A reliability assessment method for a power system including an energy storage device and a storage medium

Abstract

The present invention provides a method for evaluating the reliability of a power system including an energy storage device, comprising: determining a power system state vector and random variables in the power system reliability evaluation; wherein the random vectors of the daily net loads of all buses included in the power system state vector and the peak values ​​of the daily total net load curve of the power system included in the random variables have a reversible mapping relationship; the power system reliability evaluation method provided by the present invention, by constructing a bidirectional mapping relationship between the random vectors of the daily net loads of all buses and the peak values ​​of the daily total net load curve of the power system, models and samples the original probability density function of the peak value of the daily total net load curve of the power system during the reliability evaluation process, without the need to construct a more complex original probability density function of the random vectors of the daily net loads of all buses. This solves the problem that the probability distribution function of the time series load curve is difficult to construct and the parameters are difficult to optimize in the importance sampling method based on cross entropy.

Description

Technical Field

[0001] The invention relates to a reliability assessment method for a power system including an energy storage device, and belongs to the field of reliability assessment of power systems. Background Art

[0002] According to the calculation principle, the reliability assessment methods of power systems are divided into two categories: analytical methods and Monte Carlo-based simulation methods, and according to the sampling method, non-sequential methods and sequential methods. Among them, almost all analytical methods are non-sequential methods, while Monte Carlo methods include both sequential and non-sequential methods. In traditional power systems, the above methods have shown satisfactory computing performance. However, in order to cope with the uncertainty and volatility of renewable energy generation, more and more energy storage devices are connected to new power systems. Since non-sequential methods cannot cope with the time coupling constraints of energy storage devices, when conducting reliability assessments on new power systems containing renewable energy and energy storage devices, sequential methods must be used, among which the sequential Monte Carlo method is the most famous. However, due to the high power supply margin and small reliability index of the new power system, the number of simulation years required for the sequential Monte Carlo method to achieve the specified index accuracy is too long, resulting in the method only having theoretical feasibility and losing practical operational value. Therefore, it is necessary to study relevant measures to reduce the simulation years of the sequential Monte Carlo method while ensuring the accuracy of the estimated index.

[0003] To this end, in recent years, some experts have tried to combine the importance sampling method with the time series Monte Carlo method, and have achieved a certain improvement in computing efficiency, but the research has mainly focused on how to optimize the failure rate of traditional units, while completely ignoring the acceleration potential of the load. This is because the time period included in each sample chain of the time series Monte Carlo method is not fixed, so it is difficult to find a suitable random variable to represent the time series load and optimize its importance distribution parameters. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies in the prior art and provide a method for evaluating the reliability of a power system containing an energy storage device. By constructing a bidirectional mapping relationship between the daily net load random vectors of all buses and the peak value of the daily total net load curve of the power system, the problem that the probability distribution function of the time series load curve is difficult to construct and the parameters are difficult to optimize in the importance sampling method based on cross entropy is solved.

[0005] To achieve the above object, the present invention is implemented by adopting the following technical solutions:

[0006] In a first aspect, the present invention provides a method for evaluating the reliability of a power system including an energy storage device, comprising:

[0007] Acquire relevant data required for power system reliability assessment, and determine the power system state vector and random variables in the power system reliability assessment based on the relevant data;

[0008] The power system state vector includes: a first continuous variable; the first continuous variable includes: a random vector of daily net loads of all buses;

[0009] The random variable includes: a second continuous variable; the second continuous variable includes: a peak value of a daily total net load curve of the power system;

[0010] The first continuous variable and the second continuous variable have a reversible mapping relationship;

[0011] Fitting the original probability distribution function of the random variable, and selecting a basic form of the importance probability distribution function of the random variable;

[0012] Entering the pre-simulation phase, obtaining the non-time-series parameters of the importance probability distribution function of the random variable through non-time-series sampling and iterative updating; and inputting the non-time-series parameters into the main simulation phase;

[0013] Enter the main simulation phase, including:

[0014] Establish a minimum load shedding model that includes energy storage equipment;

[0015] Combining the non-time-sequential parameters with the reversible mapping relationship to obtain a time-sequential sample of a power system state vector;

[0016] The time-series samples of the power system state vector are input into a minimum load shedding model, and the power system reliability index is calculated according to the output of the minimum load shedding model.

[0017] Furthermore, the relevant data required for the reliability assessment of the power system include: load curves of all buses within a preset historical period, time series data curves of wind power, time series data curves of photovoltaic power, time series data curves of net loads of all buses, average failure time, average repair time, upper and lower power limits and historical operating status of conventional generators and transmission lines, and the grid structure of the power grid;

[0018] The net load time series data curves of all the buses are calculated by the load curves of all the buses, the time series data curves of wind power and the time series data curves of photovoltaic power;

[0019] The first continuous variable is the random vector of daily net load of all buses, and the calculation formula is:

[0020] ;

[0021] In the formula, is the daily load vector of all buses, is the wind power vector, is the photovoltaic power generation power vector;

[0022] The daily total net load curve of the power system is obtained by superimposing the time series data curves of all bus net loads on that day.

