Capacity credibility evaluation method and system for flexible resources to participate in capacity market

Abstract

The invention relates to the technical field of capacity credibility evaluation of capacity market subjects, in particular to a capacity credibility evaluation method for flexible resources participating in a capacity market. Comprising the following steps: sampling to generate a plurality of conventional generator set available output sequences; randomly generating output sequences of a plurality of new energy units; determining an energy storage output sequence according to the energy storage scheduling strategy; according to the available output sequence of the conventional generator set, the new energy output sequence and the energy storage output sequence, calculating an original reliability index value of the system added with the to-be-evaluated power generation resource; power generation resources to be evaluated are removed, a perfect unit with a certain capacity is connected, and a corresponding reliability index value is calculated; correcting the perfect unit capacity in the access system until the reliability index result is consistent with the original reliability index value or less than a limited error; the added perfect unit capacity is the credible capacity of the to-be-evaluated power generation resource, and the credible capacity is divided by the installed capacity of the to-be-evaluated power generation resource to obtain the capacity credibility.

Description

Technical Field

[0001] The present invention relates to the technical field of capacity credibility assessment of capacity market entities, and in particular to a capacity credibility assessment method and system for flexible resources participating in a capacity market. Background Art

[0002] Faced with the severe challenges of global climate change and environmental pollution, my country proposed the goals of carbon peak and carbon neutrality at the United Nations General Assembly, marking the upcoming profound transformation of my country's energy structure and power system. With the advancement of the dual carbon goals, my country's power system will accelerate its transformation to a new power system dominated by new energy. This transformation has put forward an urgent need for the reconstruction of the power market system. Since the launch of a new round of power system reform in my country in 2015, 14 regions have carried out the construction and trial operation of power spot markets, but the existing power market construction still faces many challenges.

[0003] First, as new energy becomes the main form of power generation in the power grid, spot market prices continue to fall. Conventional thermal power units find it difficult to obtain stable returns in an energy market that lacks a power generation surplus mechanism, and the problem of investment cost recovery needs to be solved urgently. Shandong, Guangdong and other places have begun to explore capacity compensation mechanisms, but these mechanisms are mostly approved by the government or regulatory agencies, resulting in compensation effects lagging behind market regulation. Drawing on the capacity market mechanisms of mature power markets such as PJM, MISO, and the United Kingdom in the United States, stimulating power generation investment through market competition, and ensuring reliable capacity during peak hours are of great significance for my country to build a power generation surplus mechanism.

[0004] Secondly, the increase in the proportion of new energy access has led to a significant increase in the volatility and uncertainty of the net load of the power system, posing a challenge to the safe and stable operation of the system. To this end, the power system needs a higher flexibility adjustment capability to ensure balance and stability. Although flexible resources such as energy storage have been widely explored, their energy is limited and cannot fully support the system capacity margin. Therefore, it is necessary to establish evaluation indicators in the capacity market to measure the contribution of energy storage and new energy units to system reliability and achieve reasonable clearing and pricing.

[0005] In summary, facing the goal of building a new power system, my country's power market system urgently needs to be restructured to build a reasonable power generation surplus mechanism. At the same time, with the high proportion of new energy access, the flexibility demand of the system is increasing. How to discover the value of flexibility from the perspective of the capacity market, and how to take into account the participation of emerging entities such as new energy units and energy storage in the capacity market, have become urgent issues to be solved. Summary of the invention

[0006] This study aims to propose a capacity credibility assessment method for flexible resources participating in the capacity market, which has important theoretical and practical value for promoting power market reform and improving the safety and economy of power system operation. The specific scheme is as follows: A capacity credibility assessment method for flexible resources participating in the capacity market is provided, including the following steps:

[0007] S1: Model the conventional units and generate a number of available output sequences of conventional generator sets by sampling;

[0008] S2: Model the new energy units and randomly sample and generate output sequences of several new energy units;

[0009] S3: Based on the available output sequence of the conventional generator sets and the output sequence of the new energy generator sets, determine the output sequence of the energy storage according to the energy storage scheduling strategy;

[0010] S4: Calculating the original reliability index value of the power generation resource to be evaluated in the system according to the available output sequence of conventional power generation units, the output sequence of new energy sources and the output sequence of energy storage;

[0011] S5: Remove the power generation resources to be evaluated, connect a perfect unit of a certain capacity (used to represent an idealized power generation resource), perform system reliability calculations and record the corresponding reliability index values; correct the capacity of the perfect unit connected to the system until the reliability index calculation result of the system after adding the perfect unit is consistent with the original reliability index value or less than the specified error; at this time, the capacity of the perfect unit added is the credible capacity of the power generation resources to be evaluated, and the credible capacity is divided by the installed capacity of the power generation resources to be evaluated to obtain the capacity credibility.

[0012] Furthermore, the modeling process of the conventional generator set in step S1 is as follows:

[0013] During the operation of traditional generator sets, they alternate between two different states: fault-free operation and fault shutdown and repair. The switching frequency between states is inversely proportional to the average dwell time in a specific state, that is, the switching frequency is equal to the inverse of the dwell time. Specifically, the switching frequency from the running state to the shutdown state and from the shutdown state to the running state is calculated. The following is the specific calculation formula:

[0014]

[0015] In the formula, λ is the failure rate of the generator; MTTF is the mean time between failures; μ is the repair rate of the generator; MTTR is the mean repair time; MTTF and MTTR are obtained through historical data, and the specific calculation formula is as follows:

[0016]

[0017] Where N o is the total number of failures; is the normal working time of the generator set before the nth fault; is the downtime and maintenance time of the generator at the nth fault;

[0018] The duration of a conventional generator set in two states is approximately considered to obey an exponential distribution due to the randomness of the fault, satisfying the following formula:

[0019] TTF=-MTTF×ln(u1) (5)

[0020] TTR=-MTTR×ln(u2) (6)

[0021] Where TTF and TTR are the normal operation duration and fault maintenance duration respectively; u1 and u2 are random variables with uniform distribution in the interval [0,1].

[0022] Furthermore, in step S1, the steps for generating the available output sequence of the conventional generator set are as follows:

[0023] 1) For conventional units, sample and generate the duration of normal operating status;

[0024] Specifically, assuming that the initial state of all units is a normal operating state, a random number u1 is generated by sampling in a uniform distribution in the interval [0,1], and the corresponding normal operating duration is obtained using the following formula (5);

[0025] TTF=-MTTF×ln(u1) (5)

[0026] 2) Sample the duration of the fault repair state;

[0027] Specifically, a random number u2 is generated by sampling from a uniform distribution in the interval [0,1], and the corresponding fault inspection duration is obtained using formula (6);

[0028] TTR=-MTTR×ln(u2) (6)

[0029] 3) Alternately generate the duration of each state until the length of the generator state sequence is equal to the length of the research period;

[0030] Repeat steps 1) and 2) to alternately generate the duration of the normal operation state and the fault maintenance state, and determine the state of the conventional generator set at each moment, until the length of the state sequence is greater than or equal to the total study period; if the length of the generated sequence is greater than or equal to the total study period, correction is required to correct the length of the state sequence of the generator set to the length of the total study period;

[0031] 4) Generate the available output sequence of conventional generator sets

[0032] According to the state sequence and maximum output of conventional generator sets (based on the generator's own parameters, the generator is designed with an upper limit), it is converted into the available output sequence of conventional units. The specific formula is as follows:

[0033]

[0034] Where P i,t is the available output of conventional generator set i at time t; x i,t is the state of the conventional generator set i at time t, 1 means normal startup, and 0 means fault shutdown; is the maximum output of conventional generator set i.

