Strong-randomness wind power plant confidence capacity calculation method based on acceleration time sequence Monte Carlo method

Abstract

The invention relates to a strong-randomness wind power plant confidence capacity calculation method based on an acceleration time sequence Monte Carlo method. The method comprises the following steps: step 1, determining key indexes for evaluating the power generation reliability of a wind power plant; 2, constructing a reliability index calculation model of the wind power plant power generation system based on the key indexes determined in the step 1; 3, performing reliability evaluation calculation on the reliability index calculation model of the wind power plant power generation system constructed in the step 2 by using an acceleration time sequence Monte Carlo method; and 4, solving estimated capacities under different confidence probabilities by combining a reliability evaluation calculation result obtained in the step 3 and utilizing an effective load capacity evaluation method to obtain the capacity margin of the wind power plant. According to the method, state transition of new energy and a conventional unit can be simulated through acceleration time sequence Monte Carlo simulation, and the capability of coping with randomness and volatility of wind power is enhanced.

Description

Technical Field

[0001] The invention belongs to the technical field of power system confidence capacity calculation, and relates to a confidence capacity calculation method for a strongly random wind farm, in particular to a confidence capacity calculation method for a strongly random wind farm based on an accelerated sequential Monte Carlo method. Background Art

[0002] Building a new power system with a high proportion of new energy as the main body is one of the important measures for my country to promote green and low-carbon transformation of energy production and consumption, and is also a key link in accelerating the construction of a clean, low-carbon, safe and efficient energy system. This transformation not only involves the optimization of the energy structure, but also concerns the country's sustainable development, ecological and environmental protection, and high-quality economic development.

[0003] However, the randomness, volatility and intermittency of high-proportion renewable energy and new loads have brought severe challenges to the structural form of the existing power system and the power market model. These characteristics make it impossible for the traditional power system to effectively cope with the demand for efficient consumption of renewable energy, and also make the safe and reliable access of new loads complicated and difficult. With the continuous increase in the proportion of renewable energy, the power system is facing frequent fluctuations in supply and demand, which greatly affects the stability and economy of power supply.

[0004] In this context, it is particularly important to develop a method that can accurately evaluate the confidence capacity of wind farms. As an important renewable energy source, wind power generation is affected by many factors such as wind speed, climate and environment, and has strong uncertainty. This uncertainty not only increases the difficulty of power system scheduling, but also poses a threat to the safety of its operation.

[0005] At present, the main indicators used to evaluate wind power confidence capacity include effective load capacity (ELCC) and equivalent conventional unit capacity (ECUC). Both indicators rely on the reliability calculation of the system, and their calculation methods are mainly analytical methods. Although the analytical method is numerically stable, it usually equates intermittent renewable energy to a multi-derating state unit, ignoring its timing characteristics.

[0006] Therefore, in order to solve the above problems, the present invention proposes a method for calculating the confidence capacity of a strongly random wind farm based on an accelerated sequential Monte Carlo method.

[0007] After searching, no public documents of the prior art that are identical or similar to the present invention were found. Summary of the invention

[0008] The purpose of the present invention is to overcome the shortcomings of the prior art and propose a method for calculating the confidence capacity of a strongly random wind farm based on the accelerated Monte Carlo method. The accelerated Monte Carlo simulation can simulate the state transition of new energy and conventional units, thereby enhancing the ability to cope with the randomness and volatility of wind power.

[0009] The present invention solves the practical problem by adopting the following technical solutions:

[0010] A method for calculating the confidence capacity of a strongly random wind farm based on an accelerated sequential Monte Carlo method comprises the following steps:

[0011] Step 1: Determine the key indicators for evaluating the reliability of wind farm power generation;

[0012] Step 2: Based on the key indicators determined in step 1, a reliability index calculation model for the wind farm power generation system is constructed;

[0013] Step 3: Using the accelerated sequential Monte Carlo method, a reliability evaluation calculation is performed on the reliability index calculation model of the wind farm power generation system constructed in step 2;

[0014] Step 4: Combine the reliability assessment calculation results obtained in step 3, use the effective load capacity assessment method to solve the estimated capacity under different confidence probabilities, and obtain the capacity margin of the wind farm.

