For nodes exist The amount of power loss load during the period; is the unit power loss cost; For the system Heat loss load during the period; is the unit heat loss cost.

[0075] (d) Costs of curtailing renewable energy

[0076] (12)

[0077] Where: 、 For wind farms , photovoltaic power stations exist The amount of power wasted during the time period; is the penalty coefficient for unit amount of abandoned renewable energy; For reservoirs exist Discarded water flow rate during the period; is the unit water abandonment penalty coefficient; It is the duration of a single period and the minimum time granularity that can be used when quantifying the supply and demand of system flexibility; 、 、 are the number of wind farms, photovoltaic power stations, and hydropower stations respectively.

[0078] 2) Upper-level model constraints

[0079] Constraints include two types: system-level operation constraints and technical operation constraints of various types of units.

[0080] System-level operational constraints focus on power supply and demand balance and network security operations:

[0081] (a) Node power balance constraints

[0082] (13)

[0083] Where: For nodes exist Electricity load value of the time period; For the line admittance; For nodes exist Voltage phase angle during the time period; For node m Voltage phase angle during the time period; 、 For energy storage power stations exist Discharge power and charging power during the time period; 、 For wind farms , photovoltaic power stations exist Power output during the time period; For the node A collection of connected nodes; For nodes The collection of units.

[0084] (b) Thermal power balance constraints

[0085] (14)

[0086] Where: For the system Heat load during the period;

[0087] (c) Line flow restrictions

[0088] (15)

[0089] Where: For the line The maximum flow limit.

[0090] (d) Voltage phase angle constraint

[0091] (16)

[0092] (17)

[0093] Where: is the reference node of the system.

[0094] (e) Load shedding constraints

[0095] (18)

[0096] (19)

[0097] Where: 、 are the power loss load and heat loss load at node n in the system during period t, 、 They are the upper limit of power loss load and heat loss load of the system at node n respectively.

[0098] Considering the technical characteristics of each resource, the operating output model of each type of unit is constructed:

[0099] (a) Conventional thermal power units

[0100] (20)

[0101] (twenty one)

[0102] (twenty two)

[0103] (twenty three)

[0104] Where: 、 For conventional thermal power units Maximum and minimum technical output; 、 For the crew Upward and downward climbing rates; 、 For the crew The minimum operating time and minimum shutdown time of the unit. Equation (20) is the upper and lower limit constraints of the unit output power; Equation (21) is the ramp constraint of the unit in adjacent time periods; Equations (22)-(23) are the minimum start-stop time constraints of the unit.

[0105] (b) Combined cycle gas turbine units

[0106] (twenty four)

[0107] (25)

[0108] (26)

[0109] (27)

[0110] (28)

[0111] (29)

[0112] Where: Combined cycle unit Mode transfer feasibility factor, 1 means the unit Can be obtained from Mode switched to Mode, 0 means no transfer can occur; 、 Combined cycle unit Maximum and minimum technical output in mode m; 、 For the crew Upward and downward climbing rates in M ​​mode; 、 For the crew The minimum operating time and minimum shutdown time in mode m. Equation (24) is the operating mode state constraint; Equation (25) is the mode transition constraint; Equation (26) is the upper and lower limit constraints of the unit output power; Equation (27) is the unit adjacent time period ramp constraint; Equations (28)-(29) are the minimum start-stop time constraints of the unit.

[0113] (c) Combined heat and power units

[0114] In the extraction steam operation mode, the CHP unit extracts part of the medium and low pressure steam from the steam turbine for heating, and the remaining steam continues to drive the steam turbine to generate electricity. The output power limit of the CHP unit is restricted by the heating power. Its thermoelectric characteristic curve is shown in Figure 2 .

[0115] The feasible region boundary method is used to model the operating boundary of the CHP unit:

[0116] (30)

[0117] (31)

[0118] Where, 、 For CHP units The thermal-electric operation characteristic function represents the upper and lower boundaries of the electric power output; 、 、 Respectively for units The slopes of the CD, BC, and AB sides of the operating interval boundaries; 、 Respectively for units The vertical coordinates of the power boundary points D and A under pure condensing mode operation; For the crew The ordinate of the intersection of the extended line of the operating range boundary BC and the y-axis.