[0035] Furthermore, a Markov Chain Monte Carlo simulation (MCMC) method is used to simulate the output sequence of the new energy generator set, wherein the output sequence of the new energy generator set includes a wind power output sequence and a photovoltaic output sequence.

[0036] Furthermore, the wind power output sequence is generated through the following steps:

[0037] Step 1: Discretize the historical wind power output into different states

[0038] By analyzing the historical data of wind power, according to the size of wind power output, the output at each time point is divided into N states. The original wind power output sequence {W1, W2, …, W T} is transformed into a state sequence {q1,q2,…,q T}. The corresponding conversion formula satisfies the following formula:

[0039]

[0040] In the formula, w N is the difference between the maximum and minimum values ​​of each state when wind power is divided into N states; W max , W min They are the maximum and minimum output of wind power during the study period.

[0041] Step 2: Generate wind power output state transition probability matrix

[0042] In the Markov chain model, the value of each state depends on the value of the previous state, and the states at adjacent moments satisfy the following formula:

[0043] q t+1 =Y·q t (10)

[0044] Where Y is the state transition probability matrix.

[0045] The number of changes between states is obtained through statistics of historical wind power output data, and the state transfer matrix Y is established accordingly. Each element in the matrix Y, Y r,s It indicates the probability that the wind power output will be transferred to s at the next moment under the condition that the current state is r, and the sum of the probabilities of each row should be 1. That is:

[0046] y r,s =Y(q t+1 =r|q t =s) (11)

[0047] And based on the state transfer probability matrix, a cumulative probability matrix is ​​generated to assist in the calculation. The specific calculation formula for each element in the cumulative probability transfer matrix is ​​as follows:

[0048]

[0049] In the formula, is the probability of transitioning to state 1, 2, ..., n at the next moment under the condition that the current state is m.

[0050] Step 3: Generate wind power state sequence

[0051] First, an integer u0 is sampled from a uniform distribution in the interval [1, N] to determine the initial state of wind power output. Then, a random number u1 is sampled from a uniform distribution in the interval [0, 1]; The relationship between u1 and u1 is used to determine the state of wind power at time t = 1. Then, the same method is used to generate the state of wind power at time t + 1 according to the state value at time t, until the wind power state at all times in the period to be studied is generated.

[0052] Step 4: Generate wind power output sequence

[0053] According to the wind power status, it is converted into specific wind power output. The conversion formula is as follows:

[0054] w t =(q t +u t ) N (13)

[0055] In the formula, w t is the output value at time t in the newly generated wind power output sequence.

[0056] Furthermore, the steps for generating the photovoltaic output sequence are as follows:

[0057] Step 1: Discretize the historical photovoltaic output into different states

[0058] By analyzing the historical data of photovoltaic power generation, the output of photovoltaic power generation at each time point is divided into N states according to the output of photovoltaic power generation. Transformed into a state sequence {q pv1 ,q pv2 ,…,q pvT}. The corresponding conversion formula satisfies the following formula:

[0059]

[0060] In the formula, Pv is the difference between the maximum and minimum values ​​of each state when photovoltaic power generation is divided into N states; max , Pv min They are the maximum and minimum outputs of photovoltaic power generation during the study period.

[0061] Unlike wind power, photovoltaic power does not produce power at night. Therefore, when dividing different discrete states in the above steps, only the valid data of photovoltaic power output is taken into account; secondly, photovoltaic power shows an increasing trend in the morning period and a decreasing trend in the afternoon period. To reflect this characteristic of photovoltaic power, when discretizing the photovoltaic output in this step, the same output in the morning and afternoon periods needs to be divided into different states.

[0062] Step 2: Generate photovoltaic output state transition probability matrix

[0063] In the Markov chain model, the value of each state depends on the value of the previous state, and the states at adjacent moments satisfy the following formula:

[0064] q pv,t+1 =Y·q pv,t (16)

[0065] Where Y is the state transition probability matrix.

[0066] The number of changes between states is obtained through statistics of photovoltaic historical output data, and the state transfer matrix Y is established accordingly. Each element in the matrix Y, Y pv,r,s Indicates that the current state is r pv Under the condition of pv The probability of each row should sum to 1. That is:

[0067] y pv,r,s =Y(q pv,t+1 =r pv |q pv,t =s pv ) (17)

[0068] And based on the state transfer probability matrix, a cumulative probability matrix is ​​generated to assist in the calculation. The specific calculation formula for each element in the cumulative probability transfer matrix is ​​as follows:

[0069]

[0070] In the formula, is the probability of transitioning to state 1, 2, ..., n at the next moment under the condition that the current state is m.

[0071] Step 3: Generate PV state sequence

[0072] Unlike wind power, because the output of photovoltaic power is almost equal to 0 near sunrise, the initial state of photovoltaic power does not need to be randomly generated and can be directly specified.

[0073] Step 4: Generate photovoltaic output sequence

[0074] Because the photovoltaic power output is 0 before sunrise and after sunset, the daily sunrise and sunset times are used as the start and end times of the photovoltaic power output sequence simulation in this step. The photovoltaic state is converted into the specific photovoltaic output, and the conversion formula is as follows:

[0075] pv t =(q pvt +u pvt )pv N (18)

[0076] In the formula, pv t is the output value at time t in the newly generated photovoltaic output sequence.

[0077] Furthermore, in step S3, the energy storage scheduling strategy is as follows:

[0078] First, the entire system is calculated. This system refers to the power generation capacity margin of the entire system composed of power generation resources (including conventional power generation units, new energy units, etc.), energy storage equipment, transmission network and load. The calculation formula when only conventional power generation units are considered as available resources is given. It is specifically related to the available output of all conventional power generation units in the system. The calculation formula for the power generation capacity margin at time t is:

[0079]

[0080] Where N G is the collection of conventional generator sets in the system; d t is the load output at time t.

[0081] When the system's power generation capacity margin M t When it is greater than 0, the energy storage is in a charging state, and its charging power is limited by its own maximum charging power and total energy limit.

[0082] The specific charging process is described by the following formula:

[0083]

[0084] In the formula, is the charging power of energy storage e at time t; is the maximum charging power of energy storage e; e,t is the energy stored e at time t; is the energy upper limit of energy storage e. When the system's power generation capacity margin M t When it is less than 0, the energy storage is in a discharging state, and its discharge power is limited by its own maximum discharge power and the lower limit of the total energy.

[0085] The specific discharge process can be described by the following formula:

[0086]

[0087] In the formula, is the discharge power of energy storage e at time t; is the maximum discharge power of energy storage e; is the lower limit of energy storage e;

[0088] Determine the output sequence of energy storage based on margin / shortage and energy storage dispatch strategy.

[0089] Furthermore, when there are only conventional generator sets in the system resources, the specific scheduling steps of the energy storage units are as follows:

[0090] Step 1: Calculate the system generation capacity margin / shortage M t

[0091] According to formula (14), the system's power generation capacity margin / shortage is calculated based on the available output sequence of conventional generator sets and system load demand.