[0023] Furthermore, the reversible mapping relationship between the first continuous variable and the second continuous variable includes:

[0024] Obtaining the corresponding daily net load curves of all buses according to the first continuous variable, and superimposing them to obtain the daily total net load curve of the power system, and then obtaining the peak value of the daily total net load curve of the power system, which is the corresponding second continuous variable;

[0025] According to the second continuous variable, the corresponding daily total net load curve of the power system is obtained, and each daily total net load curve of the power system has a corresponding date number; according to the date number, the daily net load curves of all buses on that day are extracted, and then the first continuous variable of T time periods is obtained, and T is a preset value.

[0026] Furthermore, the power system state vector and the random variables also include: discrete variables; the discrete variables include: the state random vector of the conventional generator set and the state random vector of the transmission line; in response to the conventional unit or the transmission line being in normal operation, the corresponding state value is 1, otherwise the corresponding state value is 0.

[0027] Furthermore, the fitting of the original probability distribution function of the random variable and the selection of a basic form of the importance probability distribution function of the random variable include:

[0028] Fitting raw probability distribution functions of discrete variables, including:

[0029] The state random vector of a conventional generator set for:

[0030] ;

[0031] In the formula, is the total number of conventional generating units; For the The operating status of conventional generating units;

[0032] The original probability distribution function of is constructed as:

[0033] ;

[0034] In the formula, Indicates The failure probability of a conventional generator set;

[0035] Random state vector of transmission line for:

[0036] ;

[0037] In the formula, is the total number of transmission lines; For the The operating status of the transmission lines;

[0038] The original probability distribution function of is constructed as:

[0039] ;

[0040] In the formula, Indicates The failure probability of a transmission line;

[0041] The second continuous variable is obtained by using kernel density estimation on the net load time series data curves of all buses. The original probability density function is:

[0042] ;

[0043] In the formula, is the number of days in the relevant data; is the kernel density estimation bandwidth; is the probability density function of the standard normal distribution; It is the first Peak value of the day;

[0044] The selecting of a basic form of the probability distribution function of the importance of the random variable includes:

[0045] and The importance probability distribution function of has the same form as the original probability distribution function:

[0046] ;

[0047] In the formula, yes The importance probability distribution function of yes The importance probability distribution function of It is The importance failure probability of each conventional generator set; It is The importance of the transmission line and the probability of failure;

[0048] The importance probability density function basic form chooses normal distribution:

[0049] ;

[0050] In the formula, and yes Parameters.

[0051] Furthermore, the non-time series parameters of the importance probability distribution function of the random variable are obtained by non-time series sampling and iterative updating, including:

[0052] S11: Set the number of samples per round to ; Initialize the importance distribution at the first iteration to the original distribution of each random vector;

[0053] S12: Sampling using the current importance distribution indivual , Samples and / T A sample of

[0054] Will / T The sample mapping is indivual Samples and , and The sample combination is Power system state vector Pre-simulation samples of

[0055] S13: Calculate the system state evaluation function for each pre-simulation sample , System power failure indication function Likelihood Ratio ,include:

[0056] The calculation formula is:

[0057] ;

[0058] In the formula, and The corresponding power system is in a fault state; and Indicates that the power system is in operation;

[0059] The calculation formula is:

[0060] ;

[0061] In the formula, For the The output power upper limit of a conventional generator set; Power system status Under this condition, the minimum load shedding model without energy storage equipment is used to obtain the Load shedding capacity of busbars; is the total number of buses;

[0062] The calculation formula is:

[0063] ;

[0064] In the formula, is the rounding symbol;

[0065] S14: All pre-simulation samples , and Substitute the update formula of importance distribution parameters and iteratively calculate the importance distribution parameters , , and ,include:

[0066] ;

[0067] ;

[0068] ;

[0069] ;

[0070] In response to all pre-simulation samples If the proportion exceeds 5%, the iteration is terminated and the current importance distribution parameters are output as non-time series parameters of the importance probability distribution function. Otherwise, a new importance distribution is constructed using the new importance distribution parameters and the process returns to S12.

[0071] Furthermore, the establishment of a minimum load shedding model including energy storage equipment includes:

[0072] ;

[0073] In the formula, Subscript the time period; It is the subscript of energy storage equipment;

[0074] It is On the bus The amount of load shedding in each period;

[0075] It is The conventional generator set is The power of each period;

[0076] It is The energy storage device is The power of each period;

[0077] It is Period The energy stored in the energy storage device;

[0078] It is The efficiency of each energy storage device;

[0079] and They are The lower and upper power limits of conventional generator sets;

[0080] It is The rated power of each energy storage device;

[0081] and They are The minimum storage energy and maximum storage energy of each energy storage device;

[0082] and They are The minimum and maximum transmission powers of the transmission lines; It is The transmission line in The transmission power of each time period;

[0083] , and They are The power transmission distribution coefficients of the transmission lines reflect , , and right impact.

