Power generation system toughness comprehensive evaluation method considering multi-fault evolution in extreme weather

Abstract

The invention discloses a power generation system toughness comprehensive evaluation method considering multi-fault evolution in extreme weather. The method comprises the following steps: constructing an extreme event fault set under a specific disaster; aiming at a single extreme event, constructing a frequency modulation process of the power system; a toughness comprehensive quantitative index is proposed from three dimensions of time, frequency and power; calculating a toughness evaluation index based on the frequency modulation process of the power system; analyzing the complex dynamic process of evolution of the extreme events in combination with the occurrence time periods of the adjacent extreme events; according to an extreme event fault set under a specific disaster, the toughness of the power generation system is evaluated in combination with a complex dynamic process of adjacent extreme event evolution. According to the method, the weak link of the power generation system under the disaster can be identified under the condition that the physical information of the extreme meteorological disaster is known, and a frequency-based toughness comprehensive quantitative evaluation index is provided aiming at the influence of continuous extreme events on the power generation system, so that the influence of the extreme meteorological disaster on the weak link of the power generation system is quantified.

Description

Technical Field

[0001] The present invention belongs to the technical field of resilient power systems, relates to the technology of power system resilience assessment, and particularly relates to a comprehensive assessment method for the resilience of a power generation system considering multi-fault evolution under extreme weather conditions. Background Art

[0002] Power system resilience mainly measures the ability of the system to reduce fault losses and quickly return to the normal state after the disturbance ends through the stages of preparation, absorption, and adaptation when suffering from extreme events and large-scale faults that cannot be predicted during the planning stage. However, the resilience of a power system is not determined by a single factor, but is jointly affected by the characteristics and interactions of its internal components.

[0003] Firstly, a power system consists of a power generation system, a high-voltage transmission system, and a distribution system, which have different characteristics that make them vulnerable to various extreme weather events or disasters. However, there are still deficiencies in the systematic assessment of power system resilience from the perspective of the power generation system. There is an urgent need to propose a systematic resilience index system and resilience assessment method to reflect the impact of extreme meteorological disasters on the weak links of the power generation system.

[0004] Secondly, the impact of extreme weather such as typhoons on the power system is highly complex. Typhoons have strong spatio-temporal persistence, with a wide range of influence and a long duration, often causing long-term and multi-frequency disturbance impacts on the power system. During the typhoon invasion, the power system may encounter multiple consecutive faults. Since the interval between adjacent faults is short, a new fault may occur before the transient process of the previous fault ends, making the fault evolution show continuous and complex dynamic characteristics. This process not only involves multiple links such as the power generation side, the transmission system, the distribution network, and the load users, but also involves the dynamic characteristics of various components and corresponding control strategies. Therefore, how to effectively characterize the fault dynamic evolution mechanism of the power generation system under extreme events such as typhoons and optimize the strategies to enhance system resilience is an important issue that needs to be solved urgently. Summary of the Invention

[0005] Object of the Invention: In order to overcome the deficiencies in the prior art, a comprehensive assessment method for the resilience of a power generation system considering multi-fault evolution under extreme weather conditions is provided. It can identify the weak links of the power generation system under disasters when the physical information of known extreme meteorological disasters is available, and propose a frequency-based comprehensive resilience quantification assessment index for the power generation system affected by continuous extreme events, so as to quantify the impact of extreme meteorological disasters on the weak links of the power generation system.

[0006] Technical Solution: To achieve the above object, the present invention provides a comprehensive assessment method for the resilience of a power generation system considering multi-fault evolution under extreme weather conditions, including the following steps:

[0007] S1: Analyze the impact process of extreme meteorological disasters on the power generation system, and construct a fault set of extreme events under specific disasters;

[0008] S2: For a single extreme event, construct the frequency modulation process of the power system, including the response and recovery processes after the frequency is disturbed;

[0009] S3: According to the transient and steady-state characteristics of the system, quantitatively define the resilience models and indicators of the system at different stages and states under impact, and propose comprehensive resilience quantification indicators from three dimensions of time, frequency, and power, including absolute indicators and relative indicators, to reflect the impact of extreme disasters on the power generation system;

[0010] S4: For the proposed comprehensive resilience quantification indicators, calculate the resilience evaluation indicators of the power generation system based on the frequency modulation process of the power system;

[0011] S5: Combine the time periods when adjacent extreme events occur, and analyze the complex dynamic process of the frequency response of the power system under the influence of extreme events;

[0012] S6: Based on steps S1 - S5, according to the fault set of extreme events under specific disasters, and combined with the complex dynamic process of the evolution of adjacent extreme events, propose a resilience evaluation process based on the system frequency to evaluate the resilience of the power generation system.

[0013] Furthermore, the method for constructing the fault set of extreme events under specific disasters in step S1 is as follows:

[0014] Input the historical typhoon data, and based on the typhoon disaster physical model and the component failure rate model of transmission lines under typhoon disasters, generate a corresponding number of fault sample sets by the Monte Carlo scenario generation method, and finally obtain the typical scenarios of each extreme event by using the scenario reduction technology based on probability distance.

[0015] The construction methods of the typhoon disaster physical model and the component failure rate model of transmission lines under typhoon disasters are as follows:

[0016] Typhoon disaster physical model:

[0017] Use the Batts typhoon model to simulate the wind speed of the typhoon wind field.

[0018] Component failure rate model:

[0019] The strength of the component follows the following normal distribution, as shown in formula (1); when the difference between the load borne by the component and its own strength is greater than 0, the component is in an unreliable operating state, and the component failure rate is calculated by formula (2).

[0020]

[0021] Where f(R) is the probability distribution of the component strength, R is the component strength, μ p and δ p are strength parameters related to the line and tower, is the failure probability of the k-th component on the transmission line at time t after the typhoon makes landfall, R k (t) and F k (t) are the component strength and wind load of the k-th component at time t after the typhoon makes landfall, respectively;

[0022] In the same overhead line, the conductors and towers are connected in series. Any line fault will cause the entire line to withdraw from operation. The series model is used to obtain the fault probability of a single line, as shown in formula (3);

[0023]

[0024] Where P i line (t) is the fault probability of the i-th transmission line at time t after the typhoon makes landfall. m1 and m2 are the numbers of towers and conductor spans in the line, respectively, and are the fault probabilities of the tower and conductor at time t after the typhoon makes landfall, respectively.

[0025] Furthermore, when constructing the frequency modulation process of the power system during a single extreme event in step S2, it includes:

[0026] When the frequency drops beyond the allowable range, all the generators participating in power generation in the system will successively carry out a primary frequency modulation process, a secondary frequency modulation process, and a tertiary frequency modulation process until the power in the system is rebalanced, the system frequency reaches a new steady-state value and this steady-state value is within the frequency safety range; the primary frequency modulation process is completed by the speed control system of the prime mover and can quickly respond to the change of the system frequency; the secondary frequency modulation process is completed by the spinning reserve and various flexible resources of the system and can achieve zero-frequency regulation; the tertiary frequency modulation process is completed by load shedding and can keep the frequency deviation caused by extreme events within the allowable range.