[0092] Step 2: Determine the energy storage charging and discharging power based on the system power generation capacity margin / shortage and energy storage charging and discharging constraints

[0093] According to the system's power generation capacity margin / shortage M t Determine the charging and discharging state of energy storage. When the system has a power generation capacity margin, that is, M t > 0, the energy storage is charged according to equations (15)-(18); when there is a shortage of power generation capacity in the system, that is, M t <0 energy storage is discharged according to equations (19)-(22).

[0094] Furthermore, during the discharge process, considering the system reliability, the energy storage discharge optimization strategy adopted is to minimize the system's LOLE hours. The specific scheduling steps are as follows:

[0095] Step 1: Calculate the system generation capacity margin / shortage M t

[0096] According to formula (14), the system's power generation capacity margin / shortage is calculated based on the available output sequence of conventional generator sets and system load demand;

[0097] Step 2: Determine the energy storage charging and discharging power according to the system power generation capacity margin / shortage and energy storage charging and discharging restriction conditions; when the system has power generation capacity margin, that is, M t >0 Energy storage is charged according to equations (15)-(18); when there is a shortage of power generation capacity in the system, that is, M t <0, calculate the current power generation capacity shortage duration T acs , the calculation formula is as follows:

[0098]

[0099] Where, T acs is the duration of the power generation capacity shortage; T c is the current moment; T is the total research period; t e is the end time of the power generation capacity shortage; this formula means finding a certain time t from the current time e So that in [T c ,t e ] The system power generation capacity margin is negative in all variables t e The maximum value in T c The difference is the duration of the generation capacity shortfall;

[0100] If the generation capacity shortage lasts for T acs When it is less than the duration of energy storage, discharge is performed according to equations (19)-(22) during the entire period of power generation capacity shortage.

[0101] When the generation capacity shortage lasts for a period of time T acs When it is greater than the duration of the energy storage, the capacity shortage at each moment within the duration of the power generation capacity shortage is sorted, as shown in the following formula (24), and then discharge is performed according to the final capacity shortage sequence and formulas (19)-(22). This discharge strategy prioritizes the energy storage to discharge at the moment when the capacity shortage is the smallest, so that the limited energy storage energy can reduce the load loss at as many moments as possible, so as to achieve the effect of reducing the number of system load loss hours.

[0102]

[0103] Where M c is the time-ordered sequence of the system generation capacity shortage from the current moment to the end of the generation capacity shortage duration; M scTo M c The sequence is sorted from small to large according to the power generation capacity deficit.

[0104] Furthermore, the specific calculation steps of the original reliability index value of the power generation resource system are as follows:

[0105] The system reliability index is calculated based on the available output sequence of conventional generator sets, the output sequence of new energy sources and the output sequence of energy storage. The reliability index includes the load loss probability LOLP, the load loss expectation value LOLE and the power shortage expectation value EENS.

[0106] Among them, LOLP (loss of load probability) is the time probability that the available power generation capacity cannot meet the load demand, that is, the probability that the power generation capacity margin is less than 0. Its specific calculation formula satisfies the following formula:

[0107] M=P ac -d (25)

[0108] LOLP=Pr(M<0) (26)

[0109] Where M is the power generation capacity margin; P ac is the available capacity for power generation; d is the load demand;

[0110] Among them, LOLE (loss of load expectation) is the conversion of LOLP calculation results into time. There are two ways to calculate annual LOLE, one is to use the daily maximum load curve of 365 days a year or the other is to use the daily load curve of 8760 hours. If the daily maximum load curve is used for calculation, the LOLE calculation formula satisfies the following formula, and the LOLE unit is day / year.

[0111] LOLE=LOLP×365 (27)

[0112] If the 8760-hour daily load curve is used, the LOLE calculation formula satisfies the following formula, and the LOLE unit is hours / year.

[0113] LOLE=LOLP×8760 (28)

[0114] Among them, EENS (Expected Electricity Shortage Value) refers to the expected value of the power loss caused by the user due to insufficient available capacity during the study period. The unit of this indicator is MW·h, and the calculation formula is as follows:

[0115]

[0116] In the formula, Pr(M t <0) is the probability that the power generation margin is less than 0 at time t; T is the research period in hours.

[0117] According to another aspect of the present invention, there is also provided a capacity credibility evaluation system for flexible resources participating in a capacity market, comprising:

[0118] A conventional generator set output sequence generation unit is used to model the conventional generator set and generate a number of available output sequences of the conventional generator set by sampling;

[0119] The output sequence generation unit of the new energy generating unit generates output sequences of several new energy generating units by sampling;

[0120] The energy storage output sequence generation unit determines the energy storage output sequence according to the energy storage scheduling strategy based on the available output sequence of the conventional generator set and the output sequence of the new energy generator set;

[0121] A system reliability index value calculation unit calculates the original reliability index value of the power generation resources to be evaluated in the system according to the available output sequence of conventional power generation units, the output sequence of new energy sources and the output sequence of energy storage units;

[0122] The capacity credibility calculation unit removes the power generation resources to be evaluated, connects a perfect unit of a certain capacity, performs system reliability calculation and records the corresponding reliability index value; corrects the capacity of the perfect unit connected to the system until the reliability index calculation result of the system after the perfect unit is added is consistent with the original reliability index value or less than the specified error; at this time, the capacity of the perfect unit added is the credible capacity of the power generation resources to be evaluated, and the credible capacity is divided by the installed capacity of the power generation resources to be evaluated to obtain the capacity credibility.

[0123] Furthermore, the conventional generator set output sequence generation unit performs modeling on the conventional generator set as follows:

[0124] During the operation of a conventional generator set, it will alternate between two different states: fault-free operation and fault shutdown for repair. The conversion frequency from the running state to the shutdown state and from the shutdown state to the running state is calculated. The specific calculation formula is as follows:

[0125]

[0126] In the formula, λ is the failure rate of the generator; MTTF is the mean time between failures; μ is the repair rate of the generator; MTTR is the mean repair time; MTTF and MTTR are obtained through historical data, and the specific calculation formula is as follows:

[0127]

[0128] Where N o is the total number of failures; is the normal working time of the generator set before the nth fault; is the downtime and maintenance time of the generator at the nth fault;

[0129] The duration of a conventional generator set in two states is approximately considered to obey an exponential distribution due to the randomness of the fault, satisfying the following formula:

[0130] TTF=-MTTF×ln(u1) (5)

[0131] TTR=-MTTR×ln(u2) (6)

[0132] Where TTF and TTR are the normal operation duration and fault maintenance duration respectively; u1 and u2 are random variables with uniform distribution in the interval [0,1].

[0133] The present invention effectively improves the operational safety and economy of the power system by proposing a method for comprehensively evaluating the capacity credibility of flexible resources participating in the capacity market. This method not only optimizes the dispatching strategy of new energy and energy storage resources and reduces operating costs, but also enhances the adaptability of the power grid to fluctuations in renewable energy, reduces environmental impacts, and promotes the use of green energy. In addition, the implementation of this invention will also help power markets and system operators to more accurately evaluate and utilize flexible resources, thereby promoting the power industry to develop in a more sustainable and environmentally friendly direction while ensuring the stability of power supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0134] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0135] The principles and features of the present invention are described below in conjunction with the accompanying drawings. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0136] refer to Figure 1 The present invention provides a capacity credibility evaluation method for flexible resources participating in the capacity market, which is characterized by comprising the following steps:

[0137] S1: Model the conventional units and generate a number of available output sequences of conventional generator sets by sampling;

[0138] S2: Model the new energy units and randomly sample and generate output sequences of several new energy units;

[0139] S3: Based on the available output sequence of the conventional generator sets and the output sequence of the new energy generator sets, determine the output sequence of the energy storage according to the energy storage scheduling strategy;

[0140] S4: Calculating the original reliability index value of the power generation resource to be evaluated in the system according to the available output sequence of conventional power generation units, the output sequence of new energy sources and the output sequence of energy storage;

[0141] S5: Remove the power generation resources to be evaluated, connect a perfect unit of a certain capacity (the perfect unit is used to represent an idealized power generation resource), calculate the system reliability and record the corresponding reliability index value; correct the capacity of the perfect unit connected to the system until the reliability index calculation result of the system after adding the perfect unit is consistent with the original reliability index value or less than the specified error; at this time, the capacity of the perfect unit added is the credible capacity of the power generation resources to be evaluated, and the credible capacity is divided by the installed capacity of the power generation resources to be evaluated to obtain the capacity credibility.