[0015] Moreover, the reliability evaluation index in step 1 includes:

[0016] (1) Probability parameter LOLP:

[0017] The calculation formula of LOLP is as follows:

[0018] LOLP=P(X≥CL)

[0019] Among them, P is the probability coefficient, X is the system's outage capacity, L is the daily maximum load, and C is the system's total installed capacity.

[0020] (2) Expected parameters:

[0021] ① Low power level LOLE

[0022] The calculation formula of LOLE is as follows:

[0023] LOLE=p[X≥(CL)]

[0024] Among them, p is the outage coefficient, X is the outage capacity of the system, L is the peak load, and C is the total installed capacity of the system.

[0025] The level of power shortage for the whole year is as follows:

[0026]

[0027] In the formula, C i is the system installed capacity in the i-th time period; and p[X≥(C i -L ijk )] represents the probability that the outage capacity is greater than or equal to the backup capacity in the i-th time period; m represents the number of time periods in a year; n i Represents the number of days in the i-th time period; L ijk It represents the peak load at hour K in the i-th time period of the j-th day.

[0028] ②Battery level EENS

[0029] The calculation formula of EENS is as follows:

[0030] EENS=(XR)×P(X)

[0031] Where P(X) is the power shortage coefficient, X is the system's outage capacity, and R is the system's spare resource capacity, that is, the system's installed capacity.

[0032] The power shortage level for a whole year is as follows:

[0033]

[0034] Where, L ijk represents the peak load at time K in the i-th time period of the j-th day; and P ijk (X) represents the probability that the outage capacity is greater than or equal to X at the Kth time in the i-th time period on the j-th day; m represents the number of time periods in a year; n i Indicates the number of days included in the i-th time period.

[0035] Moreover, the wind farm power generation system reliability index calculation model in step 2 includes: a conventional unit output model, a wind farm output model and a load curve fluctuation model;

[0036] (1) Conventional unit output model: only the normal operation and fault shutdown states are considered, and the normal operation duration t is used. 1 and fault repair time t 2 Description, usually 1 and t 2 It obeys exponential distribution, and its formula is as follows:

[0037]

[0038]

[0039] Where: λ is the failure rate; μ is the repair rate; t MTTFis the average working time; t MTTR is the mean repair time, both units are h, so t 1 and t 2 The unit is also h; u 1 and u 2 is a random number uniformly distributed between [0,1].

[0040] (2) Wind farm output model: The output power of wind turbines is directly related to wind speed, so the power simulation problem can be transformed into a wind speed simulation problem. The wind power output model is as follows:

[0041]

[0042]

[0043] Where v is the wind speed, v ci is the cut-in wind speed, v co is the cut-out wind speed, v r is the rated wind speed, P sw Output power for wind farms.

[0044] (3) Load curve fluctuation model: When constructing the load model, the change of load within the simulation period is not considered. The load model uses the annual peak load as the benchmark. The calculation formula is as follows:

[0045]

[0046] In the formula, is the annual peak load, P w (t) is the percentage of weekly peak load to annual peak load, P d (t) is the percentage of daily peak load to weekly peak load, P h (t) is the percentage of hourly peak load to daily peak load.

[0047] Since load forecasting has certain uncertainty, a random variable that obeys normal distribution can be superimposed on the original load model to correct the original load. At this time, the load curve fluctuation model can be expressed as:

[0048]

[0049] Moreover, the specific steps of step 3 include:

[0050] (1) Use the accelerated sequential Monte Carlo method to simulate the output model of the wind farm and calculate the reliability index of the system by simulating the random process of the wind farm. With a sufficiently long simulation time, a reliability index close to the expected value is obtained.

[0051] At this time, the reliability calculation of the system is as follows:

[0052]

[0053] Where I represents the reliability index of the system, which can be the lack of power time (LOLP) or the lack of power level (LOLE), etc.; T represents the simulation time, x t represents the state of the system component at time t, and f(x) is used to describe the state of the component at the time x. t When the simulation time is long enough, the reliability index will approach the expected value I, thus obtaining the reliability index of the system.

