A Method for Comprehensive Assessment of the Operational Reliability and Frequency of a High-Proportion Wind Power Power System

Abstract

The present invention relates to the technical field of high-proportion wind power power systems, and in particular to a comprehensive evaluation method for the operation reliability and frequency of a high-proportion wind power power system. This method takes into account the coupling relationship between system reliability and frequency risk, comprehensively analyzes the system parameters under various scenarios of the system, and based on the frequency-sensitive reliability model and the maximum frequency deviation under each scenario, calculates the steady-state probability of each scenario; then, by linearizing the frequency deviation curve and the unbalanced power curve of the system, and combining the probabilities of each scenario, calculates the comprehensive reliability index and the frequency deviation index of the system, so as to achieve the purpose of quantifying the system reliability level. Through the above process, the present invention can comprehensively evaluate the reliability and frequency risk of the power system within the operation reliability time scale. Compared with the prior art, the present invention improves the accuracy of the operation reliability evaluation of the high-proportion wind power power system.

Description

Technical Field

[0001] The present invention relates to the technical field of high-proportion wind power power systems, and particularly to a method for comprehensively evaluating the operation reliability and frequency of a high-proportion wind power power system. Background Art

[0002] The evaluation of the operation reliability of a high-proportion wind power power system refers to studying the reliability level of the system in the short term in the future under the current state by considering the influence of factors such as the operation state and real-time operation environment of the high-proportion wind power power system. With the low-carbon transformation globally, the grid-connected operation of high-proportion wind power has become an important feature of modern power systems. Therefore, the operation reliability of a high-proportion wind power power system has become increasingly important.

[0003] Different from conventional units, wind power generation highly depends on resource endowments, has strong uncertainty and poor adjustability. Due to the strong randomness and volatility of wind power, in the case of high wind power penetration, the system may lack sufficient regulation capacity and ramping ability. Therefore, maintaining the reliability and frequency security of the power system is very challenging. At present, the research on the reliability of high-proportion wind power power systems mostly focuses on the influence of wind power fluctuations on the medium- and long-term reliability of the system, and cannot quantify the influence of short-term system frequency fluctuations on the system reliability, thus unable to meet the actual needs of system operation reliability evaluation.

[0004] In fact, the large-scale grid connection of wind power will exacerbate the system frequency fluctuations caused by unit failures (reliability events) or wind speed fluctuations, affect the commissioning and dispatching of units, and even deteriorate the failure rate of units, increasing the operation and maintenance costs of the system. The increase in system frequency fluctuations may trigger the frequency protection actions of units and even cause generator tripping (reliability events). This means that there is a certain coupling relationship between system frequency and reliability. In addition, since wind turbines achieve maximum power tracking control through converters, the generator rotors of wind turbines are decoupled from the grid frequency and cannot achieve frequency response like conventional units. With the large-scale grid connection of wind power and the gradual retirement of conventional units, the inertia of the system is significantly reduced, the frequency regulation resources of the system are reduced, and the frequency fluctuations are more obvious.

[0005] The problem of system frequency risk in high-proportion wind power power systems is prominent, and it is necessary to expand the system's frequency regulation resources. How to control renewable energy units such as wind power and actively involve them in the system's frequency control is an effective measure to alleviate the system frequency risk at present. At present, the research on power system reliability and system frequency control is relatively fragmented and cannot depict the coupling relationship between the two. How to comprehensively evaluate the power system reliability and frequency risk within the time scale of operation reliability is a technical problem that needs to be solved urgently by those skilled in the art at present. Summary of the Invention

[0006] In view of the deficiencies of the above-mentioned existing technologies, the present invention provides a method for comprehensively evaluating the operation reliability and frequency of a high-proportion wind power power system, which can comprehensively evaluate the reliability and frequency risk of the power system within the time scale of operation reliability.