[0084] Furthermore, the method of combining the non-time-sequential parameters and the reversible mapping relationship to obtain a time-sequential sample of the power system state vector includes:

[0085] S21: Convert the importance failure probability parameters of conventional generators and transmission lines in the pre-simulation phase into corresponding new mean failure time and new mean repair time, including:

[0086] ;

[0087] ;

[0088] ;

[0089] ;

[0090] In the formula, To obtain the The average failure time of conventional generator sets;

[0091] After conversion New mean time between failures of conventional generator sets;

[0092] To obtain the The average repair time of a conventional generator set;

[0093] After conversion New mean time to repair for conventional generator sets;

[0094] To obtain the Mean time between failures of transmission lines;

[0095] After conversion New mean time between failures for transmission lines;

[0096] To obtain the Average repair time for transmission lines;

[0097] After conversion New mean time to repair for transmission lines;

[0098] S22: Generate time sequence state samples of conventional generators and transmission lines using the new mean failure time and new mean repair time of the conventional generators and transmission lines after conversion, including:

[0099] ;

[0100] ;

[0101] ;

[0102] ;

[0103] In the formula, For the The sampling duration of the operating state of a conventional generator set within a state conversion cycle;

[0104] For the The sampling duration of the fault state within one state conversion cycle of a conventional generator set;

[0105] For the The sampling duration of the operating state of a transmission line in a state conversion cycle;

[0106] For the The sampling duration of the fault state in one state conversion cycle of the transmission line;

[0107] is a first uniformly distributed random variable with a value range of (0,1);

[0108] is a second uniformly distributed random variable with a value range of (0,1);

[0109] is a third uniformly distributed random variable with a value range of (0,1);

[0110] is the fourth uniformly distributed random variable, with a value range of (0,1);

[0111] initialization , and set the maximum simulation time to N;

[0112] In the interval , No. Conventional generator sets are in operation. , No. The conventional generator set is in a fault state, and the state random vector of the conventional generator set is Take a sample and then set , repeat the Sampling until When the maximum simulation time N is reached, N of them are arranged in chronological order. sample ;

[0113] Similarly, in the interval , No. Transmission lines are in operation, in the section , No. The transmission line is in a fault state, and the state vector of the transmission line is random. Take a sample and then set , repeat the Sampling until When the maximum simulation time N is reached, N of them are arranged in chronological order. sample ;

[0114] S23: In the interval ,right Repeat the sampling N / T times, and then convert the N / T continuous variables into The samples are converted into N time-ordered sample ;

[0115] In chronological order , and Samples constitute the time series samples of the power system state vector .

[0116] Furthermore, the step of inputting the time series samples of the power system state vector into a minimum load shedding model and calculating the power system reliability index according to the output of the minimum load shedding model includes:

[0117] Substituting each power system state vector sample into the minimum load shedding model to obtain the load shedding amount of each bus;

[0118] According to the load shedding of each bus, the corresponding state vector of each power system sample is calculated in turn. , and ;

[0119] Each power system state vector sample corresponds to , and Substitute into the calculation formula of power system reliability index to obtain the power system reliability index;

[0120] The power system reliability index calculation formula includes:

[0121] ;

[0122] .

[0123] In a second aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the power system reliability assessment method as described in the first aspect is implemented.

[0124] Compared with the prior art, the present invention has the following beneficial effects:

[0125] (1) The power system reliability assessment method provided by the present invention constructs a bidirectional mapping relationship between the daily net load random vectors of all buses and the peak value of the daily total net load curve of the power system. In the reliability assessment process, the original probability density function of the peak value of the daily total net load curve of the power system is modeled and sampled, without the need to construct a more complex original probability density function of the daily net load random vectors of all buses. This solves the problem that the probability distribution function of the time series load curve is difficult to construct and the parameters are difficult to optimize in the importance sampling method based on cross entropy.

[0126] (2) The power system reliability assessment method provided by the present invention can tap the acceleration potential of load in the calculation of reliability indicators and significantly improve the efficiency of reliability assessment of new power systems containing energy storage. BRIEF DESCRIPTION OF THE DRAWINGS

[0127] Figure 1 A flow chart of a reliability assessment method for a power system including an energy storage device provided in this application;

[0128] Figure 2 This is a schematic diagram of the test system structure designed based on the IEEE RTS-79 system;

[0129] Figure 3 To test the historical output curves of the three wind farms in the system;

[0130] Figure 4 This is the historical output curve of the photovoltaic power station in the test system. DETAILED DESCRIPTION

[0131] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices. The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments.