[0054] (2) Convert the continuous process into a cumulative summation method to continue formula optimization. The specific calculation formula is as follows:

[0055]

[0056] (3) performing reliability evaluation calculation on the wind farm power generation system reliability index calculation model constructed in step 2;

[0057] Moreover, the specific steps of step 3 (3) include:

[0058] First, the reliability assessment process of the wind farm system starts with inputting relevant data and then setting the sampling year to 1;

[0059] Next, the output curves of conventional units and wind turbines are sampled respectively, including the annual output curve of conventional units, the annual time-series wind speed intensity sequence, and the outage status of wind turbines; based on these data, the time-series output curve of wind turbines is calculated and integrated to form the annual power generation capacity curve;

[0060] Subsequently, the system will compare the power balance on an hourly basis and calculate the system reliability index; if the reliability index does not reach the convergence standard, the process will check whether it exceeds the maximum age. If not, the sampling years will be increased and the above process will be repeated.

[0061] Once the reliability index converges or reaches the maximum age, the system will output the final reliability index, thus completing a comprehensive assessment of the reliability of the system containing the wind farm.

[0062] Moreover, the specific method of step 4 is:

[0063] Evaluate the load that the wind farm can carry, compare the system reliability before and after the wind farm is connected, and calculate the confidence capacity of the wind farm:

[0064] Assume that the reliability of the original system is R 0 When the system is connected to intermittent energy, the reliability level rises to R 1At this time, if the grid load is gradually increased, the system reliability will gradually decrease; when the load level drops to a certain level, the reliability level will return to R 0 , as shown below:

[0065] R 0 =f 0 (G,L)=f 1 (G+G int ,L)

[0066] Inverse the distribution on both sides of the above equation to get:

[0067] L′=f 1 -1 (G+G int ,R)

[0068] Therefore, the confidence capacity CC of the wind farm is ELCC for:

[0069]

[0070] In the formula, G int is the rated capacity of the wind farm, R is the reliability of the system, and L is the load.

[0071] Advantages and beneficial effects of the present invention:

[0072] The present invention proposes a method for calculating the confidence capacity of a strongly random wind farm based on an accelerated time series Monte Carlo method. By improving the traditional analytical method, the state transfer of new energy and conventional units can be simulated more accurately. This method enhances the ability to cope with the randomness and volatility of wind farm power generation and improves the accuracy of the evaluation. It overcomes the defects of the prior art that the time series characteristics are ignored, the ability to cope with a high proportion of new energy systems is insufficient, and the calculation efficiency is low. First, the key evaluation indicators of the reliability of wind farm power generation are determined, and these indicators can fully reflect the power generation performance of the wind farm under different conditions. According to the determined evaluation indicators, a calculation model is constructed, which can simulate the power generation behavior of the wind farm under various conditions. Then, the output model of the wind farm is simulated by the accelerated time series Monte Carlo method to evaluate its reliability at different confidence levels. Finally, combined with the reliability evaluation results of the wind farm, the effective load capacity evaluation method is used to calculate the estimated capacity under different confidence probabilities, thereby obtaining the capacity margin of the wind farm. The present invention can enhance the reliability and stability of wind farms in the power grid, enabling power operators to more clearly understand the power generation capacity of wind farms under different meteorological conditions, thereby optimizing the allocation of power resources. This not only helps to achieve the effective consumption of a high proportion of new energy, but also provides a solid foundation for the construction of a new power market. Ultimately, promoting the access of a high proportion of new energy will provide strong support for the sustainable development of my country's power system, improving energy utilization efficiency, and achieving a low-carbon economic transformation. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 It is a flow chart of a method for calculating the confidence capacity of a strongly random wind farm based on an accelerated sequential Monte Carlo method according to the present invention;

[0074] Figure 2 is a reliability assessment flow chart of a wind farm system according to the present invention;

[0075] Figure 3 It is a flow chart of solving the wind farm confidence capacity calculation model of the present invention. DETAILED DESCRIPTION

[0076] The embodiments of the present invention are further described in detail below with reference to the accompanying drawings:

[0077] A confidence capacity calculation method for strongly random wind farms based on the accelerated time series Monte Carlo method, such as Figures 1 to 3 As shown, the following steps are included:

[0078] Step 1: Determine the key indicators for evaluating the reliability of wind farm power generation;

[0079] The reliability evaluation indicators in step 1 include:

[0080] First, it is necessary to determine the key indicators for evaluating the reliability of wind farm power generation. These indicators include probabilistic parameters such as LOLP (time underpower probability) and expected parameters such as LOLE (level of underpower) and EENS (expected unsupplied energy). These parameters will help quantify the reliability of wind farms under different circumstances.