[0027] The power system frequency modulation process, including the power system frequency response and recovery process, is as follows:

[0028] The construction method of the power system frequency response and recovery model is:

[0029] Power system primary frequency modulation model:

[0030] At time t s,1 when the s-th extreme event occurs, the transfer function of the overall dynamic frequency response model of the power system is as shown in formula (4). The governor group of the system is equivalent to a virtual governor of the same form. After equivalence, formula (4) is transformed into formula (5);

[0031]

[0032] Wherein, Δf s,1* is the per-unit value of the frequency change amount in the primary frequency regulation process of the s-th extreme event, and ΔP S is the unbalanced power generated by the power system when the s-th extreme event occurs. G I (s) is the system inertia response transfer function, and H G (s) and H equ (s) are the transfer functions of the governor group of thermal power units before and after the equivalent of the whole system respectively. N G is the number of thermal power units in the system. is the rated output of thermal power unit i, and u i is the start-stop state of thermal power unit i, which is 1 when starting and 0 when stopping. k L is the load frequency regulation coefficient, and the reference value of the load is the rated value of the frequency at the t0 moment in the left neighborhood at the t moment after the typhoon makes landfall. H i and D i are the inertia time constant and damping coefficient of generator set i respectively. μ equ 、 are the equivalent droop, proportional, and integral coefficients of the system respectively; μ i 、 are the droop, proportional, and integral coefficients of unit i respectively;

[0033] Power system secondary frequency regulation model:

[0034] When the equivalent governor of the whole system returns to the steady state during the primary frequency regulation process, assume this moment is t s,2 , and the corresponding system frequency value at this time is f s,2 ; if the unbalanced power caused by the occurrence of the s-th extreme event is small and f s,2 has returned to the allowable error range, then only the resilience evaluation index during the primary frequency regulation process needs to be calculated, and the system directly enters the normal frequency stage from the frequency response stage; if the unbalanced power caused by the occurrence of the s-th extreme event is large and f s,2 has not returned to the allowable error range, then the secondary frequency regulation process needs to be carried out, and the system enters the frequency recovery stage from the frequency response stage; the frequency change is calculated using the above-mentioned power system equivalent dynamic response model, but at this time, the input of the response model is the positive reserve of thermal power units and the supporting power of all flexible loads, and the transfer function is as shown in formula (6);

[0035]

[0036] Wherein: Δf s,2*(s) is the per-unit value of the frequency change during the secondary frequency regulation process of the s-th extreme event; is the sum of the emergency power support of all flexible loads, and its form is considered as a non-delay step input opposite to the disturbance direction; is the sum of the positive reserve of the remaining operating thermal power units in the system, considered as a ramp input with delay and limited amplitude opposite to the disturbance direction; G I (s) and H equ (s) are the system inertia response transfer function and the equivalent governor transfer function of the thermal power units in the whole system respectively. The subscript i in the function is the unit number; N G is the number of thermal power units in the system; u i is the start-stop state of thermal power unit i, 1 when starting up and 0 when shutting down; γ i is the ramp rate of thermal power unit i; is the standby start-up delay of thermal power unit i; is the full standby time of thermal power unit i; and both start at the start time t of the secondary frequency regulation s,2 ;

[0037] Due to the existence of the ramp reserve of thermal power units, an analytical expression cannot be obtained when solving the maximum frequency deviation of the system. Therefore, the ramp reserve is equivalent to a step reserve. The standby response functions and parameter calculation methods of thermal power unit i before and after equivalence are shown in formulas (7)-(9);

[0038]

[0039] In the formula: y i (τ) and are the standby response functions of thermal power unit i before and after equivalence respectively; δ(i) is the unit step function; is the number of rising edges of the step reserve of thermal power unit i, and the number of steps is The number range is L i is the height of a single step of the step reserve of thermal power unit i; is the trigger time of the j-th rising edge of the step reserve of thermal power unit i, and its start time is the start time t of the secondary frequency regulation s,2 ;

[0040] Power system tertiary frequency regulation model:

[0041] When the frequency returns to the steady state during the secondary frequency regulation process, assume this moment is t s,3 , and the corresponding system frequency value at this time is f s,3 . If f s,3If it has been restored to within the allowable error range, only the resilience evaluation index during the primary and secondary frequency regulation processes needs to be calculated, and the system enters the normal frequency stage from the frequency restoration stage; if f s,3 has not been restored to within the allowable error range, then the tertiary frequency regulation process is required. The system enters the load shedding stage from the frequency restoration stage. The equivalent dynamic response model of the power system is still used to calculate the frequency change, but at this time, the input of the response model is the load shedding amount, and the transfer function is as shown in formula (10):

[0042]

[0043] where: Δf s,3* (s) is the per-unit value of the frequency change amount during the tertiary frequency regulation process of the s-th extreme event; is the load shedding amount.

[0044] Furthermore, in step S3, a comprehensive resilience quantification index is proposed from three dimensions of time, frequency, and power, specifically including:

[0045] The system frequency regulation process under extreme disaster events can be divided into four stages, namely the normal frequency stage, the frequency response stage, the frequency restoration stage, and the load shedding stage; the frequency response stage, the frequency restoration stage, and the load shedding stage correspond to the primary frequency regulation process, the secondary frequency regulation process, and the tertiary frequency regulation process respectively;

[0046] Taking the frequency as the performance index of the power generation system, the first stage is the normal frequency stage (t s,0 -t s,1 ). In this stage, the frequency of the power system is maintained near 50 Hz. Despite the power fluctuations of renewable energy and loads, the frequency is still maintained within the allowable fluctuation range; the second stage is the frequency response stage (t s,1 -t s,2 ). At time t s,1 , the s-th extreme event occurs, and the frequency starts to decline from 50 Hz. At this time, the system automatically performs the primary frequency regulation process, and the frequency drops to the lowest point f s nadir , corresponding to the time Due to the automatic adjustment of the governor and the load, the frequency gradually recovers to f s,2 ; the third stage is the frequency restoration stage (t s,2 -t s,3 ). At this time, since the frequency response stage is a droop regulation, it is necessary to increase the spinning reserve and various flexible loads to supplement the power difference. At time t s,2 , the secondary frequency regulation process is carried out, and the frequency gradually recovers from f s,2 to f s,3 . At this time, it is necessary to judge f s,3Whether it is restored to within the allowable error range. If not, it enters the fourth stage; the fourth stage is the load shedding stage (t s,3 -t s,4 ). At this time, if f s,3 is not restored to within the allowable error range, small-scale load shedding is required for restoration. The frequency is restored to f s,4 at time t s,4 , within the safe range; by analyzing the state evolution process of the power generation system before and after the occurrence of the s-th extreme event, a resilience evaluation index of the power generation system is defined based on the performance of the system frequency.

[0047] The resilience evaluation index of the power generation system includes:

[0048] Primary frequency regulation duration It reflects the response time of the system for primary frequency regulation during the s-th extreme event, as shown in formula (11)

[0049]

[0050] In the formula: t s,1 is the time when the s-th extreme event occurs, and t s,2 is the time when the primary frequency regulation process ends during the s-th extreme event;

[0051] Maximum frequency deviation Δf s max (Hz): Δf s max represents the maximum deviation of the system frequency during the s-th extreme event, and together with reflects the transient stability of the frequency, as shown in formula (12)

[0052] Δf s max =|50 - f s nadir | (12)

[0053] In the formula: f s nadir is the lowest frequency point during the primary frequency regulation process of the s-th extreme event;

[0054] Frequency recovery time It reflects the response time of the system for secondary frequency regulation during the s-th extreme event, as shown in formula (13)

[0055]

[0056] In the formula: t s,2 is the time when the primary frequency regulation process ends during the s-th extreme event, and t s,3is the moment when the secondary frequency regulation process ends in the s-th extreme event;

[0057] Final frequency recovery value f s final (Hz): f s final reflects the frequency regulation ability and disaster resistance ability of the system itself in the s-th extreme event, as shown in formula (14)

[0058] f s final = f n,3 (14)

[0059] In the formula: f n,3 is the frequency value recovered after the primary and secondary frequency regulations in the s-th extreme event;

[0060] Load shedding amount reflects the scope of the impact on the system load in the s-th extreme event.

[0061] Frequency drop / rise time T p is the expected time for the system frequency to drop from the standard frequency to the lowest point / rise to the highest point, reflecting the primary frequency regulation ability of the system. The expression is:

[0062]

[0063] In the formula: T p is the duration for the system frequency to drop to the lowest point / rise to the highest point, in seconds (s).

[0064] Frequency resistance time T r is the expected time for the system frequency to rise / fall from the lowest point / highest point to the load shedding frequency after the timely response of the system reserve capacity and resilience resources, reflecting the adequacy of the system. The expression is:

[0065]

[0066] In the formula: T r is the duration for the system frequency to rise / fall from the lowest point / highest point to the load shedding frequency, in seconds (s).

[0067] Frequency recovery time T s is the response time for the system frequency to rise / fall from the load shedding frequency to the standard frequency (within the safe range). The expression is:

[0068] T s = t s,4 - t s,3 (17)

[0069] Where: T s is the duration for the system frequency to rise / fall from the load-shedding frequency to the standard frequency (within the safe range), with the unit of second (s).