[0132] Example 1

[0133] According to a first aspect of the present invention, the present invention provides a method for evaluating the reliability of a power system including an energy storage device, such as Figure 1 As shown, the method comprises the following steps:

[0134] Step 1: Obtain relevant data required for power system reliability assessment, and determine the power system state vector and random variables in the power system reliability assessment based on the relevant data.

[0135] Specifically, the relevant data required for the reliability assessment of the power system include: the load curves of all buses within a preset historical period, the timing data curves of wind power, the timing data curves of photovoltaic power, the timing data curves of the net loads of all buses, the average failure time, the average repair time, the upper and lower power limits and the historical operating status of conventional generator sets and transmission lines, and the grid structure of the power grid; wherein the timing data curves of the net loads of all buses are calculated by the load curves of all buses, the timing data curves of wind power and the timing data curves of photovoltaic power.

[0136] In this embodiment, the power system state vector includes: a discrete variable and a first continuous variable, and the random variable includes: a discrete variable and a second continuous variable;

[0137] Specifically, the discrete variables include: the state random vector of the conventional generator set and the state random vector of the transmission line, the first continuous variable is the random vector of daily net load of all buses, and the calculation formula is:

[0138] ;

[0139] In the formula, is the daily load vector of all buses, is the wind power vector, is the photovoltaic power generation power vector;

[0140] The second continuous variable is: the peak value of the daily total net load curve of the power system, which is obtained by superimposing the time series data curves of all bus net loads on that day.

[0141] It should be noted that the first continuous variable and the second continuous variable are reversible mapping relationships.

[0142] Specifically, the daily net load curves of all corresponding buses are obtained according to the first continuous variable, and the daily total net load curve of the power system is obtained by superposition, and then the peak value of the daily total net load curve of the power system is obtained, which is the corresponding second continuous variable; the corresponding daily total net load curve of the power system is obtained according to the second continuous variable, and each daily total net load curve of the power system has a corresponding date number; the daily net load curves of all buses on that day are extracted according to the date number, and if the sampling time interval is 1 hour, the first continuous variables of 24 time periods are obtained.

[0143] Step 2: Fit the original probability distribution function of the random variable, and select the basic form of the importance probability distribution function of the random variable.

[0144] Preferably, the discrete variable adopts a 0-1 two-state model, and its original probability distribution function is constructed in combination with the input related data, and the basic form of its importance probability distribution function is the same as the original distribution; the original probability density function of the second continuous variable is constructed by using kernel density estimation on the net load time series data curves of all buses, and the basic form of its importance probability density function selects normal distribution.

[0145] Specifically, the state random vector of a conventional generator set is for:

[0146] ;

[0147] In the formula, is the total number of conventional generating units; For the The operating status of conventional generating units;

[0148] The original probability distribution function of is constructed as:

[0149] ;

[0150] In the formula, Indicates The failure probability of a conventional generator set;

[0151] Random state vector of transmission line for:

[0152] ;

[0153] In the formula, is the total number of transmission lines; For the The operating status of the transmission lines;

[0154] The original probability distribution function of is constructed as:

[0155] ;

[0156] In the formula, Indicates The failure probability of a transmission line;

[0157] and The importance probability distribution function of has the same form as the original probability distribution function:

[0158] ;

[0159] In the formula, yes The importance probability distribution function of yes The importance probability distribution function of It is The importance failure probability of each conventional generator set; It is The importance of the transmission line and the probability of failure;

[0160] Second continuous variable The original probability density function is:

[0161] ;

[0162] In the formula, is the number of days in the relevant data; is the kernel density estimation bandwidth; is the probability density function of the standard normal distribution; It is the first Peak value of the day;

[0163] Second continuous variable The basic form of the importance probability density function is:

[0164] ;

[0165] In the formula, and yes Parameters.

[0166] Step 3: Enter the pre-simulation phase, obtain the non-time-series parameters of the importance probability distribution function of the random variable through non-time-series sampling and iterative updating; and input the non-time-series parameters into the main simulation phase.

[0167] Step 31: Set the number of samples per round to ; Initialize the importance distribution at the first iteration to the original distribution of each random vector;

[0168] Step 32: Sampling using the current importance distribution indivual , Samples and / T A sample of

[0169] Will / T The sample mapping is indivual Samples and , and The sample combination is Power system state vector Pre-simulation samples of

[0170] Step 33: Calculate the system state evaluation function for each pre-simulation sample , System power failure indication function Likelihood Ratio ,include:

[0171] The calculation formula is:

[0172] ;

[0173] In the formula, and The corresponding power system is in a fault state; and Indicates that the power system is in operation;

[0174] The calculation formula is:

[0175] ;

[0176] In the formula, For the The output power upper limit of a conventional generator set; Power system status Under this condition, the minimum load shedding model without energy storage equipment is used to obtain the Load shedding capacity of busbars; is the total number of buses;

[0177] The calculation formula is:

[0178] ;

[0179] In the formula, is the rounding symbol;

[0180] S14: All pre-simulation samples , and Substitute the update formula of importance distribution parameters and iteratively calculate the importance distribution parameters , , and ,include:

[0181] ;

[0182] ;

[0183] ;

[0184] ;

[0185] In response to all pre-simulation samples If the proportion exceeds 5%, the iteration is terminated and the current importance distribution parameters are output as non-time series parameters of the importance probability distribution function. Otherwise, a new importance distribution is constructed using the new importance distribution parameters and the process returns to S12.