[0081] (2) Probability parameter LOLP:

[0082] LOLP (Loss of Load Probability) is mainly used to describe the possibility of failure in the power system. This indicator is used to indicate the proportion of time occupied by the load gap caused by equipment failure.

[0083] The calculation formula of LOLP is as follows:

[0084] LOLP=P(X≥CL)

[0085] Where P is the probability coefficient, X is the system's outage capacity, L is the daily maximum load, and C is the system's total installed capacity.

[0086] (3) Expected parameters:

[0087] ① Low power level LOLE

[0088] LOLE (Loss of Load Expectation) represents the probability of system outage and is considered when it is greater than or equal to the system's reserve capacity. The system's reserve capacity is defined as the difference between the installed capacity and the load. Typically, the system load model is analyzed using the daily peak load curve.

[0089] The calculation formula of LOLE is as follows:

[0090] LOLE=p[X≥(CL)]

[0091] Among them, p is the outage coefficient, X is the outage capacity of the system, L is the peak load, and C is the total installed capacity of the system.

[0092] The power shortage level for the whole year (unit: h / a) is as follows:

[0093]

[0094] In the formula, C i is the system installed capacity in the i-th time period; and p[X≥(C i -L ijk )] represents the probability that the outage capacity is greater than or equal to the backup capacity in the i-th time period; m represents the number of time periods in a year; n i Represents the number of days in the i-th time period; L ijk It represents the peak load at hour K in the i-th time period of the j-th day.

[0095] ②Battery level EENS

[0096] EENS (Expected Energy Not Supplied) refers to the expected value of power shortage for users due to the forced shutdown of the unit. This indicator can effectively reflect the severity of system failures, because the reliability of the power system is not only affected by power shortages, but also by the lack of power.

[0097] The calculation formula of EENS is as follows:

[0098] EENS=(XR)×P(X)

[0099] Where P(X) is the power shortage coefficient, X is the system's outage capacity, and R is the system's spare resource capacity, that is, the system's installed capacity.

[0100] The power shortage level for a whole year is as follows:

[0101]

[0102] Where, L ijkrepresents the peak load at time K in the i-th time period of the j-th day; and P ijk (X) represents the probability that the outage capacity is greater than or equal to X at the Kth time in the i-th time period on the j-th day; m represents the number of time periods in a year; n i Indicates the number of days included in the i-th time period.

[0103] Step 2: Based on the key indicators determined in step 1, a reliability index calculation model for the wind farm power generation system is constructed;

[0104] The wind farm power generation system reliability index calculation model in step 2 includes: a conventional unit output model, a wind farm output model and a load curve fluctuation model;

[0105] After determining the key indicators, a mathematical model is constructed to calculate these indicators. Intermittent energy is equivalent to a multi-state unit. The reliability models of each system when calculating the reliability of intermittent energy using the Monte Carlo simulation method are given, including: conventional unit output model, wind farm output model, and load model. These models will simulate the output of wind farms under different wind speed conditions and load fluctuations.

[0106] (1) Conventional unit output model: In order to simplify the calculation, only the normal operation and fault shutdown are considered, and the normal operation duration t 1 and fault repair time t 2 Description, usually 1 and t 2 It obeys exponential distribution, and its formula is as follows:

[0107]

[0108]

[0109] Where: λ is the failure rate; μ is the repair rate; t MTTF is the average working time; t MTTR is the mean repair time, both units are h, so t 1 and t 2 The unit is also h; u 1 and u 2 is a random number uniformly distributed between [0,1].