[0119] In addition, the CHP unit must meet the following operating constraints:

[0120] (32)

[0121] (33)

[0122] (34)

[0123] (35)

[0124] (36)

[0125] Where: For CHP units The upper limit of thermal power output; 、 For the crew Upward and downward climbing rates; 、 For the crew The minimum operating time and minimum shutdown time of the unit are: Formula (32) is the unit thermal power output limit constraint; Formula (33) is the upper and lower limit constraints of the electric power output; Formula (34) is the unit ramp constraint in adjacent time periods, which equates the output change in the cogeneration state to the electric power change in the pure condensing condition; Formulas (35)-(36) are the unit minimum start-stop time constraints.

[0126] (d) Cascade hydropower stations

[0127] (37)

[0128] (38)

[0129] (39)

[0130] (40)

[0131] (41)

[0132] (42)

[0133] (43)

[0134] (44)

[0135] Where: For hydropower stations Output coefficient; For hydropower stations exist Power generation flow during the period; For hydropower stations The effective head of For reservoirs exist Storage capacity during the time period; For reservoirs exist Natural water flow during the period; 、 For hydropower stations Maximum and minimum technical output; 、 For hydropower stations Maximum and minimum power generation flow; 、 For hydropower stations the maximum and minimum reservoir capacity; For hydropower stations Maximum water discharge rate; 、 For hydropower stations Minimum operating time and minimum downtime; For the reservoir A collection of upper-level reservoirs with direct hydrological connections; For the upper reservoir Release water to reservoir The discharge delay. Equation (37) is the hydropower coupling constraint, which expresses the relationship between the unit output and the power generation flow and water head; Equation (38) is the reservoir The spatiotemporal coupling constraint represents the water balance of the reservoir in space and time. Equation (39) is the upper and lower limit constraint of output power. Equation (40) is the upper and lower limit constraint of power generation flow. Equation (41) is the upper and lower limit constraint of reservoir storage capacity. Equation (42) is the water abandonment limit constraint. Equations (43)-(44) are the minimum start-up and shutdown time constraints of the unit.

[0136] (e) Energy storage power station

[0137] (45)

[0138] (46)

[0139] (47)

[0140] (48)

[0141] (49)

[0142] (50)

[0143] (51)

[0144] Where: 、 It is a 0-1 integer variable, representing the energy storage power station exist Discharge and charge status of the time period; 、 For energy storage power stations The minimum and maximum discharge power, 、 The charging and discharging power of the energy storage power station; 、 For energy storage power stations Minimum and maximum charging power; 、 For energy storage power stations exist 、 The amount of storage capacity during the time period; 、 For energy storage power stations Maximum and minimum capacity; 、 For energy storage power stations Discharge and charge efficiency; 、 For energy storage power stations The maximum continuous discharge and charge time of the battery is given by Equation (45). Equations (46) and (47) are the upper and lower limit constraints of the charge and discharge power. Equation (48) is the energy storage capacity constraint. Equation (49) is the relationship between the energy storage capacities in adjacent time periods. Equations (50) and (51) are the maximum continuous discharge and charge time constraints.

[0145] (f) New energy stations

[0146] (52)

[0147] (53)

[0148] (54)

[0149] (55)

[0150] Where: 、 Wind farms , photovoltaic power stations exist The predicted output for the time period.

[0151] 3) Quantification of flexible supply capacity in a single period

[0152] System flexibility capacity exhibits three characteristics: 1) Directionality, manifested in the asymmetry of the system's upward and downward adjustment capabilities; 2) State-dependence, manifested in the fact that the system's flexibility provision level at a given time period is constrained by factors such as the power source's current on / off status, output level, and ramp performance; and 3) Temporal correlation, meaning that flexibility assessment results differ at different time scales. Therefore, when defining the flexibility provision capacity of each type of unit within a single time period, it is important to fully consider the impact of these characteristics on the unit's adjustment potential and operational feasibility.