[0007] To solve the above technical problems, the present invention adopts the following technical solutions:

[0008] A method for comprehensively evaluating the operation reliability and frequency of a high-proportion wind power power system, comprising the following steps:

[0009] Step 1: Obtain the reliability parameters, electrical parameters of the units, load data of the system, and wind speed data; the units include thermal power units and wind power units;

[0010] Step 2: Based on the reliability parameters, electrical parameters, and operation condition requirements of the thermal power units, establish a frequency-sensitive reliability model of the thermal power units;

[0011] Step 3: Enumerate the fault events of each order of the system to generate system scenarios;

[0012] Step 4: Calculate the system parameters under each scenario, and establish a frequency response model considering the coordinated frequency regulation of thermal power units and wind power units;

[0013] Step 5: Analyze the frequency dynamic response process and unbalanced power change process of the system under each scenario, and solve the maximum frequency deviation, frequency deviation after primary frequency regulation, frequency deviation after secondary frequency regulation, system primary frequency regulation time, and system secondary frequency regulation time under each scenario;

[0014] Step 6: Determine whether the frequency dynamic response process of each system scenario has been fully simulated. If so, go to Step 7; if not, return to Step 4;

[0015] Step 7: Based on the frequency-sensitive reliability model and the maximum frequency deviation under each scenario, correct the unit failure rate data, and use the Markov equation to calculate the steady-state probability of each scenario;

[0016] Step 8: Linearly process the frequency deviation curve and unbalanced power curve of the system, and combine the probabilities of each scenario to calculate the comprehensive reliability index and frequency deviation index of the system, and quantify the operation reliability level of the system.

[0017] Preferably, in Step 2, the operation condition requirements include the frequency protection measures of the thermal power units, and the frequency protection measures include low-frequency protection thresholds and high-frequency protection thresholds

[0018] The frequency-sensitive reliability model is:

[0019]

[0020] In the formula, U m is the statistical average value of the failure probability of the thermal power unit; is the lower bound of the frequency for the thermal power unit to maintain normal operation; is the upper bound of the frequency for the thermal power unit to maintain normal operation, and f is the actual value of the system frequency.

[0021] Preferably, in step 3, each order of fault accidents of the system is enumerated by the state enumeration method to generate system scenarios, and the number of system scenarios generated by the enumeration is where N all is the total number of units in the system.

[0022] Preferably, in step 4, the system parameters include the equivalent ramp rate K of the system, the spinning reserve capacity SR of the system, and the equivalent inertia constant H i , the equivalent governor speed regulation coefficient R i , the system frequency deviation factor B i , and the unbalanced power ΔP of the system i ; where:

[0023]

[0024] In the formula, Δf is the frequency deviation of the system; N all is the total number of units in the system; K i is the ramp rate of the i-th unit; is the state of the i-th unit, 1 represents the normal operation state of the unit, and 0 represents the fault state of the unit; P load is the load power level, which is used as the reference value for the system power per unit;

[0025]

[0026] In the formula, P i is the actual output of the i-th unit, and P i,min is the minimum allowable output of the i-th unit;

[0027]

[0028] In the formula, H i is the inertia coefficient of the i-th unit; Cap i is the rated capacity of the i-th unit;

[0029]

[0030] In the formula, is the speed regulation coefficient of the i-th unit; f rate is the rated frequency of the system;

[0031] B i = 1 / R i + D i ;

[0032] wherein, D i is the system load regulation coefficient;

[0033] ΔP i = ΔP Wind + ΔP G ;

[0034] wherein, ΔP Wind is the system power fluctuation caused by wind speed fluctuation; ΔP G is the system power fluctuation caused by unit failure.