[0110] (2) Wind farm output model: The wind turbine output power is directly related to the wind speed, so the power simulation problem can be transformed into a wind speed simulation problem. The wind power output model is as follows:

[0111]

[0112]

[0113] where v is the wind speed, v ci is the cut-in wind speed, v co is the cut-out wind speed, v r is the rated wind speed, P sw is the output power of the wind farm.

[0114] In this embodiment, the output model of the wind farm can be divided into three states:

[0115] UP1 state: The wind turbine is in normal operation, but the wind speed is lower than the rated wind speed (Vt < Vr). In this state, the wind turbine can generate electricity, but it has not reached its maximum production capacity.

[0116] UP2 state: The wind turbine is also in normal operation, but the wind speed is higher than the rated wind speed (Vt > Vr). In this state, the wind turbine can operate at the rated power (Pt = P1) and reach its maximum production capacity.

[0117] DOWN state: The wind turbine is in a faulty or shutdown state. At this time, the output power of the wind turbine is zero (Pt = 0). Due to maintenance, faults or other reasons, the wind turbine cannot generate electricity.

[0118] (3) Load curve fluctuation model: When constructing the load model, the change of the load within the simulation time limit is not considered. The load model uses the annual peak load as the benchmark, and the calculation formula is as follows:

[0119]

[0120] In the formula, is the annual peak load, P w (t) is the percentage of the weekly peak load to the annual peak load, P d (t) is the percentage of the daily peak load to the weekly peak load, P h (t) is the percentage of the hourly peak load to the daily peak load.

[0121] Since load forecasting has a certain degree of uncertainty, a random variable obeying a normal distribution can be superimposed on the original load model to correct the original load. At this time, the load curve fluctuation model can be expressed as:

[0122]

[0123] Step 3: Use the accelerated time series Monte Carlo method to perform reliability evaluation calculations on the reliability index calculation model of the wind farm power generation system constructed in Step 2;

[0124] The specific steps of Step 3 include:

[0125] (1) Use the accelerated sequential Monte Carlo method to simulate the output model of the wind farm and calculate the reliability index of the system by simulating the random process of the wind farm. With a sufficiently long simulation time, a reliability index close to the expected value is obtained.

[0126] At this time, the reliability calculation of the system is as follows:

[0127]

[0128] Where I represents the reliability index of the system, which can be the lack of power time (LOLP) or the lack of power level (LOLE). T represents the simulation time, x t represents the state of the system component at time t, and f(x) is used to describe the state of the component at the time x. t When the simulation time is long enough, the reliability index will approach the expected value I, thus obtaining the reliability index of the system.

[0129] (2) The above solution method is a continuous process and is not convenient to solve. Therefore, the discretization method can be used to convert the continuous process into a cumulative summation method to continue formula optimization. The specific calculation formula is as follows:

[0130]

[0131] (3) Perform reliability evaluation calculation on the wind farm power generation system reliability index calculation model constructed in step 2, such as Figure 2 As shown, the calculation process is as follows:

[0132] First, the reliability assessment process of the wind farm system starts with inputting relevant data and then setting the sampling year to 1;

[0133] Next, the output curves of conventional units and wind turbines are sampled respectively, including the annual output curve of conventional units, the annual time-series wind speed intensity sequence, and the outage status of wind turbines; based on these data, the time-series output curve of wind turbines is calculated and integrated to form the annual power generation capacity curve;

[0134] Subsequently, the system will compare the power balance on an hourly basis and calculate the system reliability index; if the reliability index does not reach the convergence standard, the process will check whether it exceeds the maximum age. If not, the sampling years will be increased and the above process will be repeated.

[0135] Once the reliability index converges or reaches the maximum age, the system will output the final reliability index, thus completing a comprehensive assessment of the reliability of the system containing the wind farm.

[0136] In this embodiment, this process ensures the accuracy and reliability of the evaluation results through iterative calculation.