[0142] This solution intends to use the Monte Carlo method to establish a capacity credibility calculation method for flexible resources participating in the capacity market, generate the state sequence of conventional generators and the output sequence of new energy through sampling, and determine the charge and discharge state and charge and discharge power of energy storage according to the system's power generation capacity margin / shortage and the energy storage discharge strategy. The specific implementation of the solution of the present invention is divided into the following four major modules:

[0143] 1. Modeling of conventional generator sets

[0144] During the operation of traditional generator sets, they alternate between two different states: fault-free operation and fault shutdown for repair. This state transition occurs randomly, so in Monte Carlo simulation, it is necessary to generate a state change sequence through random sampling to determine the available output power sequence of the generator set. In order to model traditional generator sets, a two-state Markov model can be used.

[0145] The frequency of transitions between states is inversely proportional to the average dwell time in a particular state, that is, the frequency of transitions is equal to the inverse of the dwell time. Specifically, the frequency of transitions from the running state to the shutdown state and from the shutdown state to the running state can be calculated. The following is the specific calculation formula:

[0146]

[0147] Where λ is the failure rate of the generator; MTTF is the mean time to failure; μ is the repair rate of the generator; and MTTR is the mean time to repair.

[0148] Usually MTTF and MTTR are obtained through historical data. The specific calculation formula is as follows:

[0149]

[0150]

[0151] Where N o is the total number of failures; is the normal working time of the generator set before the nth fault; is the downtime and maintenance time of the generator at the nth fault;

[0152] The duration of a conventional generator set in two states is approximately considered to obey an exponential distribution due to the randomness of the fault, satisfying the following formula:

[0153] TTF=-MTTF×ln(u1) (5)

[0154] TTR=-MTTR×ln(u2) (6)

[0155] Where, TTF and TTR are normal operation duration and fault repair duration respectively. u1 and u2 are random variables with uniform distribution in the interval [0,1].

[0156] The state sequence {x1,x2,x3,…,x T}, the specific generation steps are as follows:

[0157] Step 1: Sample the duration of normal operation

[0158] Assuming that the initial state of all units is normal operation, a random number u1 is generated by sampling from a uniform distribution in the interval [0,1], and the corresponding normal operation duration is obtained using formula (5);

[0159] Step 2: Sample the duration of the troubleshooting state

[0160] A random number u2 is generated by sampling from a uniform distribution in the interval [0,1], and the corresponding fault repair duration is obtained using formula (6);

[0161] Step 3: Alternately generate the duration of each state until the length of the generator state sequence is equal to the length of the research period.

[0162] Repeat steps 1 and 2, alternately generate the duration of the normal operation state and the fault maintenance state, and determine the state of the conventional generator set at each moment until the length of the state sequence is greater than or equal to the total study period. If the length of the generated sequence is greater than or equal to the total study period, it is necessary to correct it and correct the length of the state sequence of the generator set to the length of the total study period.

[0163] Step 4: Generate the available output sequence of conventional generator sets

[0164] According to the state sequence and maximum output of conventional generator sets, it is converted into the available output sequence of conventional generator sets. The specific formula is as follows:

[0165]

[0166] Where P i,t is the available output of conventional generator set i at time t; x i,t is the state of the conventional generator set i at time t, 1 means normal startup, and 0 means fault shutdown; is the maximum output of conventional generator set i.

[0167] 2. New Energy Unit Modeling

[0168] At present, the modeling technology of renewable energy power generation series is mainly divided into two categories: direct method and indirect method. However, in actual operation, the output data of wind and solar power generation are often affected by problems such as bad data, wind / solar abandonment and data missing, which make it difficult for the quality of actual data to reach the ideal state. Therefore, some researchers tend to adopt the indirect method, which relies on wind speed or solar radiation intensity to build the model. The specific operation is to first generate the time series data of wind speed / solar radiation intensity, and then derive the power of wind and solar power generation based on the conversion relationship between wind speed and wind power generation power, solar radiation and solar power generation output. However, this indirect method may introduce certain conversion errors.

[0169] Therefore, some scholars have proposed modeling techniques based directly on power data, that is, directly using historical output data of wind or solar power generation to build models. Many researchers have modeled the power of renewable energy through techniques such as stochastic differential equations and Markov Chain Monte Carlo simulation (MCMC). Among these methods, MCMC is widely used in the modeling of renewable energy output time series because it can effectively capture the time series characteristics and autocorrelation of renewable energy output, and has been widely recognized by the industry.

[0170] This scheme considers using the Markov Chain Monte Carlo simulation MCMC method to simulate the output sequence of new energy units, which includes wind power output sequence and photovoltaic output sequence. The wind power output sequence is generated through the following steps:

[0171] Step 1: Discretize the historical wind power output into different states

[0172] By analyzing the historical data of wind power, according to the size of wind power output, the output at each time point is divided into N states. The original wind power output sequence {W1, W2, …, W T} is transformed into a state sequence {q1,q2,…,q T}. The corresponding conversion formula satisfies the following formula:

[0173]

[0174] In the formula, w N is the difference between the maximum and minimum values ​​of each state when wind power is divided into N states; W max , W min They are the maximum and minimum output of wind power during the study period.

[0175] Step 2: Generate wind power output state transition probability matrix

[0176] In the Markov chain model, the value of each state depends on the value of the previous state, and the states at adjacent moments satisfy the following formula:

[0177] q t+1 =Y·q t (10)

[0178] Where Y is the state transition probability matrix.

[0179] The number of changes between states is obtained through statistics of historical wind power output data, and the state transfer matrix Y is established accordingly. Each element in the matrix Y, Y r,s It indicates the probability that the wind power output will be transferred to s at the next moment under the condition that the current state is r, and the sum of the probabilities of each row should be 1. That is:

[0180] y r,s =Y(q t+1 =r|q t =s) (11)

[0181] And based on the state transfer probability matrix, a cumulative probability matrix is ​​generated to assist in the calculation. The specific calculation formula for each element in the cumulative probability transfer matrix is ​​as follows:

[0182]

[0183] In the formula, is the probability of transitioning to state 1, 2, ..., n at the next moment under the condition that the current state is m.

[0184] Step 3: Generate wind power state sequence

[0185] First, an integer u0 is sampled from a uniform distribution in the interval [1, N] to determine the initial state of wind power output. Then, a random number u1 is sampled from a uniform distribution in the interval [0, 1]; The relationship between u1 and u1 is used to determine the state of wind power at time t = 1. Then, the same method is used to generate the state of wind power at time t + 1 according to the state value at time t, until the wind power state at all times in the period to be studied is generated.