[0070] The frequency change duration T d is the total response time for the system frequency to change during the disaster invasion, and the expression is:

[0071] T d = T p + T r + T s (18)

[0072] Where: T p , T r , T s are respectively the frequency drop / rise time, frequency resistance time, and frequency recovery time of the system frequency, with the unit of second (s).

[0073] The frequency regulation power P R is the expected value of the frequency regulation power provided by the system reserve capacity and resilience resources within the frequency resistance time T r , and the expression is:

[0074] P R = P R,r (19)

[0075] Where: P R,r is the frequency regulation power provided by the system reserve capacity and resilience resources within the frequency resistance time T r , with the unit of megawatt (MW).

[0076] The adaptation rate R A reflects the short-time power ramp-up ability of the frequency regulation units and the abundance of the resilience resources possessed by the system. The higher the value, the stronger the anti-voltage sag characteristic of the system frequency. It is determined according to the output of each link in the resilience adaptation stage, and the expression is:

[0077]

[0078] Where: P R is the frequency regulation power, with the unit of megawatt (MW); T r is the frequency resistance time, with the unit of second (s).

[0079] The proportion of the frequency resistance time T r % can reflect the influence of different reserve capacities and resilience resources on the improvement effect of the system frequency robustness, and can be expressed as:

[0080]

[0081] Where: T p 、T r 、T s are respectively the frequency drop / rise time, frequency resistance time, and frequency recovery time of the system frequency at state i and interval j, with the unit of second (s).

[0082] The proportion of the load-shedding frequency difference Δf s % can reflect the influence of different load-shedding amounts on the system frequency recovery ability, and can be expressed as:

[0083]

[0084] Where: f * is set to the standard value of 50 Hz, and f cut 、f max / min are respectively the frequency change amount obtained by load shedding and the frequency change amount of the system frequency from the standard frequency to the lowest point / highest point, with the unit of hertz (Hz).

[0085] Furthermore, the calculation method of the power generation system resilience evaluation index in step S4 is as follows:

[0086] Solving the primary frequency regulation model of the power system:

[0087] Without considering the damping coefficient of the thermal power unit, that is, setting the generator damping coefficient D to 0; knowing that at time t s,1 the s-th extreme event occurs, formula (5) can be transformed into formula (23);

[0088]

[0089] Where: Δf s,1* (s) is the per-unit value of the frequency change amount in the primary frequency regulation process of the s-th extreme event; ΔP S is the unbalanced power caused by the occurrence of the s-th extreme event, and the fluctuation form is considered as step fluctuation; k G is the system equivalent inertia coefficient; λ, α, β are the transformation coefficients of the system dynamic frequency response model transfer function;

[0090] Let The time-domain expression of the per-unit value of the frequency change amount in the primary frequency regulation process of the s-th extreme event is obtained as formula (24)

[0091]

[0092] According to formula (24), calculate the frequency deviation f s,2* -f s,1* at the end and start of the primary frequency regulation, as shown in formula (25), the frequency deviation f s nadir*-f s,1* As shown in formula (26), the time difference between the lowest frequency point and the initial moment As shown in formula (27), the time difference between the end moment of primary frequency regulation and the lowest frequency point As shown in formula (28):

[0093]

[0094]

[0095] According to formulas (25)-(28), the calculation methods of two frequency indexes in primary frequency regulation are given as follows: The duration of primary frequency regulation The calculation is as shown in formula (29):

[0096]

[0097] The maximum frequency difference Δf s max The calculation is as shown in formula (30):

[0098] Δf s max =|f s nadir* -f s,1* | (30)

[0099] Solution of the secondary frequency regulation model of the power system:

[0100] It is known that at the end moment t s,2 of primary frequency regulation, the frequency value of the system is f s,2 ; if f s,2 has not yet recovered to the allowable error range, then the secondary frequency regulation process needs to be carried out, and the system enters the frequency recovery stage from the frequency response stage. At this time, the input of the response model is the positive reserve of thermal power units and the supporting power of all flexible loads, and formula (6) is transformed into formula (31)

[0101]

[0102] In formula (31), g(s) has the same form as Δf s,1* (s) in formula (23), and the form of the derived frequency difference is also the same. Therefore, the influence of h i (s) is mainly analyzed; through derivation, the frequency deviation f s,3* -f s,2* between the end moment and the start moment of secondary frequency regulation can be obtained. As shown in formula (34), the time difference t s,3 -t s,2 between the end moment of secondary frequency regulation and the end moment of primary frequency regulation is as shown in formula (33), and the parameters therein are calculated by formula (34):

[0103]

[0104]

[0105] where p i is the th step of thermal power unit i;

[0106] According to formulas (32)-(35), the calculation methods of two frequency indexes in primary frequency regulation are given as follows:

[0107] Frequency recovery time is calculated as shown in formula (35):

[0108]

[0109] Final value of frequency recovery f s final is calculated as shown in formula (36):

[0110] f s final =(f s,3* -f s,2* )+(f s,2* -f s,1* )+f s,1* (36)

[0111] Solution of the power system's tertiary frequency regulation model:

[0112] It is known that at the end time t s,3 of the secondary frequency regulation, the frequency value of the system is f s,3 ; if f s,3 has not yet recovered within the allowable error range, then the tertiary frequency regulation process is required. The system enters the load shedding stage from the frequency recovery stage. The same equivalent dynamic response model of the power system as that in primary frequency regulation is used to calculate the frequency change, but at this time, the input of the response model is the load shedding amount; by reducing a certain proportion of the load, the system frequency f s,4 is restored within the allowable error range, and the total load reduced at this time is the load shedding amount

[0113] Furthermore, in step S5, in combination with the time periods when adjacent extreme events occur, analyze the complex dynamic process of the frequency response of the power system under the influence of extreme events:

[0114] During the invasion of extreme events, the power system is often affected by multiple faults. This causes new faults to quickly intervene during the frequency regulation process of the power system before the transient process of a single fault is completely over, resulting in a complex dynamic process. By combining the time periods when adjacent faults occur, the frequency fluctuations, power attenuation, and oscillation characteristics experienced by the system in a short period can be analyzed in depth. Furthermore, the complex dynamic coupling relationships between various links can be revealed, which helps to understand the response behavior and recovery characteristics of the system during the fault evolution process, provides key data support for constructing a comprehensive system resilience assessment model, and lays a theoretical foundation for formulating more accurate protection and recovery strategies.

[0115] Furthermore, the resilience assessment process based on the system frequency proposed in step S6 specifically includes the following steps:

[0116] 1) Input the historical data or prediction information of a certain typhoon, the power grid structure, and the geographical location relationship between the typhoon and the power grid;

[0117] 2) Calculate the failure rates of transmission lines connected to power plants / substations at different times and construct an extreme event fault set;

[0118] 3) Determine the evaluation period and analyze the interval time of extreme events;

[0119] 4) Taking 50Hz as the initial frequency, analyze the frequency regulation process of the power system in the first extreme event, and model the response and recovery of the power system frequency in the first extreme event;

[0120] 5) Determine whether the next extreme event occurs during the frequency regulation process of the previous extreme event;

[0121] 5.1) If it occurs, calculate the initial frequency of the power system when the next extreme event occurs, and analyze the frequency regulation process of the power system in the current extreme event with this frequency as the initial frequency to complete the modeling of the response and recovery of the power system frequency;

[0122] 5.2) If it does not occur, calculate the resilience index of the power generation system in the previous extreme event;

[0123] 6) Determine whether the extreme event is over; if so, end the evaluation process; if not, continue to evaluate the next extreme event and return to execute step 5.1).

[0124] The core of the proposed solution lies in constructing a comprehensive resilience quantification index based on frequency characteristics. In the case of extremely short intervals between consecutive faults during extreme events, where the frequency modulation transient process of the previous fault has not ended when the next fault occurs, the proposed index fully considers the transient characteristics (such as frequency fluctuations) exhibited by the system when subjected to shocks, as well as the steady-state characteristics (such as system stability and power balance) demonstrated during the disaster duration or recovery phase. Thus, absolute and relative indices are proposed from three dimensions: time, frequency, and power. The absolute index, as a direct quantification benchmark, can be used to evaluate the resilience level of the system under disturbances; while the relative index focuses on comparing the impacts of different factors on the frequency resilience of the system, revealing the performance deviations during the evolution process of the system from transient to steady state. The two indices complement each other, not only comprehensively reflecting the impacts of extreme meteorological disasters on the power generation system and the fault evolution state, but also providing solid data support and theoretical basis for formulating more targeted optimization and protection strategies.