[0153] The regulation capabilities of conventional thermal power units, combined cycle gas units, and CHP units are primarily limited by their current operating point and ramping performance. The flexibility supply capability is defined as follows:

[0154] (56)

[0155] (57)

[0156] (58)

[0157] (59)

[0158] (60)

[0159] (61)

[0160] Where: Provide flexibility and assess the time granularity of capability, typically 15 minutes; 、 For conventional thermal power units exist Upward and downward flexibility supply in time periods; 、 Combined cycle unit exist Upward and downward flexibility supply in time periods; 、 For combined heat and power units exist Upward and downward flexibility supply in time periods; For the crew exist Maximum thermal power variation over time scales.

[0161] Hydropower and energy storage have advantages such as rapid response, wide regulation range, and strong climbing ability. They are typical high-performance flexible regulation resources. However, their flexibility supply capacity is limited by the capacity constraints of the energy storage medium, making it difficult to continuously provide high-power flexible regulation support over a long period of time. The flexibility supply capacity of hydropower is limited by the reservoir storage capacity and runoff conditions:

[0162] (62)

[0163] (63)

[0164] Where: 、 For hydropower stations exist Upward and downward flexibility supply in time periods;

[0165] The flexibility of energy storage is limited by the battery's state of charge:

[0166] (64)

[0167] (65)

[0168] Where: 、 For energy storage power stations exist Upward and downward flexibility supply in time periods.

[0169] 4) Lower-level problem: Stochastic production simulation model with probabilistic matching of flexible supply and demand

[0170] (a) Cluster identification of unit multi-state flexibility supply model

[0171] In order to characterize the uncertainty of the flexibility supply of units at different time scales, based on the quantitative results of the upward and downward flexibility supply capabilities of various types of units in different scheduling periods, the K-means algorithm is used for clustering identification, and the typical flexibility supply states and probabilities of the units at different time scales are extracted, and the time series changes of the unit's flexibility supply capacity are converted into discrete state distributions. Furthermore, the original probability of each state is corrected in combination with the forced outage rate of the unit, and a multi-state flexibility supply model for the unit at multiple time scales is obtained. Taking the four-state flexibility supply model of the unit as an example, its schematic diagram is shown in Figure 3 .

[0172] Assume that the unit flexibility supply has states, each state corresponds to a set of flexibility supply capabilities , flexible supply of outage volume and its probability ,satisfy

[0173] (66)

[0174] Where: Provides the maximum flexibility the unit can offer.

[0175] (b) Quantification of system flexibility requirements

[0176] The uncertainty of renewable energy output and load together leads to random fluctuations in net load. Based on the stochastic characteristics of wind power, photovoltaic power generation, and load, we establish corresponding random sample sets. Based on this, we use a heuristic synchronous back-substitution method to reduce scenarios and construct a set of typical net load scenarios that reflect the randomness of system operation. We then calculate the system flexibility demand and duration under each scenario, providing a calculation basis for the original sustained flexibility demand curve.

[0177] Wind power output is significantly affected by meteorological conditions, and its output fluctuation mainly comes from changes in wind speed. Considering the obvious autocorrelation and volatility of wind speed, the ARMA model is used to simulate wind speed. The wind speed of the wind farm at time t can be expressed as:

[0178] (67)

[0179] (68)

[0180] Where: 、 For this wind farm The mean and standard deviation of historical wind speed during the period; is the random value of wind speed; is the autoregressive coefficient; is the moving average coefficient; is a white noise term with a mean of 0 and a variance of , is the random disturbance term of wind speed at the ti-th moment, is the white noise term at the tjth moment.

[0181] There is a nonlinear mapping relationship between wind speed and wind turbine power output, which can be approximated by a piecewise function:

[0182] (69)

[0183] Where: 、 、 are the cut-in wind speed, rated wind speed and cut-out wind speed of the wind turbine respectively; is the rated power of the wind turbine, is the wind speed.