[0035] Preferably, the wind turbine is a doubly-fed wind turbine; in step 4, in the frequency response model, the doubly-fed wind turbine adopts a virtual inertia control operation mode, and the rotor kinetic energy is lifted to participate in the system frequency response process, and the following relationship is satisfied:

[0036]

[0037] wherein, f kopt is the virtual inertia factor, ω r0 is the sampling value of the rotor speed of the doubly-fed wind turbine before disturbance, k vir is the virtual inertia coefficient of the doubly-fed wind turbine, Δf is the system frequency deviation, p is the number of pole pairs of the doubly-fed wind turbine;

[0038]

[0039] wherein, ΔP WT is the active power change of the doubly-fed wind turbine, ω s is the synchronous rotor speed of the power grid, H eq is the equivalent inertia time constant of the doubly-fed wind turbine, f s is the system frequency;

[0040]

[0041] wherein, Δf kopt is the power regulation factor, K opt is the maximum power tracking coefficient, ω g is the rotor speed, ω max is the speed entering the constant power region, ω min is the starting speed of the rotor;

[0042] K OPPT = K opt (f kopt - Δf kopt );

[0043] In the formula, K OPPT is the power tracking coefficient of the fan in the virtual inertia control mode;

[0044]

[0045] In the formula, P OPPT is the output power of the fan in the virtual inertia control mode, ω0 is the rotor speed when entering the maximum power area, ω1 is the rotor speed when entering the constant speed area, and P max is the maximum output power of the doubly-fed fan.

[0046] Preferably, in step 5, the maximum frequency deviation Δf of the system max is calculated by the formula:

[0047] Δf max =-max{abs(Δf(t k ))};

[0048] In the formula, Δf(t k ) refers to the sampled value of the system frequency deviation at time t k ;

[0049] The calculation formula for the frequency deviation Δf1 after primary frequency modulation is:

[0050]

[0051] The calculation formula for the frequency deviation Δf2 after secondary frequency modulation is:

[0052]

[0053] Preferably, in step 5, the calculation process of the system primary frequency modulation time t1 and the system secondary frequency modulation time t2 includes:

[0054] Set the initial sampling time t = 0, the iteration step size is Δt, and the convergence accuracy is ε; calculate the primary frequency modulation steady-state deviation Δf1 and the secondary frequency modulation steady-state deviation Δf2 of the system in this state;

[0055] For the kth iteration cycle, the sampling time is t k = t k-1 +Δt; judge whether it satisfies the convergence condition |Δf(t k ) - Δf(t k-1 )|≤ε, |Δf(t k ) - Δf1|≤ε; if the condition is not satisfied, enter the (k + 1)th iteration cycle; if it is satisfied, output the corresponding sampling time t k as the system primary frequency modulation time t1:

[0056] Meanwhile, for the sampling time t of the k-th iteration loop k = t k-1 + Δt; It is also determined whether it satisfies the convergence condition |Δf(t k ) - Δf(t k-1 )| ≤ ε, |Δf(t k ) - Δf2| ≤ ε; If the condition is not satisfied, enter the (k + 1)-th iteration loop; if satisfied, output the corresponding sampling time t k as the secondary frequency regulation time t2 of the system:

[0057] Preferably, step 7 includes:

[0058] S71. Based on the frequency-sensitive reliability model and the maximum frequency deviation in each scenario, correct the unit failure probability parameters, so that

[0059]

[0060] wherein, U new is the corrected unit failure probability; λ new is the corrected unit failure rate; μ is the average repair rate of the unit;

[0061] S72. Based on the corrected generator failure rate and the Markov process, calculate the steady-state probability of each scenario.

[0062] Preferably, S72 includes:

[0063] S721. Construct a state space diagram according to the component state transition;

[0064] S722. Construct a state transition matrix T according to the state space diagram;

[0065] S723. Calculate the probability vector P of each state according to the Markov equation to obtain the steady-state probability of each scenario:

[0066] P(T - I) = 0;

[0067] wherein, I is the identity matrix.