[0125] Advantages: Compared with the prior art, the resilience assessment index defined by the present invention can accurately and comprehensively reflect the resilience level of the power generation system under extreme events. It can identify the weak links of the power generation system under disasters when the physical information of known extreme meteorological disasters is available, propose a systematic resilience assessment index based on frequency for the power generation system, thereby quantifying the impacts of extreme meteorological disasters on the weak links of the power generation system, and providing reliable data for the system energy management system and safety control system. Brief Description of the Drawings

[0126] Figure 1 Schematic diagram of the resilience state of the power generation system before and after a single extreme event in the present invention;

[0127] Figure 2 Flowchart of the resilience assessment of the power generation system based on frequency adopted by the present invention. Detailed Embodiments

[0128] The present invention will be further clarified below in conjunction with the drawings and specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and not to limit the scope of the present invention. After reading the present invention, various equivalent modifications made by those skilled in the art fall within the scope defined by the appended claims of this application.

[0129] The present invention provides a comprehensive resilience assessment method for a power generation system considering multi-fault evolution under extreme weather, referring to Figure 2 , and includes the following steps:

[0130] S1: Analyze the impact process of extreme meteorological disasters on the power generation system and construct an extreme event fault set under specific disasters;

[0131] S2: For a single extreme event, construct the frequency regulation process of the power system, including the response and recovery processes after the frequency is disturbed;

[0132] S3: According to the transient and steady-state characteristics of the system, quantitatively define the resilience models and indicators of the system at different stages and states under shocks, and propose comprehensive resilience quantification indicators from three dimensions of time, frequency, and power, including absolute indicators and relative indicators, to reflect the impact of extreme disasters on the power generation system;

[0133] S4: For the proposed comprehensive resilience quantification indicators, calculate the resilience evaluation indicators of the power generation system based on the frequency regulation process of the power system;

[0134] S5: Combine the time periods when adjacent extreme events occur, and analyze the complex dynamic process of the frequency response of the power system under the influence of extreme events;

[0135] S6: Based on steps S1 - S5, according to the extreme event fault set under specific disasters, and combining the complex dynamic process of the evolution of adjacent extreme events, propose a resilience evaluation process based on the system frequency to evaluate the resilience of the power generation system.

[0136] The method for selecting and constructing the extreme event fault set under specific disasters in step S1 is as follows:

[0137] Input the typhoon historical data. Based on the typhoon disaster physical model and the component failure rate model of transmission lines under typhoon disasters, use the Monte Carlo scenario generation method to generate a corresponding number of fault sample sets, and finally use the scenario reduction technology based on probability distance to obtain the typical scenarios of each extreme event.

[0138] The methods for constructing the typhoon disaster physical model and the component failure rate model of transmission lines under typhoon disasters include:

[0139] Typhoon disaster physical model:

[0140] Use the Batts typhoon model to simulate the typhoon wind field wind speed.

[0141] Component failure rate model:

[0142] The strength of the component follows the following normal distribution, as shown in formula (1); when the difference between the load borne by the component and its own strength is greater than 0, the component is in an unreliable operating state, and the component failure rate is calculated by formula (2).

[0143]

[0144] In the formula, f(R) is the probability distribution of the component strength, R is the component strength, μ p and δ p are strength parameters related to the line and tower, is the failure probability of the k-th component on the transmission line at time t after the typhoon makes landfall, R k (t) and F k (t) are the component strength and wind load of the k-th component at time t after the typhoon makes landfall, respectively;

[0145] In the same overhead line, conductors and poles are connected in series. Any line fault will cause the entire line to withdraw from operation. The series model is used to obtain the fault probability of a single line, as shown in formula (3);

[0146]

[0147] In the formula, P i line (t) is the fault probability of the i-th transmission line at time t after the typhoon makes landfall. m1 and m2 are the numbers of poles and conductor spans in the line, respectively. and are the fault probabilities of poles and conductors at time t after the typhoon makes landfall, respectively.

[0148] In step S2, when constructing a single extreme event, the frequency regulation process of the power system:

[0149] When the frequency drops beyond the allowable range, all generators participating in power generation in the system will successively undergo a primary frequency regulation process, a secondary frequency regulation process, and a tertiary frequency regulation process until the power in the system is rebalanced, the system frequency reaches a new steady-state value and this steady-state value is within the frequency safety range; the primary frequency regulation process is completed by the speed control system of the prime mover and can quickly respond to the change in the system frequency; the secondary frequency regulation process is completed by the spinning reserve and various flexible resources of the system and can achieve a zero-frequency regulation; the tertiary frequency regulation process is completed by load shedding and can keep the frequency deviation caused by extreme events within the allowable range.

[0150] The frequency regulation process of the power system, including the frequency response and recovery process of the power system, is as follows:

[0151] The construction method of the power system frequency response and recovery model is:

[0152] Power system primary frequency regulation model:

[0153] At time t s,1 when the s-th extreme event occurs, the transfer function of the overall dynamic frequency response model of the power system is as shown in formula (4). The governor group of the system is equivalent to a virtual governor of the same form. After equivalence, formula (4) is transformed into formula (5);

[0154]

[0155] In the formula, Δf s,1*is the per-unit value of the frequency change during the primary frequency regulation process of the s-th extreme event, ΔP S is the unbalanced power generated by the power system during the occurrence of the s-th extreme event, G I (s) is the system inertia response transfer function, H G (s) and H equ (s) are the transfer functions before and after the equivalent of the governor group of thermal power units in the whole system respectively, N G is the number of thermal power units in the system, is the rated output of thermal power unit i, u i is the start-stop state of thermal power unit i, 1 when starting up and 0 when shutting down, k L is the load frequency regulation coefficient, the reference value of the load is the rated value of the frequency at the t0 moment in the left neighborhood at the t moment after the typhoon makes landfall, H i and D i are the inertia time constant and damping coefficient of generator set i respectively, μ equ 、 are the equivalent droop, proportional, and integral coefficients of the system respectively; μ i 、 are the droop, proportional, and integral coefficients of unit i respectively;

[0156] Power system secondary frequency regulation model:

[0157] When the equivalent governor of the whole system returns to the steady state during the primary frequency regulation process, assuming this moment is t s,2 , the corresponding system frequency value at this time is f s,2 ; if the unbalanced power caused by the occurrence of the s-th extreme event is small and f s,2 has returned to the allowable error range, then only the resilience evaluation index during the primary frequency regulation process needs to be calculated, and the system directly enters the normal frequency stage from the frequency response stage; if the unbalanced power caused by the occurrence of the s-th extreme event is large and f s,2 has not returned to the allowable error range, then the secondary frequency regulation process needs to be carried out, and the system enters the frequency recovery stage from the frequency response stage; the above power system equivalent dynamic response model is used to calculate the frequency change, but at this time the input of the response model is the positive reserve of thermal power units and the support power of all flexible loads, and the transfer function is as shown in formula (6);

[0158]

[0159] In the formula: Δf s,2* (s) is the per-unit value of the frequency change during the secondary frequency regulation process of the s-th extreme event; is the sum of the emergency power support amounts of all flexible loads, and its form is considered as a non-delay step input opposite to the disturbance direction; It is the sum of the positive reserves of the remaining operating thermal power units in the system, considered as a ramp input with a time delay and a limited amplitude in the direction opposite to the disturbance; G I (s) and H equ (s) are the system inertia response transfer function and the equivalent governor transfer function of the entire system of thermal power units respectively. The subscript i in the function is the unit number; N G is the number of thermal power units in the system; u i is the start-stop state of thermal power unit i, which is 1 when the unit is on and 0 when the unit is off; γ i is the ramp rate of thermal power unit i; is the standby start-up time delay of thermal power unit i; is the full standby time of thermal power unit i; and Both start at the start time t of secondary frequency regulation s,2 ;