[0186] Step 4: Enter the main simulation phase, which includes:

[0187] Step 41: Establish a minimum load shedding model including energy storage equipment, including:

[0188] ;

[0189] In the formula, Subscript the time period; It is the subscript of energy storage equipment;

[0190] It is On the bus The amount of load shedding in each period;

[0191] It is The conventional generator set is The power of each period;

[0192] It is The energy storage device is The power of each period;

[0193] It is Period The energy stored in the energy storage device;

[0194] It is The efficiency of each energy storage device;

[0195] and They are The lower and upper power limits of conventional generator sets;

[0196] It is The rated power of each energy storage device;

[0197] and They are The minimum storage energy and maximum storage energy of each energy storage device;

[0198] and They are The minimum and maximum transmission powers of the transmission lines; It is The transmission line in The transmission power of each time period;

[0199] , and They are The power transmission distribution coefficients of the transmission lines reflect , , and right impact.

[0200] Step 42: Combining the non-time-sequential parameters and the reversible mapping relationship to obtain a time-sequential sample of the power system state vector, including:

[0201] Step 421: converting the importance failure probability parameters of conventional generator sets and transmission lines in the pre-simulation phase into corresponding new mean failure time and new mean repair time, including:

[0202] ;

[0203] ;

[0204] ;

[0205] ;

[0206] In the formula, To obtain the The average failure time of conventional generator sets;

[0207] After conversion New mean time between failures of conventional generator sets;

[0208] To obtain the The average repair time of a conventional generator set;

[0209] After conversion New mean time to repair for conventional generator sets;

[0210] To obtain the Mean time between failures of transmission lines;

[0211] After conversion New mean time between failures for transmission lines;

[0212] To obtain the Average repair time for transmission lines;

[0213] After conversion New mean time to repair for transmission lines;

[0214] Step 422: Generate a time sequence state sample of the conventional generator set and the transmission line using the new mean failure time and the new mean repair time of the conventional generator set and the transmission line after the conversion, including:

[0215] ;

[0216] ;

[0217] ;

[0218] ;

[0219] In the formula, For the The sampling duration of the operating state of a conventional generator set within a state conversion cycle;

[0220] For the The sampling duration of the fault state within one state conversion cycle of a conventional generator set;

[0221] For the The sampling duration of the operating state of a transmission line in a state conversion cycle;

[0222] For the The sampling duration of the fault state in one state conversion cycle of the transmission line;

[0223] is a first uniformly distributed random variable with a value range of (0,1);

[0224] is a second uniformly distributed random variable with a value range of (0,1);

[0225] is a third uniformly distributed random variable with a value range of (0,1);

[0226] is the fourth uniformly distributed random variable, with a value range of (0,1);

[0227] Step 423: Initialization , and set the maximum simulation time to N.

[0228] Step 424: In the interval , No. Conventional generator sets are in operation. , No. The conventional generator set is in a fault state, and the state random vector of the conventional generator set is Take a sample and then set , repeat the Sampling until When the maximum simulation time N is reached, N of them are arranged in chronological order. sample ;

[0229] Similarly, in the interval , No. Transmission lines are in operation, in the section , No. The transmission line is in a fault state, and the state vector of the transmission line is random. Take a sample and then set , repeat the Sampling until When the maximum simulation time N is reached, N of them are arranged in chronological order. sample ;

[0230] Step 425: In the interval ,right Repeat the sampling N / T times, and then convert the N / T continuous variables into The samples are converted into N time-ordered sample ;

[0231] In chronological order , and Samples constitute the time series samples of the power system state vector .

[0232] Step 43: inputting the time-series samples of the power system state vector into the minimum load shedding model, and calculating the power system reliability index according to the output of the minimum load shedding model, including:

[0233] Step 431: Substitute each power system state vector sample into the minimum load shedding model to obtain the load shedding amount of each bus;

[0234] Step 432: Calculate the corresponding state vector of each power system sample in turn according to the load shedding amount of each bus , and ;

[0235] Step 433: Corresponding to each power system state vector sample , and Substitute into the calculation formula of power system reliability index to obtain the power system reliability index;

[0236] The power system reliability index calculation formula includes:

[0237] ;

[0238] .