[0186] Step 4: Generate wind power output sequence

[0187] According to the wind power status, it is converted into specific wind power output. The conversion formula is as follows:

[0188] w t =(q t +u t ) N (13)

[0189] In the formula, w t is the output value at time t in the newly generated wind power output sequence;

[0190] The photovoltaic output sequence can be generated using the same steps, but since photovoltaic output has obvious time differences, the specific steps are slightly different. First, photovoltaics do not produce power at night, so when dividing different discrete states in the above step 1, only the valid data of photovoltaic output is taken into account; second, photovoltaics show an increasing trend in the morning period and a decreasing trend in the afternoon period. To reflect this characteristic of photovoltaics, when the photovoltaic output is discretized in step 1, the same output in the morning and afternoon periods is divided into different states; third, the photovoltaic output near sunrise is almost equal to 0, so the initial state of photovoltaics in step 3 does not need to be randomly generated and can be directly specified. Fourth, the photovoltaic output is 0 before sunrise and after sunset, so in step 4, the daily sunrise and sunset times are used as the start and end times of the photovoltaic output sequence simulation. The specific steps for generating the photovoltaic output sequence are as follows:

[0191] Step 1: Discretize the historical photovoltaic output into different states

[0192] By analyzing the historical data of wind power, according to the size of photovoltaic output, the output at each time point is divided into N states. Transformed into a state sequence {q pv1 ,q pv2 ,…,q pvT}. The corresponding conversion formula satisfies the following formula:

[0193]

[0194] In the formula, Pv is the difference between the maximum and minimum values ​​of each state when photovoltaic power generation is divided into N states; max , Pv min They are the maximum and minimum outputs of photovoltaic power generation during the study period.

[0195] Unlike wind power, photovoltaic power does not produce power at night. Therefore, when dividing different discrete states in the above steps, only the valid data of photovoltaic power output is taken into account; secondly, photovoltaic power shows an increasing trend in the morning period and a decreasing trend in the afternoon period. To reflect this characteristic of photovoltaic power, when discretizing the photovoltaic output in this step, the same output in the morning and afternoon periods needs to be divided into different states.

[0196] Step 2: Generate photovoltaic output state transition probability matrix

[0197] In the Markov chain model, the value of each state depends on the value of the previous state, and the states at adjacent moments satisfy the following formula:

[0198] q pv,t+1 =Y·q pv,t (16)

[0199] Where Y is the state transition probability matrix.

[0200] The number of changes between states is obtained through statistics of photovoltaic historical output data, and the state transfer matrix Y is established accordingly. Each element in the matrix Y, Y pv,r,s Indicates that the current state is r pv Under the condition of pv The probability of each row should sum to 1. That is:

[0201] y pv,r,s =Y(q pv,t+1 =r pv |q pv,t =s pv ) (17)

[0202] And based on the state transfer probability matrix, a cumulative probability matrix is ​​generated to assist in the calculation. The specific calculation formula for each element in the cumulative probability transfer matrix is ​​as follows:

[0203]

[0204] In the formula, is the probability of transitioning to state 1, 2, ..., n at the next moment under the condition that the current state is m.

[0205] Step 3: Generate PV state sequence

[0206] Unlike wind power, because the output of photovoltaic power is almost equal to 0 near sunrise, the initial state of photovoltaic power does not need to be randomly generated and can be directly specified.

[0207] Step 4: Generate photovoltaic output sequence

[0208] Because the photovoltaic power output is 0 before sunrise and after sunset, the daily sunrise and sunset times are used as the start and end times of the photovoltaic power output sequence simulation in this step. The photovoltaic state is converted into the specific photovoltaic output, and the conversion formula is as follows:

[0209] pv t =(q pvt +u pvt )pv N (18)

[0210] In the formula, pv t is the output value at time t in the newly generated photovoltaic output sequence.

[0211] 3. Energy storage charging and discharging model and its scheduling strategy

[0212] For energy storage, its charging and discharging process is not only affected by various aspects of the system at the current moment, but also needs to consider the impact of adjacent moments. In the reliability assessment of energy storage, the specific charging and discharging process needs to be considered to accurately consider the contribution of energy storage to system reliability. Its charging and discharging process is mainly affected by the available capacity of the system, its own maximum charging and discharging power, energy upper limit and energy at adjacent moments. The main function of energy storage equipment is to provide frequency regulation, peak regulation, emergency backup and other services to cope with the volatility and uncertainty of renewable energy power generation. When calculating the charging and discharging strategy of energy storage, it is first necessary to understand the system's power generation capacity margin when relying only on conventional generators without the participation of energy storage, so as to determine when the energy storage needs to be charged and discharged to supplement the system's power generation capacity.

[0213] In actual power system operation, the output of new energy sources fluctuates greatly and is difficult to predict, while the charging and discharging of energy storage is relatively easy to control. Therefore, in the initial energy storage scheduling strategy, the output of new energy sources is not considered first, and the power generation capacity margin is calculated based only on the output of conventional generators. This can simplify the calculation process and highlight the regulatory role of energy storage equipment in the system.

[0214] First, calculate the power generation capacity margin of the system, which refers to the overall system consisting of power generation resources (including conventional generators, new energy generators, etc.), energy storage equipment, transmission network and load. This solution gives the calculation formula when only conventional generators are considered as available resources. It is specifically related to the available output of all conventional generators in the system. The calculation formula for power generation capacity margin at time t is:

[0215]

[0216] Where N G is the collection of conventional generator sets in the system; d t is the load output at time t.

[0217] When the system's power generation capacity margin M t When it is greater than 0, the energy storage is in a charging state, and its charging power is limited by its own maximum charging power and total energy upper limit. The specific charging process can be described by the following formula:

[0218]

[0219] In the formula, is the charging power of energy storage e at time t; is the maximum charging power of energy storage e; e,t is the energy stored e at time t; is the energy upper limit of energy storage e. When the system's power generation capacity margin M t When it is less than 0, the energy storage is in a discharge state, and its discharge power is limited by its own maximum discharge power and the lower limit of total energy. The specific discharge process can be described by the following formula:

[0220]

[0221] In the formula, is the discharge power of energy storage e at time t; is the maximum discharge power of energy storage e; is the lower limit of energy storage e.

[0222] Determine the output sequence of energy storage based on margin / shortage and energy storage dispatch strategy.

[0223] When there are only conventional generators in the system resources, the specific scheduling steps of the energy storage units are as follows:

[0224] Step 1: Sampling and generating the available output sequence of conventional generator sets

[0225] According to the MTTF and MTTR of conventional generator sets, the available output sequence of conventional generator sets within the study period is generated by random sampling as described in “I. Modeling of conventional generator sets”.

[0226] Step 2: Calculate the system generation capacity margin / shortage M t

[0227] According to formula (14), the system's power generation capacity margin / shortage is calculated based on the available output sequence of conventional generator sets and system load demand.

[0228] Step 3: Determine the energy storage charging and discharging power based on the system power generation capacity margin / shortage and energy storage charging and discharging constraints

[0229] According to the system's power generation capacity margin / shortage M t Determine the charging and discharging state of energy storage. When the system has a power generation capacity margin, that is, M t> 0, the energy storage is charged according to equations (15)-(18); when there is a shortage of power generation capacity in the system, that is, M t <0 energy storage is discharged according to equations (19)-(22).