[0068] Preferably, in step 8, the system comprehensive reliability index includes the expected energy not supplied (EENS) of the system and the expected indirect energy not supplied (EIENS) of the system; the expected energy not supplied (EENS) is used to describe the power outage risk; the expected indirect energy not supplied (EIENS) is used to describe the energy not supplied during the frequency regulation process; wherein:

[0069]

[0070] wherein, N is the number of system states to be considered, Pri is the probability of system state i, and t3 is the duration of load shedding in system state i;

[0071]

[0072] IENS i,1 = ΔP i × t1;

[0073]

[0074] In the formula, IENS i,1 is the expected value of indirect energy shortage during primary frequency regulation, and IENS i,2 is the expected value of indirect energy shortage during secondary frequency regulation.

[0075] Preferably, in S8, the frequency deviation index includes the expected number of times of system low-frequency overlimit ENUF, the expected duration of system low-frequency EUFD, the probability Pr LFE that a system low-frequency event occurs, and the probability Pr RLFE that a system low-frequency event occurs and recovers; where:

[0076]

[0077] In the formula, NUF i is the number of times of low-frequency overlimit in the i-th state. If there is no frequency overlimit in this state, then NUF i = 0; If the system frequency finally recovers to the normal range after primary or secondary frequency regulation, then NUF i = 1; Otherwise, NUF i = 2;

[0078]

[0079] In the formula, UFD i is the duration that the frequency of the i-th state is lower than the lower limit of the normal range;

[0080]

[0081] In the formula, L i is a binary logic variable used to determine whether state i meets the condition. When the maximum frequency deviation of the system in state i exceeds the lower limit, L i = 1; Otherwise, L i = 0;

[0082]

[0083] In the formula, Δf m is the steady-state frequency deviation of the system; R viis a binary logic variable used to determine whether the steady-state frequency deviation of state i meets the condition. If the deviation is within the allowable range, R vi = 1; otherwise, R vi = 0.

[0084] Compared with the prior art, the present invention has the following beneficial effects:

[0085] 1. The present invention proposes a method for comprehensively evaluating the operation reliability and frequency of a high-proportion wind power power system, which evaluates the operation reliability of the high-proportion wind power power system. This method considers the coupling relationship between system reliability and frequency risk, comprehensively analyzes the system parameters under each scenario of the system, and calculates the steady-state probability of each scenario based on the frequency-sensitive reliability model and the maximum frequency deviation under each scenario; then, by linearizing the frequency deviation curve and the unbalanced power curve of the system, and combining the probabilities of each scenario, the comprehensive reliability index and frequency deviation index of the system are calculated, so as to achieve the purpose of quantifying the system reliability level.

[0086] Through the above process, the present invention comprehensively considers the coupling relationship between system reliability and frequency risk, and can comprehensively evaluate the power system reliability and frequency risk within the operation reliability time scale. Compared with the prior art, the present invention improves the accuracy of the operation reliability evaluation of the high-proportion wind power power system.

[0087] 2. In order to solve the problem of scarce frequency regulation resources in the high-proportion wind power power system, the present invention proposes a frequency response model in which thermal power and wind power participate synergistically, which takes into account both the primary / secondary frequency modulation, ramp rate limit, and reserve limit of thermal power, and the virtual inertia response process of wind turbines. Through this frequency response model, the situation of frequency deterioration in the case of a high-proportion wind power system can be effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] In order to make the purpose, technical solution and advantages of the invention clearer, the present invention will be further described in detail below with reference to the drawings, where:

[0089] Figure 1 is the flowchart in the embodiment;

[0090] Figure 2 is the schematic structural diagram of the frequency response model in the embodiment;

[0091] Figure 3 is the schematic diagram of the virtual inertia control operation curve of the wind turbine in the embodiment;

[0092] Figure 4 is the schematic diagram of the correction process of the unit failure rate in the embodiment;

[0093] Figure 5Schematic diagram of the frequency deviation and unbalanced power curve of the linearization system in the embodiment;

[0094] Figure 6 Example diagram of the system frequency fluctuation of the test system in the specific example of the embodiment under the same wind speed and different wind power penetration rates;