[0239] Example 2

[0240] This embodiment provides the performance of the method of the present invention in the reliability assessment of a power system including an energy storage device. Figure 2 The modified IEEE RTS-79 test system shown is carried out.

[0241] Compared with the original IEEE RTS-79 test system, the improved system adds three wind farms W1-W3 and one photovoltaic power station PV1, each with an installed capacity of 150MW. In addition, a battery energy storage system is added, with a rated capacity and power of 200MWh and 100MW respectively. The maximum and minimum values ​​of the state of charge are 0.9 and 0.1 respectively, and the charging and discharging efficiency coefficients are both set to 0.95.

[0242] Figure 3 The historical output curves of three wind farms in the test system are shown.

[0243] Figure 4 The historical output curve of the photovoltaic power station in the test system is shown.

[0244] In the improved IEEE RTS-79 system, W1-W3 and PV1 are connected to busbars 1, 2, 13 and 21 respectively. Battery energy storage is installed in the PV1 station and is therefore also connected to busbar 21. Other unit, line and load data are consistent with the original IEEE RTS-79 test system.

[0245] The following two methods were used as control groups to estimate the LOLP and EPNS index values ​​100 times respectively with the method proposed in this application (method 3). The coefficient of variance of LOLP and EPNS was calculated to demonstrate the computational efficiency and accuracy of the method proposed in this application.

[0246] Method 1: Traditional Sequential Monte Carlo Method;

[0247] Method 2: Timing importance sampling method without considering load acceleration potential;

[0248] The relevant calculation conditions are set as follows:

[0249] 1) In each iteration of the pre-simulation phase, 30,000 samples are extracted for both Method 2 and Method 3.

[0250] 2) The simulation period for all methods is set to 50 years.

[0251] The calculation results are shown in the following table.

[0252] ;

[0253] From the table we can see that:

[0254] 1) The average values ​​of LOLP and EPNS obtained by the three methods are almost the same, which verifies the correctness of the method proposed in this application.

[0255] 2) When the sample size of all methods is set to the same value (i.e., 50 years in this numerical test), the variance coefficient of the method proposed in this application is much smaller than that of the other two methods, which verifies that the method proposed in this application has higher accuracy.

[0256] 3) It can be further deduced from the variance coefficient that if Method 1 and Method 2 try to achieve the same variance coefficient value as Method 3, the sample size of the first two methods needs to be increased by at least 18.45 times and 4.54 times, respectively. In other words, the calculation speed of the method proposed in this application is 18.45 times and 4.54 times faster than that of Method 1 and Method 2, respectively, which proves the calculation efficiency improvement effect of the proposed method.

[0257] Example 3

[0258] According to a second aspect of the present invention, a computer storage medium is further provided, on which a computer program is stored, and when the computer program is executed by a processor, the power system reliability assessment method as described in Example 1 is implemented. The storage medium may include, for example, a storage component of a tablet computer, a hard disk of a computer, a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a portable compact disk read-only memory (CD-ROM), a USB memory, or any combination of the above storage media. The computer storage medium may be any combination of one or more computer storage media.

[0259] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for evaluating the reliability of a power system including an energy storage device, characterized in that: include: Acquire relevant data required for power system reliability assessment, and determine the power system state vector and random variables in the power system reliability assessment based on the relevant data; The power system state vector includes: a first continuous variable; the first continuous variable includes: a random vector of daily net loads of all buses; The random variable includes: a second continuous variable; the second continuous variable includes: a peak value of a daily total net load curve of the power system; The first continuous variable and the second continuous variable have a reversible mapping relationship; Fitting the original probability distribution function of the random variable, and selecting a basic form of the importance probability distribution function of the random variable; Entering the pre-simulation phase, obtaining the non-time-series parameters of the importance probability distribution function of the random variable through non-time-series sampling and iterative updating; and inputting the non-time-series parameters into the main simulation phase; Enter the main simulation phase, including: Establish a minimum load shedding model that includes energy storage equipment; Combining the non-time-sequential parameters with the reversible mapping relationship to obtain a time-sequential sample of a power system state vector; The time-series samples of the power system state vector are input into a minimum load shedding model, and the power system reliability index is calculated according to the output of the minimum load shedding model.

2. The power system reliability assessment method according to claim 1, characterized in that: The relevant data required for the reliability assessment of the power system include: load curves of all buses within a preset historical period, time series data curves of wind power, time series data curves of photovoltaic power, time series data curves of net loads of all buses, average failure time, average repair time, upper and lower power limits and historical operating status of conventional generators and transmission lines, and the grid structure of the power grid; The net load time series data curve of all buses is obtained by calculating the load curve of all buses, the time series data curve of wind power and the time series data curve of photovoltaic power; the first continuous variable L net is the random vector of the daily net load of all buses, and the calculation formula is: L net =L-P w -P PV Where L is the daily load vector of all buses, P w is the wind power vector, P PV is the photovoltaic power generation power vector; The daily total net load curve of the power system is obtained by superimposing the time series data curves of all bus net loads on that day.