[0184] Photovoltaic power generation output is mainly affected by solar irradiance, and the two are approximately linearly proportional. Studies have shown that solar irradiance in a certain period of time follows a Beta distribution, and photovoltaic power generation is also considered to follow a Beta distribution, with a probability density function of:

[0185] (70)

[0186] (71)

[0187] (72)

[0188] Where: is the gamma function; 、 is the shape parameter of the Beta distribution; is the rated power of the photovoltaic generator set; 、 is the mean and standard deviation of solar irradiance during this period, For photovoltaic power stations The predicted output, For photovoltaic power stations Power output.

[0189] To characterize the uncertainty of load forecasting, this implementation assumes that the load forecast error has a mean of 0 and a standard deviation of The Gaussian distribution of is used, and on this basis, an error sample set is generated to correct the load forecast value and obtain the actual load value after disturbance:

[0190] (73)

[0191] (74)

[0192] Where: is the load forecast error; is the load forecast value.

[0193] (c) System flexibility index evaluation

[0194] The equivalent energy function method (EEFM) is an efficient stochastic production simulation method for power systems. This method considers the failure probability of generator sets and, without explicitly constructing a large number of production scheduling paths, evaluates the reliability and economic efficiency of system operation through energy convolution calculations. To quantitatively assess system flexibility levels across multiple timescales, this implementation utilizes the EEFM to perform a probabilistic matching analysis between system flexibility supply capacity and flexibility demand. This calculation then calculates the system flexibility shortfall and the probability of flexibility shortfall, which are then returned to the upper layer as the lower-level flexibility verification result.

[0195] The principle of calculating system flexibility index based on equivalent continuous flexibility demand curve is shown in Figure 4 In the figure, the horizontal axis represents the system System flexibility requirements at time granularity. The vertical axis represents the duration of flexibility requirements. Any point on the curve Indicates that the flexibility requirement of the system is greater than or equal to The duration of . is the original continuous flexibility demand curve, which can be obtained by statistical calculation of the net load series; In order to equate the random influence of multi-state unit m to flexibility demand, the curve The modified equivalent continuous flexibility demand curve; It is the equivalent continuous flexibility demand curve formed after the convolution calculation of all units is completed. is the number of units; is the total duration of the evaluation cycle; The maximum value for original flexibility is required; The sum of all units' maximum flexibility supply; is the expected value of the system's insufficient flexibility time, and the shaded area is the expected unbalanced electricity caused by insufficient flexibility supply.

[0196] EEFM maps each level of system flexibility demand to the power demand at that level into a discrete function relationship. When considering generator failures, it directly performs probability correction and convolution calculations on the discrete power functions of each flexibility demand level. The original power function is defined as follows:

[0197] (75)

[0198] (76)

[0199] Where: is the step length of the flexibility demand level interval, which is the common divisor of the maximum flexibility supply of all units; is the maximum discretization interval index corresponding to the flexibility demand level, Indicates that it is not greater than integer; for exist arrive The area under this curve is the amount of electricity corresponding to this level of flexibility demand.

[0200] Before completion After the convolution operation of the multi-state machine group, the equivalent power function is formed ,definition , then arrange multi-state machine groups The equivalent power function after is:

[0201] (77)

[0202] In arranging After the multi-state unit is running, the expected power shortage value of the system is

[0203] (78)

[0204] (79)

[0205] Where: Before The unit has met the interval flexibility needs, is the independent variable of the function, which represents the level of system flexibility demand. is the probability of state s, Provides capacity for maximum flexibility of unit l.

[0206] After all the multi-state machine group convolution operations are completed, the system flexibility index can be calculated. The flexibility deficit, that is, the expected unbalanced power caused by insufficient flexibility supply, is

[0207] (80)

[0208] The probability of insufficient system flexibility is:

[0209] (81)

[0210] Where: After arranging the operation of the nth multi-state unit, the system power is less than the expected value. is the equivalent power function after arranging n units, is the function value when the flexibility demand level is Jn after the system arranges the first n units, is the total duration of the evaluation cycle, is the step length of the flexibility demand level interval.