[0160] Due to the existence of ramp reserve of thermal power units, an analytical expression cannot be obtained when solving the maximum frequency deviation of the system. Therefore, the ramp reserve is equivalent to step reserve. The standby response function and parameter calculation method of thermal power unit i before and after equivalence are shown in formulas (7)-(9);

[0161]

[0162] In the formula: y i (τ) and are the standby response functions of thermal power unit i before and after equivalence respectively; δ(i) is the unit step function; is the number of rising edges of the step reserve of thermal power unit i, and the number of steps is The number range is L i is the height of a single step of the step reserve of thermal power unit i; is the triggering time of the j-th rising edge of the step reserve of thermal power unit i, and its start time is the start time t of secondary frequency regulation s,2 ;

[0163] Power system tertiary frequency regulation model:

[0164] When the frequency returns to the steady state during secondary frequency regulation, assume this moment is t s,3 , and the corresponding system frequency value at this time is f s,3 . If f s,3 has returned to the allowable error range, then only the resilience evaluation index during primary and secondary frequency regulation needs to be calculated, and the system enters the normal frequency stage from the frequency recovery stage; if f s,3If it has not yet recovered within the allowable error range, three frequency regulation processes are required. The system enters the load shedding stage from the frequency recovery stage. The above-mentioned equivalent dynamic response model of the power system is still used to calculate the frequency change, but at this time, the input of the response model is the load shedding amount, and the transfer function is shown in formula (10):

[0165]

[0166] In the formula: Δf s,3* (s) is the per-unit value of the frequency change amount during the three-frequency regulation process of the s-th extreme event; is the load shedding amount.

[0167] In step S3, comprehensive resilience quantification indicators are proposed from three dimensions of time, frequency, and power:

[0168] The system frequency regulation process under extreme disaster events can be divided into four stages, namely the normal frequency stage, the frequency response stage, the frequency recovery stage, and the load shedding stage; the frequency response stage, the frequency recovery stage, and the load shedding stage correspond to the primary frequency regulation process, the secondary frequency regulation process, and the tertiary frequency regulation process respectively;

[0169] As Figure 1 shown, taking frequency as the performance index of the power generation system, the first stage is the normal frequency stage (t s,0 -t s,1 ). In this stage, the frequency of the power system is maintained near 50 Hz. Despite the power fluctuations of renewable energy and loads, the frequency is still maintained within the allowable fluctuation range; the second stage is the frequency response stage (t s,1 -t s,2 ). At time t s,1 , the s-th extreme event occurs, and the frequency starts to decline from 50 Hz. At this time, the system automatically performs the primary frequency regulation process, and the frequency drops to the lowest point f s nadir , corresponding to the time Due to the automatic regulation of the governor and the load, the frequency gradually recovers to f s,2 ; the third stage is the frequency recovery stage (t s,2 -t s,3 ). At this time, because the frequency response stage is a droop regulation, it is necessary to increase the spinning reserve and various flexible loads to supplement the power difference. At time t s,2 , the secondary frequency regulation process is carried out, and the frequency gradually recovers from f s,2 to f s,3 . At this time, it is necessary to judge whether f s,3 has recovered within the allowable error range. If not, it enters the fourth stage; the fourth stage is the load shedding stage (t s,3 -t s,4 ). At this time, fs,3 If it cannot be restored within the allowable error range, small-scale load shedding is required for restoration. The frequency is restored to f at time t s,4 and is within the safe range; by analyzing the state evolution process of the power generation system before and after the occurrence of the s-th extreme event, a resilience evaluation index of the power generation system is defined based on the performance of the system frequency. s,4

[0170] Primary frequency regulation duration It reflects the response time of the system for primary frequency regulation during the s-th extreme event, as shown in Equation (11)

[0171]

[0172] In the formula: t s,1 is the time when the s-th extreme event occurs, and t s,2 is the time when the primary frequency regulation process ends during the s-th extreme event;

[0173] Maximum frequency deviation Δf s max (Hz): Δf s max represents the maximum deviation of the system frequency during the s-th extreme event, and together with reflects the transient stability of the frequency, as shown in Equation (12)

[0174] Δf s max = |50 - f s nadir | (12)

[0175] In the formula: f s nadir is the lowest frequency point during the primary frequency regulation process of the s-th extreme event;

[0176] Frequency recovery time It reflects the response time of the system for secondary frequency regulation during the s-th extreme event, as shown in Equation (13)

[0177]

[0178] In the formula: t s,2 is the time when the primary frequency regulation process ends for the s-th extreme event, and t s,3 is the time when the secondary frequency regulation process ends during the s-th extreme event;

[0179] Final frequency recovery value f s final (Hz): f s finalIt reflects the frequency regulation ability and disaster resistance ability of the system itself in the s-th extreme event, as shown in formula (14).

[0180] f s final = f n,3 (14)

[0181] In the formula: f n,3 is the frequency value restored after the primary and secondary frequency regulation of the system in the s-th extreme event;

[0182] Load shedding amount It reflects the range of influence on the system load in the s-th extreme event.

[0183] Frequency drop / rise time T p is the expected value of the time when the system frequency drops from the standard frequency to the lowest point / rises to the highest point, reflecting the primary frequency regulation ability of the system. The expression is:

[0184]

[0185] In the formula: T p is the duration when the system frequency drops to the lowest point / rises to the highest point, with the unit of second (s).

[0186] Frequency resistance time T r is the expected value of the time when the system frequency rises / falls from the lowest point / highest point to the load shedding frequency after the timely response of the system reserve capacity and resilience resources, reflecting the adequacy of the system. The expression is:

[0187]

[0188] In the formula: T r is the duration when the system frequency rises / falls from the lowest point / highest point to the load shedding frequency, with the unit of second (s).

[0189] Frequency recovery time T s is the response time when the system frequency rises / falls from the load shedding frequency to the standard frequency (within the safe range). The expression is:

[0190] T s = t s,4 - t s,3 (17)

[0191] In the formula: T s is the duration when the system frequency rises / falls from the load shedding frequency to the standard frequency (within the safe range), with the unit of second (s).

[0192] Frequency change duration T dThe total response time for the system frequency to change during a disaster attack is expressed as:

[0193] T d = T p + T r + T s (18)

[0194] Where: T p , T r , T s are the frequency drop / rise time, frequency resistance time, and frequency recovery time of the system frequency, respectively, with the unit of second (s).

[0195] The frequency regulation power P R is the expected value of the frequency regulation power provided by the system reserve capacity and resilience resources during the frequency resistance time T r and is expressed as:

[0196] P R = P R,r (19)

[0197] Where: P R,r is the frequency regulation power provided by the system reserve capacity and resilience resources during the frequency resistance time T r , with the unit of megawatt (MW).

[0198] The adaptation rate R A reflects the short-time power ramp-up ability of the frequency regulation unit and the abundance of the resilience resources possessed by the system. The higher the value, the stronger the anti-sag characteristic of the system frequency. It is determined according to the output of each link in the resilience adaptation stage and is expressed as:

[0199]

[0200] Where: P R is the frequency regulation power, with the unit of megawatt (MW); T r is the frequency resistance time, with the unit of second (s).

[0201] The proportion of the frequency resistance time T r % can reflect the influence of different reserve capacities and resilience resources on the improvement effect of the system frequency robustness and can be expressed as:

[0202]

[0203] Where: T p , T r , T s are the frequency drop / rise time, frequency resistance time, and frequency recovery time of the system frequency at state i and interval j, respectively, with the unit of second (s).

[0204] Percentage of load shedding frequency difference Δf s % can reflect the influence of different load shedding amounts on the system frequency recovery ability, and can be expressed as:

[0205]

[0206] In the formula: f * is set to the standard value of 50 Hz, f cut , f max / min are respectively the frequency change amount obtained by load shedding and the frequency change amount when the system frequency drops from the standard frequency to the lowest point / rises to the highest point, with the unit of Hertz (Hz).