[0095] Figure 7 Schematic diagram of the difference in system frequency deviation when the wind turbines of the test system in the specific example of the embodiment participate in system frequency control or not under the same power fluctuation conditions;

[0096] Figure 8 Schematic diagram of the influence on the calculation of the system reliability index EENS when considering or not considering the coupling relationship between reliability and frequency risk of the test system in the specific example of the embodiment; Detailed implementation method

[0097] The following is a further detailed description through specific implementation methods:

[0098] Embodiment:

[0099] As Figure 1 shown, a method for comprehensively evaluating the operation reliability and frequency of a high-proportion wind power power system is disclosed in this embodiment, including the following steps:

[0100] Step 1: Obtain the reliability parameters, electrical parameters, load data of the system, and wind speed data of the units; the units include thermal power units and wind turbines. Specifically, in implementation, the wind turbines are doubly-fed wind turbines; the reliability parameters, including the failure rate data and repair rate data of thermal power units, come from the historical statistical data of thermal power units.

[0101] Step 2: Based on the reliability parameters, electrical parameters, and operating condition requirements of thermal power units, establish a frequency-sensitive reliability model of thermal power units. Among them, the operating condition requirements include the frequency protection measures of thermal power units, and the frequency protection measures include the low-frequency protection threshold and the high-frequency protection threshold

[0102] The frequency-sensitive reliability model is:

[0103]

[0104] In the formula, U m is the statistical average value of the failure probability of thermal power units; is the lower frequency bound for thermal power units to maintain normal operation; is the upper frequency bound for thermal power units to maintain normal operation, and f is the actual value of the system frequency.

[0105] Step 3: Enumerate the fault events of each order of the system to generate system scenarios. Specifically, in implementation, the fault accidents of each order of the system are enumerated by the state enumeration method to generate system scenarios, and the number of system scenarios generated is where N all is the total number of units in the system.

[0106] Step 4: Calculate the system parameters under each scenario and establish a frequency response model considering the coordinated frequency regulation of thermal power units and wind power units. The architecture of the frequency response model is as Figure 2 shown. Specifically, the system parameters include the equivalent ramp rate K of the system, the spinning reserve capacity SR of the system, the equivalent inertia constant H i of the system, the equivalent governor speed regulation coefficient R i of the system, the system frequency deviation factor B i of the system, and the unbalanced power ΔP i of the system; where:

[0107]

[0108] In the formula, Δf is the frequency deviation of the system; N all is the total number of units in the system; K i is the ramp rate of the i-th unit; state i is the state of the i-th unit, 1 represents the normal operation state of the unit, and 0 represents the unit is in a fault state; P load is the load power level, which is used as the reference value for the system power per unit.

[0109]

[0110] In the formula, P i is the actual output of the i-th unit, and P i,min is the minimum allowable output of the i-th unit;

[0111]

[0112] In the formula, H i is the inertia coefficient of the i-th unit; Cap i is the rated capacity of the i-th unit;

[0113]

[0114] In the formula, is the speed regulation coefficient of the i-th unit; f rate is the rated frequency of the system;

[0115] B i = 1 / R i + D i ;

[0116] In the formula, D i is the system load regulation coefficient;

[0117] ΔP i =ΔP Wind +ΔP G ;

[0118] In the formula, ΔP Wind is the system power fluctuation caused by wind speed fluctuation; ΔP G is the system power fluctuation caused by unit failure.

[0119] In the frequency response model, the virtual inertia control operation mode is adopted for the wind turbine (the virtual inertia control operation curve of the wind turbine is as shown in Figure 3 ), and the rotor kinetic energy is lifted to participate in the system frequency response process, and the following relationship is satisfied:

[0120]

[0121] In the formula, f kopt is the virtual inertia factor, ω r0 is the sampling value of the rotor speed of the doubly-fed wind turbine before the disturbance, k vir is the virtual inertia coefficient of the doubly-fed wind turbine, Δf is the system frequency deviation, and p is the number of pole pairs of the doubly-fed wind turbine;