[0211] (d) Upper model calculation boundary correction

[0212] When the lower-level flexibility assessment results show that the system flexibility index for one or more time periods exceeds the preset threshold, it indicates that the current scheduling plan has a risk of insufficient flexibility and the upper-level scheduling boundary needs to be revised. This implementation method adopts the method of introducing a flexibility supply and demand balance constraint for key time periods in the upper-level model. The constraint form is as follows:

[0213] (82)

[0214] Where: The middle period of the current optimization round The theoretical flexibility supply capacity of the system's starting units, that is, its ramping capability; Provides the capability result for the theoretical flexibility of the system during the last iteration, This is the flexibility shortfall for this period, calculated from the lower-level evaluation in the previous iteration. This constraint essentially increases the overall system flexibility by increasing the system's operating capacity during critical periods.

[0215] 5) Model transformation and solution

[0216] (a) Model linearization

[0217] The upper-level sequential production simulation model contains nonlinear expressions, which are linearized using the linearization method:

[0218] The quadratic functions in equations (3) and (4) can be linearized using piecewise linearization technology.

[0219] For the nonlinear cost expression of CHP unit shown in formula (5), the following linear formula is used as an equivalent replacement:

[0220] (83)

[0221] Where: 、 、 For the crew cost coefficient.

[0222] The variables involved in formula (7) to formula (10) are all 0-1 integer variables, which can be equivalently converted into linear constraints. Taking the expression as an example, it can be converted into:

[0223] (84)

[0224] (85)

[0225] (86)

[0226] Equation (30) represents the lower limit of the CHP unit operation, which can be equivalently expressed using the following formula:

[0227] (87)

[0228] In formula (33) and The 0-1 integer variable is multiplied by the continuous variable and the large M method is used to linearize it. As an example, it can be equivalently transformed into the following linear constraint form:

[0229] (88)

[0230] Where: is a sufficiently large constant, For combined heat and power units Maximum power during a time period.

[0231] At this point, the sequential production simulation model has been converted into a mixed-integer linear programming (MILP) model, which can be solved using mature commercial solvers.

[0232] (b) Model solution

[0233] The iterative solution process of the power system time series-stochastic production simulation two-layer model considering flexibility requirements proposed in this embodiment is as follows: Figure 5 As shown. In this two-layer model, the upper layer completes medium- and long-term simulations based on deterministic boundaries, avoiding the calculation of a large number of random scenarios and significantly improving computational efficiency; the lower layer uses probabilistic scenario sets and multi-state modeling to characterize the impact of source-load uncertainty on the short-time-scale flexibility supply and demand balance. When the evaluation results of the lower layer show that the system flexibility index exceeds the threshold, a correction signal of the decision boundary is fed back to the upper layer, and the rolling re-optimization of the scheduling strategy is driven by adjusting the flexibility supply and demand balance constraints. This two-layer model iteratively solves between deterministic optimization and random evaluation verification, and ultimately achieves system economic operation under a certain level of flexibility.

[0234] Example

[0235] This embodiment selects the modified IEEE 30-node system for verification. The number and installed capacity of each type of power generation entity in the test system are shown in Table 1, and the system topology is shown in Figure 6A typical sample set of wind power and photovoltaic power forecast output and system load was generated based on the 2024 statistical data of a northwestern province. The penalty for wind and solar curtailment was set at 300 yuan / MW·h; the penalty for hydropower curtailment was 2 yuan / m³; and the penalty for load loss was 80,000 yuan / MW·h. The maximum load of the system was 970.23 MW, with maximum upward and downward flexibility demands of 89.96 MW and 77.05 MW, respectively. A full-year production simulation was conducted with a 15-minute time step and the minimum time granularity for flexibility assessment. Table 1 shows the installed capacity of the main power generation entities.

[0236]

[0237] The principal component clustering algorithm is used to identify the typical net load patterns of the simulation system in spring, summer, autumn and winter. The results are shown in Figure 7 The system generally faces varying degrees of net load timing fluctuations in different seasons, especially during critical periods such as the morning peak, afternoon photovoltaic power generation, and evening drop-off. Taking summer as an example, the net load shows a clear "flat-valley-peak" characteristic, with a steep climb during the evening peak. The midday load surge caused by extreme high temperatures and the sudden drop in photovoltaic output in sunny weather lead to large fluctuations in the net load valley between typical days, further exacerbating the dynamic complexity of the system. Overall, the system is highly dependent on flexible adjustment resources, and the overall flexibility demand is at a high level.