[0207] The calculation method of the power generation system resilience index in step S4 is as follows:

[0208] Solving the primary frequency regulation model of the power system:

[0209] Without considering the damping coefficient of thermal power units, that is, setting the generator damping coefficient D to 0; knowing that at time t s,1 when the s-th extreme event occurs, formula (5) can be transformed into formula (23);

[0210]

[0211] In the formula: Δf s,1* (s) is the per-unit value of the frequency change amount in the primary frequency regulation process of the s-th extreme event; ΔP S is the unbalanced power caused by the occurrence of the s-th extreme event, and the fluctuation form is considered as step fluctuation; k G is the system equivalent inertia coefficient; λ, α, β are the transformation coefficients of the transfer function of the system dynamic frequency response model;

[0212] Let The time-domain expression of the per-unit value of the frequency change amount in the primary frequency regulation process of the s-th extreme event is obtained as formula (24)

[0213]

[0214] According to formula (24), calculate the frequency deviation f s,2* -f s,1* at the end of the primary frequency regulation and the start time, as shown in formula (25), the frequency deviation f s nadir* -f s,1* between the lowest frequency point and the initial frequency, as shown in formula (26), the time difference between the lowest frequency point and the initial time, as shown in formula (27), the time difference between the end time of the primary frequency regulation and the lowest frequency point, as shown in formula (28):

[0215]

[0216]

[0217] According to formulas (25)-(28), the calculation methods of two frequency indexes in primary frequency regulation are given as follows:

[0218] Primary frequency regulation duration The calculation is as shown in formula (29):

[0219]

[0220] Maximum frequency deviation Δf s max The calculation is as shown in formula (30):

[0221] Δf s max =|f s nadir* -f s,1* | (30)

[0222] Solution of the secondary frequency regulation model of the power system:

[0223] Given the end time t of primary frequency regulation s,2 The frequency value of the system is f s,2 ; if f s,2 has not yet recovered to the allowable error range, then the secondary frequency regulation process needs to be carried out, and the system enters the frequency recovery stage from the frequency response stage. At this time, the input of the response model is the positive reserve of thermal power units and the supporting power of all flexible loads, and formula (6) is transformed into formula (31)

[0224]

[0225] In formula (31), g(s) has the same form as Δf s,1* (s) in formula (23), and the derived form of the frequency difference is also the same. Therefore, the influence of h i (s) is mainly analyzed; through derivation, the frequency deviation f s,3* -f s,2* at the end time and the start time of secondary frequency regulation can be obtained. As shown in formula (34), the time difference t s,3 -t s,2 between the end time of secondary frequency regulation and the end time of primary frequency regulation is as shown in formula (33), and the parameters therein are calculated by formula (34):

[0226]

[0227]

[0228] where p i is the th step of thermal power unit i;

[0229] According to formulas (32)-(35), the calculation methods of two frequency indexes in primary frequency regulation are given as follows:

[0230] Frequency recovery time is calculated as shown in formula (35):

[0231]

[0232] Final value of frequency recovery f s final is calculated as shown in formula (36):

[0233] f s final =(f s,3* -Δf s,2* )+(f s,2* -f s,1* )+f s,1* (36)

[0234] Solution of the tertiary frequency regulation model of the power system:

[0235] Given that at the end time t s,3 of secondary frequency regulation, the frequency value of the system is f s,3 ; if f s,3 has not yet recovered within the allowable error range, then a tertiary frequency regulation process is required. The system enters the load shedding stage from the frequency recovery stage. The same equivalent dynamic response model of the power system as in primary frequency regulation is used to calculate the frequency change, but at this time, the input of the response model is the load shedding amount; by reducing a certain proportion of the load, the system frequency f s,4 is restored within the allowable error range. At this time, the total amount of load reduced is the load shedding amount

[0236] In step S5, in combination with the time periods when adjacent extreme events occur, analyze the complex dynamic process of the frequency response of the power system under the influence of extreme events:

[0237] During the invasion of extreme events, the power system is often affected by multiple faults. This causes new faults to quickly intervene during the frequency regulation process of the power system before the transient process of a single fault is completely over, resulting in a complex dynamic process. By obtaining the time periods when adjacent faults occur and combining with the frequency regulation process of the power system, the response and recovery process of the power system frequency after being disturbed under the influence of multiple faults can be obtained. Therefore, the frequency fluctuations, power attenuation, and oscillation characteristics experienced by the system in a short period can be deeply analyzed, and then the complex dynamic coupling relationships among various links can be revealed (the dynamic coupling relationships are mainly simulated by the typhoon model, the coupling of typhoons and meteorological events of secondary disasters accompanied by typhoons, and their joint influence on the dynamic process of the power system from the perspectives of time and space), providing key data support for constructing a comprehensive system resilience assessment model and laying a theoretical foundation for formulating more accurate protection and recovery strategies.

[0238] In step S6, a resilience assessment process based on the system frequency is proposed, as Figure 2 shown, and it specifically includes the following steps:

[0239] 1) Input the historical data or prediction information of a certain typhoon, the power grid structure, and the geographical location relationship between the typhoon and the power grid;

[0240] 2) Calculate the failure rates of transmission lines connected to power plants / substations at different times and construct an extreme event fault set;

[0241] 3) Determine the evaluation time period and analyze the interval time of extreme events;

[0242] 4) Taking 50Hz as the initial frequency, analyze the frequency regulation process of the power system in the first extreme event, analyze the frequency regulation process of the power system in the first extreme event, and model the response and recovery of the power system frequency in the first extreme event;

[0243] 5) Judge whether the next extreme event occurs during the frequency regulation process of the previous extreme event;

[0244] 5.1) If it occurs, calculate the initial frequency of the power system when the next extreme event occurs, and analyze the frequency regulation process of the power system in the current extreme event with this frequency as the initial frequency, and complete the modeling of the response and recovery of the power system frequency;

[0245] 5.2) If it does not occur, calculate the resilience index of the power generation system in the previous extreme event;

[0246] 6) Judge whether the extreme event ends; if it is satisfied, end the evaluation process; if it is not satisfied, continue to evaluate the next extreme event and return to execute step 5.1).

[0247] The resilience evaluation index defined by the present invention can accurately and comprehensively reflect the resilience level of the power generation system under extreme events.

Claims

1. A comprehensive evaluation method for power generation system resilience considering multi-fault evolution under extreme weather conditions, characterized in that: The steps include: S1: Analyze the impact of extreme meteorological disasters on power generation systems and construct extreme event fault sets under specific disasters; S2: For a single extreme event, construct the frequency regulation process of the power system, including the response and recovery process after the frequency is disturbed; S3: Based on the transient and steady-state characteristics of the system, quantitatively define the resilience model and indicators of the system at different stages and states under impact, and propose comprehensive resilience quantitative indicators from the three dimensions of time, frequency, and power, including absolute and relative indicators, to reflect the impact of extreme disasters on the power generation system; S4: Based on the proposed comprehensive quantitative index of resilience and the frequency regulation process of the power system, the resilience evaluation index of the power generation system is calculated; S5: Combined with the periods of adjacent extreme events, the complex dynamic process of the frequency response of the power system under the influence of extreme events is analyzed; S6: Based on steps S1 to S5, according to the extreme event fault set under specific disasters and the complex dynamic process of the evolution of adjacent extreme events, a resilience assessment process based on system frequency is proposed to evaluate the resilience of the power generation system.

2. According to claim 1, a comprehensive evaluation method for the resilience of a power generation system considering the evolution of multiple faults under extreme weather conditions is characterized in that: The method for constructing the extreme event fault set under a specific disaster in step S1 is: Typhoon historical data is input, and based on the typhoon disaster physical model and the component failure rate model of the transmission line under typhoon disaster, the Monte Carlo scenario generation method is used to generate a corresponding number of fault sample sets. Finally, the scenario reduction method based on probability distance is used to obtain the typical scenarios of various extreme events.