3. The method according to claim 2, characterized in that The reversible mapping relationship between the first continuous variable and the second continuous variable includes: Obtaining the corresponding daily net load curves of all buses according to the first continuous variable, and superimposing them to obtain the daily total net load curve of the power system, and then obtaining the peak value of the daily total net load curve of the power system, which is the corresponding second continuous variable; According to the second continuous variable, the corresponding daily total net load curve of the power system is obtained, and each daily total net load curve of the power system has a corresponding date number; according to the date number, the daily net load curves of all buses on that day are extracted, and then the first continuous variable of T time periods is obtained, and T is a preset value.

4. The method according to claim 2, characterized in that: The power system state vector and the random variables also include: discrete variables; The discrete variables include: a state random vector of a conventional generator set and a state random vector of a transmission line; in response to the conventional generator set or the transmission line being in a normal operating state, the corresponding state value is 1, otherwise the corresponding state value is 0.

5. The method according to claim 4, characterized in that The fitting of the original probability distribution function of the random variable and selecting a basic form of the importance probability distribution function of the random variable include: Fitting raw probability distribution functions of discrete variables, including: The state random vector x of a conventional generator set c for: x c ={x c,i |i=1,...,n c } Where n c is the total number of conventional generator sets; x c,i is the operating status of the i-th conventional generator set; x c The original probability distribution function of is constructed as: In the formula, θ c,i represents the failure probability of the i-th conventional generator set; The state random vector x of the transmission line l for: x l {x l,j |j=1,…,n l } Where n l is the total number of transmission lines; x l,j is the operating status of the jth transmission line; x l The original probability distribution function of is constructed as: In the formula, θ l,j represents the failure probability of the jth transmission line; The second continuous variable L is obtained by using kernel density estimation on the net load time series data curves of all buses. peak The original probability density function is: Where D is the number of days in the relevant data; h is the kernel density estimation bandwidth; is the probability density function of the standard normal distribution; L peak,d is the peak value on the daily total net load curve of the power system on the dth day; The selecting of a basic form of the probability distribution function of the importance of the random variable includes: x c and x l The importance probability distribution function of has the same form as the original probability distribution function: In the formula, g(x c ) is x c The importance probability distribution function of g(x l ) is x l The importance probability distribution function of c,i is the importance failure probability of the i-th conventional generator set; η l,j is the importance failure probability of the jth transmission line; L peak The importance probability density function basic form chooses normal distribution: In the formula, μ net and σ net is g(L peak ) parameters.

6. The method according to claim 5, characterized in that The non-time series parameters of the importance probability distribution function of the random variable are obtained by non-time series sampling and iterative updating, including: S11: Set the number of samples in each round of sampling to n s ; Initialize the importance distribution at the first iteration to the original distribution of each random vector; S12: Sample n using the current importance distribution s x c 、x l The sample and n s / T L peak A sample of n s / T L peak The sample mapping is n s L net Sample, and x c 、x l and L net The sample combination is n s A pre-simulation sample of the power system state vector x; S13: Calculate the system state evaluation function S(x), system power shortage indication function I(x) and likelihood ratio Q(x) for each pre-simulation sample s ),include: The calculation formula of I(x) is: In the formula, S(x) < 0 and I(x) = 1 correspond to the power system being in a fault state; S(x) ≥ 0 and I(x) = 0 indicate that the power system is in an operating state; The calculation formula of S(x) is: In the formula, G i is the upper limit of the output power of the i-th conventional generator set; L sh,b is the load shedding amount of the bth bus obtained by the minimum load shedding model without energy storage equipment under the power system state x; n b is the total number of buses; Q(x s ) is calculated as: In the formula, is the rounding symbol; S14: S(x), I(x) and Q(x) of all pre-simulation samples s ) is substituted into the update formula of importance distribution parameter, and the importance distribution parameter η is iteratively calculated. c,i , η l,j , μ net and σ net ,include: In response to the proportion of I(x)=1 in all pre-simulation samples exceeding 5%, the iteration is terminated, and the current importance distribution parameters are output as non-time parameters of the importance probability distribution function. Otherwise, a new importance distribution is constructed using the new importance distribution parameters, and return to S12.