[0238] 1) System production simulation results

[0239] Conduct production simulation for the target year of the system and evaluate the annual flexibility indicators of the system in various directions under various probability scenarios. The results are as follows: Figure 8 shown.

[0240] Overall, the system's upward flexibility margin is more insufficient than its downward margin, showing a certain asymmetry. The main reason is that during the period of low net load, the number of operating units in the system is small, and the upward ramp capacity that can be called upon is limited, which restricts the system's upward flexibility supply capacity. As a result, the system is more likely to face the risk of flexibility supply shortage when facing a rapid increase in net load. In addition, scenario 4 presents a typical "low probability, high risk" feature. Although this scenario has the lowest probability of occurrence, its upward EFNS index value is as high as over 100 MW·h, which is significantly higher than the average level of other scenarios, indicating that the system faces a more severe risk of upward flexibility supply and demand imbalance under this scenario.

[0241] 2) Comparative analysis of different optimization solutions

[0242] In order to further compare and verify the effectiveness of the proposed model in improving the flexibility of system operation, this embodiment sets up four comparative simulation schemes:

[0243] Option 1: Only traditional deterministic time-series production simulation is performed, system curtailment of wind and solar power is not allowed, and energy storage is not considered for regulation.

[0244] Option 2: Based on Option 1, the system is allowed to maintain power balance by curtailing wind and solar power.

[0245] Solution 3: Based on Solution 2, the energy storage power station is considered to participate in the regulation, which is the first optimization result of the iterative process of the method proposed in this embodiment.

[0246] Solution 4: Based on Solution 3, a flexible supply and demand probabilistic matching and feedback correction mechanism is introduced, which is the two-layer model proposed in this embodiment.

[0247] Table 2 shows the annual operating cost structure and flexibility shortfall of the system under different operating schemes. As can be seen from the results, Scheme 1 fails to consider curtailed wind and solar power and the regulatory role of energy storage power stations. Consequently, the system fails to meet power balance constraints during periods of negative net load and ramping constraints during periods of significant net load fluctuations, thus failing to generate an effective operating scheme. Scheme 2 introduces a penalty mechanism for curtailed wind and solar power, but still fails to consider the participation of energy storage in regulation. The overall system regulation capacity is limited, resulting in a flexibility shortfall of 497.5 MW·h, significant curtailment and load shedding losses, and the highest total operating cost, 2.27 billion yuan. Compared with Scheme 2, Scheme 3 significantly alleviates the system's flexibility regulation pressure after introducing energy storage resources, significantly reducing curtailment costs and total costs, and reducing the flexibility shortfall, demonstrating the important role of fast and flexible regulation resources in improving the system's flexibility supply capacity. Furthermore, Scheme 4 further introduces flexibility supply and demand balance constraints, significantly reducing the system's flexibility shortfall to 84.7 MW·h and a total operating cost of 1.44 billion yuan, slightly higher than Scheme 3 but significantly better than Scheme 2. Compared to Option 3, which doesn't consider flexibility supply and demand balance, Option 4 increases total system operating costs by 3.8%, but reduces the flexibility shortfall by 64.7%. This option effectively improves system operational flexibility while maintaining excellent economic efficiency, demonstrating the effectiveness of the proposed method in identifying and mitigating flexibility supply and demand imbalances under complex operating conditions. Table 2 shows the annual operating costs and flexibility shortfalls.