3. According to claim 2, a comprehensive evaluation method for the resilience of a power generation system considering the evolution of multiple faults under extreme weather conditions is characterized in that: The construction method of the typhoon disaster physical model and the component failure rate model of the transmission line under the typhoon disaster is as follows: Typhoon disaster physical model: The Batts typhoon model is used to simulate the wind speed of the typhoon wind field; Component failure rate model: The strength of the component obeys the following normal distribution, as shown in formula (1); when the difference between the load borne by the component and its own strength is greater than 0, the component is in an unreliable operating state, and the component failure rate is calculated by formula (2); Where f(R) is the probability distribution of component strength, R is the component strength, μ p and δ p are the strength parameters related to the line and tower respectively. is the failure probability of the kth component on the transmission line at time t after the typhoon lands, R k (t) and F k (t) are the component strength and wind load of the kth component at time t after the typhoon landed; In the same overhead line, the conductors and towers are connected in series. Any line failure will cause the entire line to be out of operation. The series model is used to obtain the single line failure probability, as shown in formula (3); Where P i line (t) is the failure probability of the ith transmission line at time t after the typhoon lands, m1 and m2 are the number of towers and conductor spans in the line, respectively. and are the failure probabilities of towers and conductors at time t after the typhoon lands.

4. According to claim 1, a comprehensive evaluation method for the resilience of a power generation system considering the evolution of multiple faults under extreme weather conditions is characterized in that: When a single extreme event occurs in step S2, the frequency regulation process of the power system includes: When the frequency drops beyond the allowable range, all generators participating in power generation in the system will undergo primary, secondary and tertiary frequency regulation processes in succession until the power in the system is rebalanced and the system frequency reaches a new steady-state value that is within the frequency safety range; the primary frequency regulation process is completed by the speed regulation system of the prime mover, which responds quickly to changes in the system frequency; the secondary frequency regulation process is completed by the system's rotating reserve and various flexible resources, allowing the frequency to be adjusted without difference; the tertiary frequency regulation process is completed by load shedding, which can keep the frequency deviation caused by extreme events within the allowable range.

5. According to claim 4, a comprehensive evaluation method for power generation system resilience considering multiple fault evolution under extreme weather conditions is characterized in that: The power system frequency regulation process in step S2 includes the power system frequency response and recovery process, which is specifically as follows: Constructing power system frequency response and recovery model: Power system primary frequency regulation model: In t s,1 The transfer function of the overall dynamic frequency response model of the power system when the sth extreme event occurs is shown in formula (4). The system speed regulator group is equivalent to a virtual speed regulator of the same form. After the equivalent, formula (4) is transformed into formula (5); Where Δf s,1* is the per-unit value of the frequency change during the frequency modulation process of the sth extreme event, ΔP S is the unbalanced power generated by the power system when the sth extreme event occurs, G I (s) is the system inertial response transfer function, H G (s) and H equ (s) are the transfer functions of the thermal power unit speed governor group before and after the whole system is equivalent, N G is the number of thermal power units in the system, is the rated output of thermal power unit i, u i is the start / stop status of thermal power unit i, which is 1 when it is on and 0 when it is off. L is the load frequency adjustment coefficient, the base value of the load is the rated value of the frequency at time t0 in the left neighborhood at time t after the typhoon landed, H i and D i are the inertia time constant and damping coefficient of generator set i, μ equ , They are the system equivalent adjustment, proportional and integral coefficients respectively; μ i , are the adjustment difference, proportional and integral coefficients of unit i respectively; Secondary frequency regulation model of power system: When the equivalent speed regulator of the whole system returns to steady state during a frequency modulation process, the time is assumed to be t s,2 , the corresponding system frequency value is f s,2 ; If the unbalanced power caused by the sth extreme event is small, f s,2 If the unbalanced power caused by the sth extreme event is large, f s,2 If it has not recovered to the allowable error range, a secondary frequency regulation process is required, and the system enters the frequency recovery stage from the frequency response stage. The above-mentioned power system equivalent dynamic response model is used to calculate the frequency change, but at this time the input of the response model is the support power of the thermal power unit and all flexible loads. The transfer function is shown in formula (6); Where: Δf s,2* (s) is the per-unit value of the frequency change in the secondary frequency modulation process of the sth extreme event; It is the sum of the emergency power support of all flexible loads, and its form is considered as a time-delayless step input in the opposite direction of the disturbance; G is the sum of the positive reserve of the remaining operating thermal power units in the system, which is considered to be a time-delayed, limited-amplitude ramp input in the opposite direction of the disturbance; I (s) and H equ (s) are the system inertia response transfer function and the equivalent governor transfer function of the whole system of thermal power units, respectively. The subscript i in the function is the unit number; N G is the number of thermal power units in the system; u i is the start / stop state of thermal power unit i, which is 1 when it is on and 0 when it is off; γ i is the ramp rate of thermal power unit i; is the standby start-up delay of thermal power unit i; is the full standby time of thermal power unit i; and The starting time of the secondary frequency modulation is t s,2 ; Due to the existence of the ramp reserve of the thermal power unit, it is impossible to obtain an analytical expression when solving the maximum frequency difference of the system. Therefore, the ramp reserve is equivalent to the step reserve. The reserve response function and parameter calculation method of the thermal power unit i before and after the equivalent are shown in formulas (7)-(9); Where: y i (τ) and are the standby response functions of thermal power unit i before and after equivalent respectively; δ(i) is the unit step function; is the number of rising edges of the step reserve of thermal power unit i, and the number of steps is The number range is L i is the height of the single-step spare ladder of the i-step thermal power unit; is the trigger time of the jth rising edge of the step standby of thermal power unit i, and its starting time is the starting time of the secondary frequency modulation t s,2 ; Power system tertiary frequency regulation model: When the frequency returns to steady state during the secondary frequency modulation process, assume that the time is t s,3 , the corresponding system frequency value is f s ,3 If f s,3 If it has recovered to within the allowable error range, then only the resilience evaluation index in the primary and secondary frequency modulation processes needs to be calculated, and the system enters the normal frequency stage from the frequency recovery stage; if f s,3 If the frequency has not yet recovered to the allowable error range, three frequency modulation processes are required. The system enters the load reduction stage from the frequency recovery stage. The above-mentioned power system equivalent dynamic response model is still used to calculate the frequency change, but the input of the response model is the load reduction amount at this time. The transfer function is shown in formula (10): Where: Δf s,3* (s) is the per-unit value of the frequency change during the tertiary frequency modulation process of the sth extreme event; The load reduction amount.

6. According to claim 5, a comprehensive evaluation method for power generation system resilience considering multiple fault evolution under extreme weather conditions, characterized in that: In step S3, comprehensive quantitative indicators of toughness are proposed from three dimensions: time, frequency, and power, specifically including: The system frequency regulation process under extreme disaster events is divided into four stages, namely, the normal frequency stage, the frequency response stage, the frequency recovery stage, and the load shedding stage; the frequency response stage, the frequency recovery stage, and the load shedding stage correspond to the primary frequency regulation process, the secondary frequency regulation process, and the tertiary frequency regulation process, respectively; Taking frequency as the performance indicator of the power generation system, the first stage is the normal frequency stage (t s,0 -t s,1 ); The second stage is the frequency response stage (t s,1 -t s,2 ), at t s,1 At time s, the extreme event occurs and the frequency starts to drop from 50Hz. At this time, the system automatically performs a frequency modulation process and the frequency drops to the lowest point f s nadir , the corresponding time is Due to the automatic regulation of the speed regulator and load, the frequency gradually returns to f s,2 ; The third stage is the frequency recovery stage (t s,2 -t s,3 ), at this time, due to the frequency response stage, there is a difference in regulation, and it is necessary to increase the spinning reserve and various flexible loads to supplement the power difference. s,2 The secondary frequency modulation process is carried out at all times, and the frequency is changed from f s,2 Gradually recover to f s,3 , then we need to judge f s,3 Whether it is restored to the allowable error range, if not, it enters the fourth stage; the fourth stage is the load reduction stage (t s,3 -t s,4 ), at this time f s,3 If the frequency is not restored to within the allowable error range, a small range of load reduction is required to restore the frequency. s,4 Time to restore to f s,4 , which is within a safe range; by analyzing the state evolution of the power generation system before and after the sth extreme event, the resilience evaluation index of the power generation system is defined based on the performance of the system frequency.