[0137] Step 4: Combine the reliability assessment calculation results obtained in step 3, use the effective load capacity assessment method to solve the estimated capacity under different confidence probabilities, and obtain the capacity margin of the wind farm;

[0138] The specific method of step 4 is:

[0139] Evaluate the load that the wind farm can carry, compare the system reliability before and after the wind farm is connected, and calculate the confidence capacity of the wind farm:

[0140] Effective Load Carry Capability (ELCC) is considered to be the preferred standard for evaluating the confidence capacity of new power sources in the system. The definition of effective load capacity refers to the ratio of the load that the new power source can carry to the capacity of the new power source under the condition that the reliability level of the system remains unchanged after the power system is connected to the new power source. As a new energy source, the confidence capacity of wind farms can also be measured by the above evaluation method.

[0141] Assume that the reliability of the original system is R 0 When the system is connected to intermittent energy, the reliability level rises to R 1 At this time, if the grid load is gradually increased, the system reliability will gradually decrease; when the load level drops to a certain level, the reliability level will return to R 0 , as shown below:

[0142] R 0 =f 0 (G,L)=f 1 (G+G int ,L)

[0143] Inverse the distribution on both sides of the above equation to get:

[0144] L′=f 1 -1 (G+G int ,R)

[0145] Therefore, the confidence capacity CC of the wind farm is ELCC for:

[0146]

[0147] In the formula, G int is the rated capacity of the wind farm, R is the reliability of the system, and L is the load.

[0148] Then the calculation results of wind farm capacity are completed as shown in Table 1.

[0149] Table 1 Confidence capacity assessment results of some wind farms to be built in a certain area

[0150]

[0151] It should be emphasized that the embodiments of the present invention are illustrative rather than restrictive. Therefore, the present invention includes but is not limited to the embodiments described in the specific implementation modes. Any other implementation modes derived by those skilled in the art based on the technical solutions of the present invention also fall within the scope of protection of the present invention.

Claims

1. A method for calculating the confidence capacity of a strongly random wind farm based on an accelerated sequential Monte Carlo method, characterized in that: The following steps are involved: Step 1: Determine the key indicators for evaluating the reliability of wind farm power generation; Step 2: Based on the key indicators determined in step 1, a reliability index calculation model for the wind farm power generation system is constructed; Step 3: Using the accelerated sequential Monte Carlo method, a reliability evaluation calculation is performed on the reliability index calculation model of the wind farm power generation system constructed in step 2; Step 4: Combine the reliability assessment calculation results obtained in step 3, use the effective load capacity assessment method to solve the estimated capacity under different confidence probabilities, and obtain the capacity margin of the wind farm.

2. The method for calculating the confidence capacity of a strongly random wind farm based on the accelerated sequential Monte Carlo method according to claim 1, characterized in that: The reliability evaluation indicators in step 1 include: (1) Probability parameter LOLP: The calculation formula of LOLP is as follows: LOLP=P(X≥CL) Among them, P is the probability coefficient, X is the system's outage capacity, L is the daily maximum load, and C is the system's total installed capacity; (2) Expected parameters: ① Low power level LOLE The calculation formula of LOLE is as follows: LOLE=p[X≥(CL)] Among them, p is the outage coefficient, X is the outage capacity of the system, L is the peak load, and C is the total installed capacity of the system; The level of power shortage for the whole year is as follows: In the formula, C i is the system installed capacity in the i-th time period; and p[X≥(C i -L ijk )] represents the probability that the outage capacity is greater than or equal to the backup capacity in the i-th time period; m represents the number of time periods in a year; n i Represents the number of days in the i-th time period; L ijk represents the peak load at hour K in the i-th time period of the j-th day; ②Battery level EENS The calculation formula of EENS is as follows: EENS=(XR)×P(X) In the formula, P(X) is the power shortage coefficient, X is the system's outage capacity, and R is the system's spare resource capacity, that is, the system's installed capacity; The power shortage level for a whole year is as follows: Where, L ijk represents the peak load at time K in the i-th time period of the j-th day; and P ijk (X) represents the probability that the outage capacity is greater than or equal to X at the Kth time in the i-th time period on the j-th day; m represents the number of time periods in a year; n i Indicates the number of days included in the i-th time period.