[0248]

[0249] Figure 9The evaluation results of the upward and downward flexibility indicators for each month under each operating scenario are presented. It can be seen that flexibility is relatively abundant in the summer, while various flexibility indicators are relatively low. This is because photovoltaic power generation efficiency decreases in high temperatures, and the system net load fluctuates more gradually than in other seasons, resulting in less regulation pressure. Regarding upward flexibility, in Scenario 2, the EFNS and FNS indicators are significantly higher in several months with significant flexibility demand. For example, the EFNS in November reached 80 MW·h. This indicates that without considering energy storage and a stochastic flexibility assessment feedback mechanism, the system's upward flexibility is severely insufficient, making it difficult to effectively cope with uncertain net load fluctuations. In Scenario 3, after introducing energy storage to participate in regulation, the overall system flexibility level is significantly improved, while the flexibility indicator evaluation results are significantly reduced. The Scenario 4 model proposed in this example considers the balance between flexibility supply and demand, effectively identifying the system's flexibility supply and demand contradictions. Further feedback and corrections are then made to the scheduling scheme, controlling the flexibility indicators within the threshold range. This results in relatively optimal upward flexibility for each month, with both the EFNS and FNS indicators at their lowest levels. For example, in November, the upward EFNS dropped to 15 MW·h. In terms of downward flexibility, Option 2 also exhibits a high risk of insufficient flexibility in typical months such as April and October. After energy storage is involved in regulation, the downward flexibility level of the system in Option 3 is significantly improved, and can basically meet the flexibility index threshold requirements. The flexibility gap in Option 4 is further reduced, and the flexibility index in each month is maintained at the lowest level, the FNS index is close to 0, and the EFNS index is also within an acceptable range.

[0250] The results show that the proposed model effectively makes up for the shortcomings of existing production simulation methods in terms of flexibility supply and demand modeling, which is difficult to take into account both the system timing operation and the source and load uncertainty characteristics at the same time. It can effectively identify the risk of insufficient system flexibility, reasonably arrange and deploy various types of flexible adjustment resources, and significantly reduce the risk of power outages and load losses while maintaining operational economy, thereby improving the level of system operational flexibility.

[0251] In summary, the present invention defines the flexibility demand of the power system as the random fluctuations in net load caused by the superposition of renewable energy output and load changes, revealing the flexibility supply level of the system under different operating conditions, and providing support for coordinating the system's operating economy and flexibility under safety constraints.

[0252] It should be emphasized that the above are only preferred embodiments of the present invention and do not limit the present invention in any form. Any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A temporal stochastic production simulation method considering flexibility demand in the power market environment is characterized by: The following steps are involved: Step 1: Construct a power system sequential-stochastic production simulation framework considering flexibility requirements in a market environment. The simulation framework is a two-layer simulation framework, including an upper-layer model and a lower-layer model. Step 2: The upper-level model aims to minimize the total system operating cost and obtains a start-stop scheduling plan that satisfies the system power balance and unit technical characteristics. On this basis, it calculates the timing flexibility supply capacity of each type of unit to provide input basis for the lower-level model. Step 3: The lower-level model describes the system's random flexibility demand based on typical probabilistic scenarios, constructs a multi-state flexibility supply model for units at different time scales, and uses the equivalent power function method to perform probabilistic matching of system flexibility supply and demand at multiple time scales, thus achieving flexibility indicator evaluation at short time scales. Step 4: Model transformation and solution.

2. The temporal random production simulation method considering flexibility demand in the power market environment according to claim 1 is characterized in that: In step 2, the minimized total system operating cost is: (1) In formula (1): is the total operating cost of the system; 、 are the unit fuel cost and startup cost respectively; The system load loss cost, Renewable energy curtailment costs, of which: (2) (6) (11) (12) In formulas (2), (6), (11), and (12), 、 、 represent the fuel costs of conventional thermal power units, combined cycle units, and cogeneration units respectively; 、 、 、 Represents conventional thermal power units , combined heat and power units , hydropower station , combined cycle units from Mode switched to The single startup cost of the model, 、 、 It is a 0-1 integer variable, representing a conventional thermal power unit , combined heat and power units , hydropower station exist Whether the time period changes from shutdown to operation, It is a 0-1 integer variable, representing the combined cycle unit of Mode in The operating status of the time period, 、 、 、 、 They are thermal power units, combined cycle units, Combined cycle unit, Model: Number of combined cycle units and cogeneration units; For nodes exist The amount of power loss load during the period; is the unit power loss cost; For the system Heat loss load during the period; is the unit heat loss cost; 、 For wind farms , photovoltaic power stations exist The amount of power wasted during the time period; is the penalty coefficient for unit amount of abandoned renewable energy; For reservoirs exist Discarded water flow rate during the period; is the unit water abandonment penalty coefficient; It is the duration of a single period and the minimum time granularity that can be used when quantifying the supply and demand of system flexibility; 、 、 are the number of wind farms, photovoltaic power stations, and hydropower stations respectively.