[0122]

[0123] In the formula, ΔP WT is the change in active power of the doubly-fed wind turbine, ω s is the synchronous rotor speed of the power grid, H eq is the equivalent inertia time constant of the doubly-fed wind turbine, f s is the system frequency;

[0124]

[0125] In the formula, Δf kopt is the power regulation factor, K opt is the maximum power tracking coefficient, ω g is the rotor speed, ω max is the speed entering the constant power region, ω min is the starting speed of the rotor;

[0126] K OPPT =K opt (f kopt -Δf kopt );

[0127] In the formula, K OPPTis the power tracking coefficient of the wind turbine in the virtual inertia control mode (also known as: OPPT mode);

[0128]

[0129] wherein, P OPPT is the output power of the wind turbine in the virtual inertia control mode, ω0 is the rotor speed when entering the maximum power region, ω1 is the rotor speed when entering the constant speed region, and P max is the maximum output power of the doubly-fed wind turbine.

[0130] Step 5: Analyze the frequency dynamic response process and unbalanced power change process of the system under each scenario, and solve the maximum frequency deviation, frequency deviation after primary frequency modulation, frequency deviation after secondary frequency modulation, system primary frequency modulation time, and system secondary frequency modulation time under each scenario.

[0131] Specifically, when implemented, the calculation formula for the maximum frequency deviation Δf max of the system is:

[0132] Δf max =-max{abs(Δf(t k ))};

[0133] wherein, Δf(t k ) refers to the sampled value of the system frequency deviation at time t k ;

[0134] The calculation formula for the frequency deviation Δf1 after primary frequency modulation is:

[0135]

[0136] The calculation formula for the frequency deviation Δf2 after secondary frequency modulation is:

[0137]

[0138] The calculation process of the system primary frequency modulation time t1 and the system secondary frequency modulation time t2 includes:

[0139] Set the initial sampling time t = 0, the iteration step size is Δt, and the convergence accuracy is ε; calculate the primary frequency modulation steady-state deviation Δf1 and the secondary frequency modulation steady-state deviation Δf2 of the system in this state;

[0140] For the kth iteration cycle, the sampling time is t k =t k-1 +Δt; judge whether it satisfies the convergence condition |Δf(t k )-Δf(t k-1 |)≤ε,|Δf(t k) - |Δf1| ≤ ε; If the condition is not satisfied, enter the (k + 1)-th iteration loop; if satisfied, output the corresponding sampling time t k As the primary frequency regulation time t1 of the system:

[0141] Meanwhile, for the sampling time t of the k-th iteration loop k = t k-1 + Δt; Also judge whether it satisfies the convergence condition |Δf(t k ) - Δf(t k-1 )| ≤ ε, |Δf(t k ) - Δf2| ≤ ε; If the condition is not satisfied, enter the (k + 1)-th iteration loop; if satisfied, output the corresponding sampling time t k As the secondary frequency regulation time t2 of the system:

[0142] Step 6: Judge whether the frequency dynamic response processes of all scenarios of the system are all simulated. If so, go to Step 7; if not, return to Step 4.

[0143] Step 7: Based on the frequency-sensitive reliability model and the maximum frequency deviation under each scenario, correct the unit failure rate data, and use the Markov equation to calculate the steady-state probability of each scenario; the correction process of the unit failure rate is as Figure 4 shown.

[0144] Specifically, Step 7 includes:

[0145] S71. Based on the frequency-sensitive reliability model and the maximum frequency deviation under each scenario, correct the unit failure probability parameters to make

[0146]

[0147] where U new is the corrected unit failure probability; λ new is the corrected unit failure rate; μ is the average repair rate of the unit;

[0148] S72. Based on the corrected generator failure rate and the Markov process, calculate the steady-state probability of each scenario.