[0230] The above scheduling process for energy storage is equivalent to discharging immediately when the system has a power generation capacity shortage, which is in line with the natural reaction of energy storage. However, in the actual operation and scheduling process of energy storage, when energy storage faces a power generation capacity shortage in the system, it may adopt certain strategies to optimize its own charging and discharging sequence. Moreover, the above optimization strategies are mainly analyzed from the perspective of energy storage operation, and the optimization scheduling strategy of energy storage from the perspective of system reliability is not considered.

[0231] From the perspective of system reliability, the main factor that has an important impact on the contribution of energy storage reliability is its discharge process. Energy storage can dynamically change the discharge strategy adopted when there is a shortage of power generation capacity in the system, such as minimizing the number of LOLE hours, minimizing the power shortage at each moment, etc., thereby reducing the corresponding reliability evaluation index value and improving the credibility of energy storage capacity. In the discharge process, considering the system reliability, the energy storage discharge optimization strategy to be adopted in this plan is to minimize the number of LOLE hours of the system. The specific scheduling steps are given below:

[0232] 1) Calculate the system power generation capacity margin / shortage M t

[0233] According to formula (14), the system's power generation capacity margin / shortage is calculated based on the available output sequence of conventional generator sets and system load demand;

[0234] 2) Determine the energy storage charging and discharging power according to the system power generation capacity margin / shortage and energy storage charging and discharging restriction conditions. When the system has power generation capacity margin, that is, M t >0 Energy storage is charged according to equations (15)-(18); when there is a shortage of power generation capacity in the system, that is, M t <0, calculate the current power generation capacity shortage duration T acs , the calculation formula is as follows:

[0235]

[0236] Where, T acs is the duration of the power generation capacity shortage; T c is the current moment; T is the total research period; t e is the end time of the power generation capacity shortage; this formula means finding a certain time t from the current time e So that in [T c ,t e ] The system power generation capacity margin is negative in all variables t eThe maximum value in T c The difference is the duration of the generation capacity shortfall.

[0237] If the generation capacity shortage lasts for T acs When it is less than the duration of energy storage, discharge is performed according to equations (19)-(22) during the entire period of power generation capacity shortage.

[0238] When the generation capacity shortage lasts for a period of time T acs When it is greater than the duration of the energy storage, the capacity shortage at each moment within the duration of the power generation capacity shortage is sorted, as shown in the following formula (24), and then discharge is performed according to the final capacity shortage sequence and formulas (19)-(22). This discharge strategy prioritizes the energy storage to discharge at the moment when the capacity shortage is the smallest, so that the limited energy storage energy can reduce the load loss at as many moments as possible, so as to achieve the effect of reducing the number of system load loss hours.

[0239]

[0240] Where M c is the time-ordered sequence of the system generation capacity shortage from the current moment to the end of the generation capacity shortage duration; M sc To M c The sequence is sorted from small to large according to the power generation capacity deficit.

[0241] IV. Calculation of Capacity Credibility of Capacity Market Entities

[0242] This scheme intends to use the EFC indicator to evaluate the capacity credibility of the capacity market entities. Therefore, in addition to the necessary reliability calculations, it is also necessary to add the perfect unit capacity to the system that does not contain the subject to be evaluated and repeatedly try to change the corresponding capacity, so that the system containing the subject to be evaluated and the system with the perfect unit capacity added have the same reliability. In this process, a certain search method is needed to change the perfect unit capacity added to the original system to reduce the search time. Conventional methods include binary search or truncation method. Since the number of iterations required for the two in actual calculations is not much different, and the principle of binary search is simple, this scheme uses binary search to search for the added perfect unit capacity.

[0243] The specific steps for calculating the capacity credibility of capacity market entities are as follows:

[0244] Step 1: Calculate the system reliability index based on the available output sequence of conventional generator sets, the output sequence of new energy sources and the output sequence of energy storage; the reliability index includes the load loss probability LOLP, the expected value of load loss LOLE and the expected value of power shortage EENS.

[0245] In this step, the load loss of the system is calculated based on the output sequence of each resource in the system (including conventional units, new energy units and energy storage) and the following formulas (25)-(29) and the corresponding reliability index is recorded. When calculating the reliability of the new energy capacity, the reliability index is calculated based on the available output sequence of the conventional generator set and the new energy output sequence; when calculating the reliability of the energy storage capacity, the new energy output sequence is replaced by the output sequence of the energy storage; when calculating the capacity reliability of the new energy and energy storage combined entity, the new energy and energy storage output sequences are added to the system.

[0246] Among them, LOLP (loss of load probability) is the time probability that the available power generation capacity cannot meet the load demand, that is, the probability that the power generation capacity margin is less than 0. Its specific calculation formula satisfies the following formula:

[0247] M=P ac -d (25)

[0248] LOLP=Pr(M<0) (26)

[0249] Where M is the power generation capacity margin; P ac is the available capacity for power generation; d is the load demand;

[0250] Among them, LOLE (loss of load expectation) is the conversion of LOLP calculation results into time. There are two ways to calculate annual LOLE, one is to use the daily maximum load curve of 365 days a year or the other is to use the daily load curve of 8760 hours. If the daily maximum load curve is used for calculation, the LOLE calculation formula satisfies the following formula, and the LOLE unit is day / year.

[0251] LOLE=LOLP×365 (27)

[0252] If the 8760-hour daily load curve is used, the LOLE calculation formula satisfies the following formula, and the LOLE unit is hours / year.

[0253] LOLE=LOLP×8760 (28)

[0254] Among them, EENS (Expected Electricity Shortage Value) refers to the expected value of the power loss caused by the user due to insufficient available capacity during the study period. The unit of this indicator is MW·h, and the calculation formula is as follows:

[0255]

[0256] In the formula, Pr(M t <0) is the probability that the power generation margin is less than 0 at time t; T is the research period in hours.

[0257] Step 2: Calculate the reliability index after the perfect capacity is connected to the system

[0258] Remove the subject to be evaluated from the original system, connect a certain number of perfect units, calculate the system reliability and record the corresponding reliability indicators.

[0259] Step 3: Repeat step 2 and use the binary method to change the capacity of the perfect unit connected to the original system until the reliability index calculation result of the system after adding the perfect unit is consistent with that in step 1 or less than the limited error.

[0260] Step 4: The perfect unit capacity in step 3 is the credible capacity of the subject to be evaluated. Divide the credible capacity by the installed capacity of the subject to be evaluated to obtain the capacity credibility of the subject to be evaluated.

[0261] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A capacity credibility assessment method for flexible resources participating in the capacity market, characterized in that: The following steps are included: Modeling conventional power generating units, and generating several available output sequences of conventional power generating units by sampling; Sampling to generate output sequences of several new energy units; Based on the available output sequence of the conventional generator sets and the output sequence of the new energy generator sets, the output sequence of the energy storage is determined according to the energy storage dispatching strategy; The original reliability index value of the power generation resource to be evaluated is added to the calculation system according to the available output sequence of conventional power generation units, the output sequence of new energy sources and the output sequence of energy storage; Remove the power generation resources to be evaluated, connect a perfect unit of a certain capacity, calculate the system reliability and record the corresponding reliability index value; correct the capacity of the perfect unit connected to the system until the reliability index calculation result of the system after adding the perfect unit is consistent with the original reliability index value or less than the specified error; at this time, the capacity of the perfect unit added is the credible capacity of the power generation resources to be evaluated, and divide the credible capacity by the installed capacity of the power generation resources to be evaluated to obtain the capacity credibility.