3. The temporal random production simulation method considering flexibility demand in the power market environment according to claim 2 is characterized in that: In step 2, in the calculation of the sequential flexibility supply capacity of each type of unit, the flexibility supply capacity of conventional thermal power units, combined cycle gas units, and cogeneration units is: (56) (57) (58) (59) (60) (61) In formulas (56) to (61): Provide flexibility to assess the time granularity of capacity; 、 For conventional thermal power units exist Upward and downward flexibility supply in time periods; 、 Combined cycle unit exist Upward and downward flexibility supply in time periods; 、 For combined heat and power units exist Upward and downward flexibility supply in time periods; For the crew exist Maximum thermal power variation on time scale, 、 、 It is a 0-1 integer variable, representing a thermal power unit , combined cycle units The operating status of mode m and cogeneration unit c in period t, 、 、 、 、 Conventional thermal power units Maximum power, minimum power, Power, ramp-up rate, ramp-down rate during the time period, 、 、 、 、 are the maximum power, minimum power, Power, ramp-up rate, ramp-down rate during the time period, 、 、 、 、 are the maximum power, minimum power, Power, ramp-up rate, ramp-down rate during the time period, is the thermal power output of unit c during period t, It is the reduction in power generation caused by unit heat supply increment under the condition of constant total steam inlet volume of unit c.

4. The temporal random production simulation method considering flexibility demand in the power market environment according to claim 2 is characterized in that: In step 2, in the calculation of the sequential flexibility supply capacity of each type of unit, the flexibility supply capacity of hydropower is: (62) (63) In formulas (62) and (63): 、 For hydropower stations exist Upward and downward flexibility supply of time period, is a 0-1 integer variable, representing a hydropower station The running status in period t, 、 、 Hydropower Station Maximum power, minimum power, The power of the period, is the output coefficient, 、 Hydropower Station The maximum and minimum power generation flow, is the effective water head, 、 、 The maximum storage capacity, minimum storage capacity, Storage capacity during the period, Provides flexibility to assess the time granularity of capability, is the discharge delay from the upper reservoir k to reservoir h, For reservoir h in The natural water flow during the period, is the hydropower unit k in the period Power generation flow, is the maximum water rejection flow of unit h.

5. The temporal random production simulation method considering flexibility demand in the power market environment according to claim 1 is characterized in that: In step 3, the equivalent electricity function method (EEFM) is used to perform a probabilistic matching analysis on the system flexibility supply capacity and flexibility demand, and then calculate the system flexibility shortage and the probability of flexibility deficiency, which are returned to the upper layer as the lower layer flexibility verification result.

6. The temporal random production simulation method considering flexibility demand in the power market environment according to claim 5 is characterized in that: The Equivalent Electricity Function Method (EEFM) maps each system flexibility demand level and the electricity demand at that level into a discrete function relationship. When considering generator failures, the discrete electricity function of each flexibility demand level is directly subjected to probability correction and convolution calculation. The original electricity function is: (75) (76) In formulas (75) and (76): is the step length of the flexibility demand level interval, is the maximum discretization interval index corresponding to the flexibility demand level, Indicates that the value is not greater than integer; for exist arrive The area under this curve.

7. The temporal random production simulation method considering flexibility demand in the power market environment according to claim 1 is characterized in that: The step 4 specifically includes: The upper-level sequential production simulation model contains nonlinear expressions, which are linearized using a linearization method; the sequential production simulation model is converted into a mixed integer linear programming model, which is then solved using a commercial solver.

Images