7. The method according to claim 6, characterized in that The step of establishing a minimum load shedding model including an energy storage device comprises: In the formula, t is the subscript of the time period; k is the subscript of the energy storage device; L sh,b,t is the load shedding amount on the bth bus in the tth period; P c,i,t is the power of the i-th conventional generator set in the t-th period; P e,k,t is the power of the kth energy storage device in the tth period; E e,k,t is the energy stored in the kth energy storage device in the tth period; η k is the efficiency of the kth energy storage device; P c,min,i and P c,max,i are the lower and upper limits of the power of the i-th conventional generator set respectively; P e,rated,k is the rated power of the kth energy storage device; E e,min,k and E e,max,k are the minimum storage energy and the maximum storage energy of the kth energy storage device, respectively; P l,min,j and P l,max,j are the minimum transmission power and maximum transmission power of the jth transmission line respectively; P l,j,t is the transmission power of the jth transmission line in the tth period; F i,j 、F k,j and F b,j are the power transmission distribution coefficients of the jth transmission line, respectively reflecting P c,i,t , P e,k,t , L net,b,t and L sh,b,t P l,j,t impact.

8. The method according to claim 7, characterized in that The method of combining the non-time series parameters and the reversible mapping relationship to obtain a time series sample of the power system state vector comprises: S21: converting the importance failure probability parameters of conventional generator sets and transmission lines in the pre-simulation stage into corresponding new mean failure time and new mean repair time, comprising: MTTF c,new,i =(MTTF c,i +MTTR c,i )(1-η c,i ) MTTR c,new,i =(MTTF c,i +MTTR c,i )n c,i MTTF l,new,j =(MTTF l,j +MTTR l,j )(1-η l,j ) MTTR l,new,j =(MTTF l,j +MTTR l,j )n l,j Where, MTTF c,i is the average failure time of the i-th conventional generator set; MTTF c,new,i is the new mean failure time of the i-th conventional generator set after conversion; MTTR c,i is the average repair time of the i-th conventional generator set; MTTR c,new,i is the new mean repair time of the i-th conventional generator set after conversion; MTTF l,j is the average failure time of the jth transmission line obtained; MTTF l,new,j is the new mean failure time of the jth transmission line after conversion; MTTR l,j is the average repair time of the jth transmission line obtained; MTTR l,new,j is the new average repair time of the jth transmission line after conversion; S22: Generate time sequence state samples of conventional generators and transmission lines using the new mean failure time and new mean repair time of the conventional generators and transmission lines after conversion, including: t o,i =-MTTF c,new,i ·ln u1 t f,i =-MTTR c,new,i ·ln u2 t o,j =-MTTF l,new,j ·ln u3 t f,j =-MTTR l,new,j ·ln u4 Where, t o,i is the sampling duration of the operating state of the i-th conventional generator set within a state conversion cycle; t f,i is the sampling duration of the fault state in one state conversion cycle of the i-th conventional generator set; t o,j is the sampling duration of the operating state of the j-th transmission line in one state conversion cycle; t f,j is the sampling duration of the fault state in one state conversion cycle of the jth transmission line; u1 is the first uniformly distributed random variable, with a value range of (0,1); u2 is a second uniformly distributed random variable with a value range of (0,1); u3 is a third uniformly distributed random variable with a value range of (0,1); u4 is the fourth uniformly distributed random variable, with a value range of (0,1); Initialize t=0 and set the maximum simulation time to N; In the interval [t,t+t o,i ), the i-th conventional generator set is in operation, in the interval [t+t o,i ,t+t o,i +t f,i ), the i-th conventional generator set is in a fault state, and the state random vector x of the conventional generator set is c Take a sample and then set t = t + t o,i +t f,i , repeat for x c Sampling is performed until t reaches the maximum simulation time N, and N x values ​​are obtained in chronological order. c Sample {x c,1 ,x c,2 …, x c,N }; Similarly, in the interval [t,t+t o,j ), the jth transmission line is in operation, in the interval [t+t o,j ,t+t o,j +t f,j ), the jth transmission line is in a fault state, and the state random vector x l Take a sample and then set t = t + t o,j +t f,j , repeat for x l Sampling is performed until t reaches the maximum simulation time N, and N x values ​​are obtained in chronological order. l Sample {x l,1 ,x l,2 …, x l,N }; S23: In the interval [0,N), for g(L peak ) Repeat the sampling N / T times, and then convert N / T L peak The sample is converted into N L in chronological order net Sample {L net,1 ,L net,2 …, L net,N }; x in chronological order c 、x l and L net The samples constitute the time series samples of the power system state vector {x1, x2…, x N }.

9. The method according to claim 8, characterized in that The step of inputting the time series samples of the power system state vector into a minimum load shedding model and calculating the power system reliability index according to the output of the minimum load shedding model comprises: Substituting each power system state vector sample into the minimum load shedding model to obtain the load shedding amount of each bus; According to the load shedding amount of each bus, S(x), I(x) and Q(x) corresponding to each power system state vector sample are calculated in sequence using the method described in claim 6; Substitute S(x), I(x) and Q(x) corresponding to each power system state vector sample into the power system reliability index calculation formula to obtain the power system reliability index; The power system reliability index calculation formula includes:

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the power system reliability assessment method according to any one of claims 1 to 9 is implemented.

Images