7. A comprehensive evaluation method for power generation system resilience considering multiple fault evolution under extreme weather conditions according to claim 6, characterized in that: The resilience evaluation index of the power generation system in step S3 includes: Duration of one FM It reflects the response time of the system to perform a frequency modulation at the sth extreme event, as shown in formula (11): Where: t s,1 is the time when the sth extreme event occurs, t s,2 is the moment when a frequency modulation process ends in the sth extreme event; Maximum frequency difference Δf s max (Hz):Δf s max represents the maximum deviation of the system frequency during the sth extreme event, and Together they reflect the transient stability of the frequency, as shown in formula (12): Where: f s nadir is the lowest frequency point in the frequency modulation process of the sth extreme event; Frequency recovery time It reflects the response time of the system to perform secondary frequency modulation in the sth extreme event, as shown in formula (13): Where: t s,2 is the moment when the frequency modulation process of the sth extreme event ends, t s,3 is the moment when the secondary frequency modulation process ends in the sth extreme event; The frequency recovers to the final value f s final (Hz): f s final It reflects the frequency regulation capability and disaster resistance capability of the system itself in the sth extreme event, as shown in formula (14): f s final =f n,3 (14) Where: f n,3 is the frequency value recovered after the system performs one or two frequency modulations in the sth extreme event; Load reduction It reflects the extent to which the system load is affected in the sth extreme event; Frequency drop / rise time T p It is the expected time for the system frequency to drop from the standard frequency to the lowest point / rise to the highest point, reflecting the primary frequency modulation capability of the system. The expression is: Where: T p The duration of the system frequency dropping to the lowest point / rising to the highest point; Frequency resistance time T r It is the expected time for the system frequency to rise / fall from the lowest / highest point to the load shedding frequency after the system spare capacity and resilience resources respond in time, reflecting the system's abundance. The expression is: Where: T r It is the duration of the system frequency rising / falling from the lowest / highest point to the load shedding frequency; Frequency recovery time T s is the response time of the system frequency rising / falling from the load shedding frequency to the standard frequency, and the expression is: T s =t s,4 -t s,3 (17) Where: T s The duration of the system frequency rising / falling from the load shedding frequency to the standard frequency; Frequency change duration T d is the total response time of the system frequency changing during the disaster attack process, expressed as: T d =T p +T r +T s (18) Where: T p 、T r 、T s They are the frequency drop / rise time, frequency resistance time, and frequency recovery time of the system frequency; Resistance to frequency modulation power P R The frequency resistance time T r The expected value of frequency regulation power provided by the internal system spare capacity and resilience resources is expressed as: P R =P R,r (19) Where: P R,r is the frequency resistance time T r Frequency regulation power provided by internal system spare capacity and resilience resources; Adaptation rate R A It reflects the short-time power ramping capability of the frequency-regulated units and the abundance of the resilience resources of the system. The expression is: Where: P R To resist frequency modulation power; T r To defend frequency against time; Frequency resistance time ratio T r % reflects the impact of different spare capacities and resilience resources on the improvement of system frequency robustness, and the expression is: Where: T p 、T r 、T s They are the frequency drop / rise time, frequency resistance time, and frequency recovery time of the system frequency in state i and interval j respectively; Load shedding frequency difference ratio Δf s % reflects the influence of different load shedding amounts on the system frequency recovery force, and the expression is: Where: f * Set to the standard value of 50Hz, f cut 、f max / min They are the frequency change obtained by load shedding and the frequency change when the system frequency drops from the standard frequency to the lowest point / rises to the highest point.

8. The comprehensive evaluation method for power generation system resilience considering multiple fault evolution under extreme weather conditions according to claim 7 is characterized in that: The calculation of the power generation system resilience evaluation index in step S4 includes: Solution of the primary frequency regulation model of the power system: Ignore the damping coefficient of the thermal power unit, that is, set the generator damping coefficient D to 0; s,1 When the sth extreme event occurs, formula (5) is transformed into formula (23); Where: Δf s,1* (s) is the per-unit value of the frequency change during the frequency modulation process of the sth extreme event; ΔP S is the unbalanced power caused by the sth extreme event, and the fluctuation form is considered to be a step fluctuation; k G is the system equivalent inertia coefficient; λ, α, β are the transformation coefficients of the transfer function of the system dynamic frequency response model; make The time domain expression of the per-unit value of the frequency change in the primary frequency modulation process of the sth extreme event is obtained as formula (24): According to formula (24), the frequency deviation f between the end time and the start time of a frequency modulation is calculated. s,2*- f s,1* As shown in formula (25), the frequency deviation f between the lowest frequency point and the initial frequency is s nadir*- f s,1* As shown in formula (26), the time difference between the lowest frequency point and the initial time is As shown in formula (27), the time difference between the end of a frequency modulation and the lowest frequency point is As shown in formula (28): According to formulas (25)-(28), the calculation methods of the two frequency indicators in a frequency modulation are given as follows: Duration of a frequency modulation The calculation is shown in formula (29): Maximum frequency difference The calculation is shown in formula (30): Solution of the secondary frequency regulation model of the power system: It is known that the end time of a frequency modulation is t s,2 The frequency value of the system is f s,2 If f s,2 If it has not recovered to the allowable error range, a secondary frequency regulation process is required. The system enters the frequency recovery stage from the frequency response stage. At this time, the input of the response model is the support power of the thermal power unit and all flexible loads. Formula (6) is transformed into formula (31): g(s) in formula (31) and Δf in formula (23) s,1* (s) has the same form, and the derived frequency difference has the same form, so we focus on analyzing h i (s); The frequency deviation f between the end and start time of the secondary frequency modulation can be obtained by derivation s,3* -f s,2* As shown in formula (34), the time difference t between the end time of the secondary frequency modulation and the end time of the primary frequency modulation is s,3 -t s,2 As shown in formula (33), the parameters are calculated by formula (34): In the formula, p i is the first Step ladder; According to formulas (32)-(35), the calculation methods of the two frequency indicators in primary frequency modulation are given as follows: Frequency recovery time The calculation is shown in formula (35): Frequency recovery final value The calculation is shown in formula (36): f s final =(f s,3* -f s,2* )+(f s,2* -f s,1* )+f s,1* (36) Solution of the three-frequency regulation model of the power system: The end time of the secondary frequency modulation is known to be t s,3 The frequency value of the system is f s,3 If f s,3 If the frequency has not yet recovered to the allowable error range, a third frequency regulation process is required. The system enters the load reduction stage from the frequency recovery stage. The frequency change is calculated using the power system equivalent dynamic response model that is the same as the primary frequency regulation. However, the input of the response model is the load reduction amount. By reducing a certain proportion of the load, the system frequency f s,4 When the load is restored to the allowable error range, the total amount of load reduced is the load reduction.

9. The comprehensive evaluation method for resilience of power generation system considering multi-fault evolution under extreme weather conditions according to claim 1 is characterized in that: In step S5, the complex dynamic process of the frequency response of the power system under the influence of the extreme event is analyzed in combination with the period of occurrence of the adjacent extreme event, including: By combining the time periods when adjacent faults occur, we conduct in-depth analysis of the frequency fluctuations, power attenuation, and oscillation characteristics experienced by the system in a short period of time, thereby revealing the complex dynamic coupling relationship between each link.

10. The comprehensive evaluation method for resilience of power generation system considering multi-fault evolution under extreme weather conditions according to claim 1 is characterized in that: The step S6 proposes a resilience assessment process based on system frequency, which specifically includes the following steps: 1) Input the historical data or forecast information of a certain typhoon, the grid structure, and the geographical location relationship between the typhoon and the grid; 2) Calculate the failure rate of the transmission lines connected to the power plant / substation at different times and construct the extreme event failure set; 3) Determine the evaluation period and analyze the interval between extreme events; 4) Taking 50 Hz as the initial frequency, analyze the frequency modulation process of the power system in the first extreme event, analyze the frequency modulation process of the power system in the first extreme event, and model the response and recovery of the power system frequency in the first extreme event; 5) Determine whether the next extreme event will occur during the frequency modulation of the previous extreme event: 5.1) If it occurs, calculate the initial value of the frequency of the power system when the next extreme event occurs, use this frequency as the initial frequency to analyze the frequency regulation process of the power system in the current extreme event, and complete the response and recovery modeling of the power system frequency; 5.2) If it has not occurred, calculate the power generation system resilience index of the previous extreme event; 6) Determine whether the extreme event has ended; if so, end the assessment process; If not, continue to evaluate the next extreme event and return to step 5.1).

Images