2. A method for evaluating the capacity credibility of flexible resources participating in the capacity market as claimed in claim 1, characterized in that: The modeling process for conventional generator sets is as follows: During the operation of a conventional generator set, it will alternate between two different states: fault-free operation and fault shutdown for repair. The conversion frequency from the running state to the shutdown state and from the shutdown state to the running state is calculated. The specific calculation formula is as follows: In the formula, λ is the failure rate of the generator; MTTF is the mean time between failures; μ is the repair rate of the generator; MTTR is the mean repair time; MTTF and MTTR are obtained through historical data, and the specific calculation formula is as follows: Where N o is the total number of failures; is the normal working time of the generator set before the nth fault; is the downtime and maintenance time of the generator at the nth fault; The duration of a conventional generator set in two states is approximately considered to obey an exponential distribution due to the randomness of the fault, satisfying the following formula: TTF=-MTTF×ln(u1) (5) TTR=-MTTR×ln(u2) (6) Where TTF and TTR are the normal operation duration and fault maintenance duration respectively; u1 and u2 are random variables with uniform distribution in the interval [0,1].

3. A capacity credibility assessment method for flexible resources participating in a capacity market as claimed in claim 1, characterized in that: The steps for generating the available output sequence of conventional generator sets are as follows: 1) For conventional units, sample and generate the duration of normal operating status; Specifically, assuming that the initial state of all units is a normal operating state, a random number u1 is generated by sampling in a uniform distribution in the interval [0,1], and the corresponding normal operating duration is obtained using the following formula (5); TTF=-MTTF×ln(u1) (5) 2) Sample the duration of the fault repair state; Specifically, a random number u2 is generated by sampling from a uniform distribution in the interval [0,1], and the corresponding fault inspection duration is obtained using formula (6); TTR=-MTTR×ln(u2) (6) 3) Repeat steps 1) and 2) to alternately generate the duration of each state until the length of the generator set state sequence is equal to the length of the research period; 4) Generate the available output sequence of conventional generator sets According to the state sequence and maximum output of conventional generator sets, it is converted into the available output sequence of conventional generator sets. The specific formula is as follows: Where P i,t is the available output of conventional generator set i at time t; x i,t is the state of the conventional generator set i at time t, 1 means normal startup, and 0 means fault shutdown; is the maximum output of conventional generator set i.

4. A capacity credibility assessment method for flexible resources participating in a capacity market as claimed in claim 1, characterized in that: The Markov chain Monte Carlo simulation (MCMC) method is used to simulate the output sequence of the new energy generator set, wherein the output sequence of the new energy generator set includes a wind power output sequence and a photovoltaic output sequence; The wind power output sequence is generated through the following steps: Step 1: Discretize the historical wind power output into different states: By analyzing the historical data of wind power, according to the size of wind power output, the output at each time point is divided into N states. The original wind power output sequence {W1, W2, …, W T } is transformed into a state sequence {q1,q2,…,q T }; The corresponding conversion formula satisfies the following formula: In the formula, w N is the difference between the maximum and minimum values ​​of each state when wind power is divided into N states; W max , W min are the maximum and minimum output of wind power during the study period; Step 2: Generate wind power output state transition probability matrix: In the Markov chain model, the value of each state depends on the value of the previous state, and the states at adjacent moments satisfy the following formula: q t+1 =Y·q t (10) Where Y is the state transition probability matrix; The number of changes between states is obtained by counting the historical wind power output data, and the state transfer matrix Y is established accordingly; each element in the matrix Y, Y r,s It indicates the probability that the wind power output will be transferred to s at the next moment under the condition that the current state is r. The sum of the probabilities of each row should be 1, that is: y r,s =Y(q t+1 =r|q t =s) (11) And based on the state transfer probability matrix, a cumulative probability matrix is ​​generated to assist in the calculation. The specific calculation formula for each element in the cumulative probability transfer matrix is ​​as follows: In the formula, is the probability of transitioning to state 1, 2, ..., n at the next moment under the condition that the current state is m; Step 3: Generate wind power state sequence First, an integer u0 is sampled from a uniform distribution in the interval [1, N] to determine the initial state of wind power output; then a random number u1 is sampled from a uniform distribution in the interval [0, 1]; and then the random numbers u1 and u2 are compared in turn. The relationship between u1 and u2 is used to determine the state of wind power at time t=1; then the same method is used to generate the state of wind power at time t+1 according to the state value at time t, until the wind power state at all times in the period to be studied is generated; Step 4: Generate wind power output sequence According to the wind power status, it is converted into specific wind power output. The conversion formula is as follows: w t =(q t +u t )w N (13) In the formula, w t is the output value at time t in the newly generated wind power output sequence; The steps for generating the photovoltaic output sequence are as follows: Step 1: Discretize the historical photovoltaic output into different states By analyzing the historical data of photovoltaic power generation, the output of photovoltaic power generation at each time point is divided into N states according to the size of photovoltaic power generation. The original photovoltaic power generation sequence {Pv1, Pv2, …, Pv T } is transformed into a state sequence {q pv1 ,q pv2 ,…,q pvT }; The corresponding conversion formula satisfies the following formula: Where Pv N Pv is the difference between the maximum and minimum values ​​of each state when photovoltaic power generation is divided into N states; max , Pv min They are the maximum and minimum outputs of photovoltaic power generation during the study period; Step 2: Generate photovoltaic output state transition probability matrix In the Markov chain model, the value of each state depends on the value of the previous state, and the states at adjacent moments satisfy the following formula: q pv,t+1 =Y·q pv,t (16) Where Y is the state transition probability matrix; The number of changes between states is obtained by counting the historical photovoltaic output data, and the state transfer matrix Y is established accordingly; each element in the matrix Y, Y pv,r,s Indicates that the current state is r pv Under the condition of pv The probability of each row should sum to 1; that is: y pv,r,s =Y(q pv,t+1 =r pv |q pv,t =s pv ) (17) And based on the state transfer probability matrix, a cumulative probability matrix is ​​generated to assist in the calculation. The specific calculation formula for each element in the cumulative probability transfer matrix is ​​as follows: In the formula, is the probability of transitioning to state 1, 2, ..., n at the next moment under the condition that the current state is m; Step 3: Generate PV state sequence Because the output of photovoltaic power is almost equal to 0 near sunrise, the initial state of photovoltaic power generation does not need to be randomly generated and can be directly specified; Step 4: Generate photovoltaic output sequence: Because the output of photovoltaic power is 0 before sunrise and after sunset, the daily sunrise and sunset times are used as the start and end times of the photovoltaic output sequence simulation in this step; the photovoltaic state is converted into the specific photovoltaic output, and the conversion formula is as follows: pv t =(q pvt +u pvt )pv N (18) In the formula, pv t is the output value at time t in the newly generated photovoltaic output sequence.