3. The method for calculating the confidence capacity of a strongly random wind farm based on the accelerated sequential Monte Carlo method according to claim 1, characterized in that: The wind farm power generation system reliability index calculation model in step 2 includes: a conventional unit output model, a wind farm output model and a load curve fluctuation model; (1) Conventional unit output model: only the normal operation and fault shutdown states are considered, and the normal operation duration t1 and fault repair time t2 are used to describe them. Usually t1 and t2 follow exponential distribution, and the formula for obtaining them is as follows: Where: λ is the failure rate; μ is the repair rate; t MTTF is the average working time; t MTTR is the average repair time, both units are h, so t1 and t2 units are also h; u1 and u2 are random numbers uniformly distributed between [0,1]; (2) Wind farm output model: The output power of wind turbines is directly related to wind speed, so the power simulation problem can be transformed into a wind speed simulation problem. The wind power output model is as follows: Where v is the wind speed, v ci is the cut-in wind speed, v co is the cut-out wind speed, v r is the rated wind speed, P sw Output power for wind farms; (3) Load curve fluctuation model: When constructing the load model, the change of load within the simulation period is not considered. The load model uses the annual peak load as the benchmark. The calculation formula is as follows: In the formula, is the annual peak load, P w (t) is the percentage of weekly peak load to annual peak load, P d (t) is the percentage of daily peak load to weekly peak load, P h (t) is the percentage of hourly peak load to daily peak load; A random variable that obeys normal distribution is superimposed on the original load model to correct the original load. At this time, the load curve fluctuation model can be expressed as:

4. The method for calculating the confidence capacity of a strongly random wind farm based on the accelerated sequential Monte Carlo method according to claim 1, characterized in that: The specific steps of step 3 include: (1) Use the accelerated sequential Monte Carlo method to simulate the output model of the wind farm and calculate the reliability index of the system by simulating the random process of the wind farm. With a sufficiently long simulation time, a reliability index close to the expected value is obtained. At this time, the reliability calculation of the system is as follows: Where I represents the reliability index of the system, which can be the lack of power time (LOLP) or the lack of power level (LOLE), etc.; T represents the simulation time, x t represents the state of the system component at time t, and f(x) is used to describe the state of the component at the time x. t The performance of the system under simulation; when the simulation time is long enough, the reliability index will approach the expected value I, thus obtaining the reliability index of the system; (2) Convert the continuous process into a cumulative summation method to continue formula optimization. The specific calculation formula is as follows: (3) Perform reliability assessment calculation on the wind farm power generation system reliability index calculation model constructed in step 2.

5. The method for calculating the confidence capacity of a strongly random wind farm based on the accelerated sequential Monte Carlo method according to claim 4 is characterized in that: The specific steps of step (3) of step 3 include: First, the reliability assessment process of the wind farm system starts with inputting relevant data and then setting the sampling year to 1; Next, the output curves of conventional units and wind turbines are sampled respectively, including the annual output curve of conventional units, the annual time-series wind speed intensity sequence, and the outage status of wind turbines; based on these data, the time-series output curve of wind turbines is calculated and integrated to form the annual power generation capacity curve; Subsequently, the system will compare the power balance on an hourly basis and calculate the system reliability index. If the reliability index does not reach the convergence standard, the process will check whether it exceeds the maximum age. If not, the sampling years will be increased and the above process will be repeated. Once the reliability index converges or reaches the maximum age, the system will output the final reliability index, thus completing a comprehensive assessment of the reliability of the system containing the wind farm.

6. The method for calculating the confidence capacity of a strongly random wind farm based on the accelerated sequential Monte Carlo method according to claim 1, characterized in that: The specific method of step 4 is: Evaluate the load that the wind farm can carry, compare the system reliability before and after the wind farm is connected, and calculate the confidence capacity of the wind farm: Assuming that the reliability of the original system is R0, when the system is connected to intermittent energy, the reliability level rises to R1. At this time, if the grid load is gradually increased, the system reliability gradually decreases; when the load level drops to a certain level, the reliability level returns to R0, as shown below: R0=f0(G,L)=f1(G+G int ,L) Inverse the distribution on both sides of the above equation to get: Therefore, the confidence capacity CC of the wind farm is ELCC for: In the formula, G int is the rated capacity of the wind farm, R is the reliability of the system, and L is the load.

Images