5. A capacity credibility assessment method for flexible resources participating in a capacity market as claimed in claim 1, characterized in that: The scheduling strategy for energy storage is as follows: First, the power generation capacity margin of the entire system is calculated, and the calculation formula when only conventional generators are considered as available resources is given. It is specifically related to the available output of all conventional generators in the system. The calculation formula for the power generation capacity margin at time t is: Where N G is the collection of conventional generator sets in the system; d t is the load output at time t; When the system's power generation capacity margin M t When it is greater than 0, the energy storage is in charging state, and its charging power is limited by its maximum charging power and total energy limit; The specific charging process is described by the following formula: In the formula, is the charging power of energy storage e at time t; is the maximum charging power of energy storage e; E e,t is the energy stored e at time t; is the energy upper limit of energy storage e; When the system's power generation capacity margin M t When it is less than 0, the energy storage is in a discharging state, and its discharge power is limited by its own maximum discharge power and the lower limit of total energy; The specific discharge process can be described by the following formula: In the formula, is the discharge power of energy storage e at time t; is the maximum discharge power of energy storage e; is the lower limit of energy storage e; Determine the output sequence of energy storage based on margin / shortage and energy storage dispatch strategy.

6. A capacity credibility assessment method for flexible resources participating in a capacity market as claimed in claim 5, characterized in that: When there are only conventional generators in the system resources, the specific scheduling steps of the energy storage units are as follows: Step 1: Calculate the system generation capacity margin / shortage M t According to formula (14), the system's power generation capacity margin / shortage is calculated based on the available output sequence of conventional generator sets and system load demand; Step 2: Determine the energy storage charging and discharging power based on the system power generation capacity margin / shortage and energy storage charging and discharging constraints According to the system's power generation capacity margin / shortage M t Determine the charging and discharging state of energy storage. When the system has a power generation capacity margin, that is, M t > 0, the energy storage is charged according to equations (15)-(18); when there is a shortage of power generation capacity in the system, that is, M t <0 energy storage is discharged according to equations (19)-(22).

7. A capacity credibility assessment method for flexible resources participating in a capacity market as claimed in claim 5, characterized in that: During the discharge process, considering the system reliability, the energy storage discharge optimization strategy adopted is to minimize the system's LOLE hours. The specific scheduling steps are as follows: Step 1: Calculate the system generation capacity margin / shortage M t According to formula (14), the system's power generation capacity margin / shortage is calculated based on the available output sequence of conventional generator sets and system load demand; Step 2: Determine the energy storage charging and discharging power according to the system power generation capacity margin / shortage and energy storage charging and discharging restriction conditions; when the system has power generation capacity margin, that is, M t >0 Energy storage is charged according to equations (15)-(18); when there is a shortage of power generation capacity in the system, that is, M t <0, calculate the current power generation capacity shortage duration T acs , the calculation formula is as follows: Where, T acs is the duration of the power generation capacity shortage; T c is the current moment; T is the total research period; t e is the end time of the power generation capacity shortage; this formula means finding a certain time t from the current time e So that in [T c ,t e ] The system power generation capacity margin is negative in all variables t e The maximum value in T c The difference is the duration of the generation capacity shortfall; If the generation capacity shortage lasts for T acs When it is less than the duration of energy storage, discharge is performed according to equations (19)-(22) during the entire duration of the power generation capacity shortage; When the generation capacity shortage lasts for a period of time T acs When it is greater than the duration of the energy storage, the capacity shortage at each moment within the duration of the power generation capacity shortage is sorted, as shown in the following formula (24), and then discharge is performed according to the final capacity shortage sequence and formulas (19)-(22); this discharge strategy prioritizes the energy storage to discharge at the moment when the capacity shortage is the smallest, so that the limited energy storage energy can reduce the load loss at as many moments as possible, so as to achieve the effect of reducing the number of system load loss hours; Where M c is the time-ordered sequence of the system generation capacity shortage from the current moment to the end of the generation capacity shortage duration; M sc To M c The sequence is sorted from small to large according to the power generation capacity deficit.

8. A capacity credibility assessment method for flexible resources participating in a capacity market as claimed in claim 1, characterized in that: The specific calculation steps of the original reliability index value of the power generation resource system are as follows: The system reliability index is calculated based on the available output sequence of conventional generator sets, the output sequence of new energy sources and the output sequence of energy storage. The reliability index includes the load loss probability LOLP, the load loss expectation value LOLE and the power shortage expectation value EENS. Among them, LOLP (loss of load probability) is the time probability that the available power generation capacity cannot meet the load demand, that is, the probability that the power generation capacity margin is less than 0; its specific calculation formula satisfies the following formula: M=P ac -d (25) LOLP=Pr(M<0) (26) Where M is the power generation capacity margin; P ac is the available capacity for power generation; d is the load demand; Among them, LOLE (loss of load expectation) is the conversion of LOLP calculation results into time. There are two methods to calculate annual LOLE, namely, using the daily maximum load curve of 365 days a year or using the daily load curve of 8760 hours. If the daily maximum load curve is used for calculation, the LOLE calculation formula satisfies the following formula, and the LOLE unit is day / year; LOLE=LOLP×365 (27) If the 8760-hour daily load curve is used, the LOLE calculation formula satisfies the following formula, and the LOLE unit is hours / year; LOLE=LOLP×8760 (28) Among them, EENS (Expected Electricity Shortage Value) refers to the expected value of the power loss caused by the user due to insufficient available capacity during the study period. The unit of this indicator is MW·h, and the calculation formula is as follows: In the formula, Pr(M t <0) is the probability that the power generation margin is less than 0 at time t; T is the research period in hours.

9. A capacity credibility assessment system for flexible resources participating in the capacity market, characterized in that: include, A conventional generator set output sequence generation unit is used to model the conventional generator set and generate a number of available output sequences of the conventional generator set by sampling; The output sequence generation unit of the new energy generating unit generates output sequences of several new energy generating units by sampling; The energy storage output sequence generation unit determines the energy storage output sequence according to the energy storage scheduling strategy based on the available output sequence of the conventional generator set and the output sequence of the new energy generator set; A system reliability index value calculation unit calculates the original reliability index value of the power generation resources to be evaluated in the system according to the available output sequence of conventional power generation units, the output sequence of new energy sources and the output sequence of energy storage units; The capacity credibility calculation unit removes the power generation resources to be evaluated, connects a perfect unit of a certain capacity, performs system reliability calculation and records the corresponding reliability index value; corrects the capacity of the perfect unit connected to the system until the reliability index calculation result of the system after the perfect unit is added is consistent with the original reliability index value or less than the specified error; at this time, the capacity of the perfect unit added is the credible capacity of the power generation resources to be evaluated, and the credible capacity is divided by the installed capacity of the power generation resources to be evaluated to obtain the capacity credibility.

10. A capacity credibility evaluation system for flexible resources participating in a capacity market as claimed in claim 9, characterized in that: The conventional generator set output sequence generation unit performs modeling of the conventional generator set as follows: During the operation of a conventional generator set, it will alternate between two different states: fault-free operation and fault shutdown for repair. The conversion frequency from the running state to the shutdown state and from the shutdown state to the running state is calculated. The specific calculation formula is as follows: In the formula, λ is the failure rate of the generator; MTTF is the mean time between failures; μ is the repair rate of the generator; MTTR is the mean repair time; MTTF and MTTR are obtained through historical data, and the specific calculation formula is as follows: Where N o is the total number of failures; is the normal working time of the generator set before the nth fault; is the downtime and maintenance time of the generator at the nth fault; The duration of a conventional generator set in two states is approximately considered to obey an exponential distribution due to the randomness of the fault, satisfying the following formula: TTF=-MTTF×ln(u1) (5) TTR=-MTTR×ln(u2) (6) Where TTF and TTR are the normal operation duration and fault maintenance duration respectively; u1 and u2 are random variables with uniform distribution in the interval [0,1].

Images