[0149] Among them, S72 includes:

[0150] S721. Construct a state space diagram according to the component state transition;

[0151] S722. Construct a state transition matrix T according to the state space diagram; Taking the two-component four-state as an example, its system state transition matrix is:

[0152]

[0153] S723. Calculate the probability vector P for each state according to the Markov equation to obtain the steady-state probabilities of each scenario:

[0154] P(T - I) = 0;

[0155] In the formula, I is the identity matrix.

[0156] Step 8: Linearize the frequency deviation curve and the unbalanced power curve of the system, and combine the probabilities of each scenario to calculate the comprehensive reliability index and the frequency deviation index of the system, and quantify the operating reliability level of the system.

[0157] Among them, the linearization of the frequency deviation curve and the unbalanced power curve of the system is as Figure 5 shown. Considering that the relationship between the frequency regulation process and time is not completely linear, the primary frequency regulation lasts for several seconds, and then during the secondary frequency regulation process, the frequency almost linearly returns to the rated value according to the ramp rate of the system units. During the primary frequency regulation process, since the average value of the changing power is approximately equal to ΔP i , therefore, ΔP i is used as the power shortage during the primary frequency regulation process. According to the linearized frequency and power change curves during the secondary frequency regulation process, the power shortage during the secondary frequency control process is calculated.

[0158] Specifically, the comprehensive reliability index of the system includes the expected energy not supplied (EENS) of the system and the expected indirect energy not supplied (EIENS) of the system. The expected energy not supplied (EENS) is used to describe the power outage risk. If the system frequency cannot return to the allowed normal range after the regulation process, then load shedding is required to restore the system frequency. The expected indirect energy not supplied (EIENS) is used to describe the energy shortage during the frequency regulation process. The power imbalance during the frequency regulation process is not directly reflected in the load shedding, but in the stability of the system frequency. Among them:

[0159]

[0160] In the formula, N is the number of system states to be considered, Pr i is the probability of system state i, and t3 is the duration of load shedding in system state i;

[0161]

[0162] IENS i,1 = ΔP i × t1;

[0163]

[0164] In the formula, IENS i,1Indirect energy deficiency expectation in the primary frequency regulation process, IENS i,2 Indirect energy deficiency expectation in the secondary frequency regulation process.

[0165] The frequency deviation index includes the expected number of times of system low frequency crossing ENUF, the expected duration of system low frequency EUFD, the probability Pr of the system having a low frequency event LFE and the probability Pr of the system having a low frequency event and recovering RLFE ; where:

[0166]

[0167] In the formula, NUF i is the number of times of low frequency crossing in the i-th state. If there is no frequency crossing in this state, then NUF i = 0; if the system frequency finally returns to the normal range after primary or secondary frequency regulation, then NUF i = 1; otherwise NUF i = 2;

[0168]

[0169] In the formula, UFD i is the duration that the frequency of the i-th state is lower than the lower limit of the normal range;

[0170]

[0171] In the formula, L i is a binary logic variable used to judge whether the i-th state meets the condition. When the maximum frequency deviation of the system in the i-th state exceeds the lower limit, L i = 1; otherwise L i = 0;

[0172]

[0173] In the formula, Δf m is the steady-state frequency deviation of the system; R vi is a binary logic variable used to judge whether the steady-state frequency deviation of the i-th state meets the condition. If the deviation is within the allowable range, R vi = 1, otherwise R vi = 0.

[0174] The present invention proposes a comprehensive evaluation method for the operation reliability and frequency of a high-proportion wind power power system, which evaluates the operation reliability of the high-proportion wind power power system. This method considers the coupling relationship between system reliability and frequency risk, comprehensively analyzes the system parameters under various scenarios of the system, and calculates the steady-state probability of each scenario based on the frequency-sensitive reliability model and the maximum frequency deviation under each scenario. Then, by linearizing the frequency deviation curve and the unbalanced power curve of the system and combining the probabilities of each scenario, the comprehensive reliability index and the frequency deviation index of the system are calculated, so as to achieve the purpose of quantifying the system reliability level. Through the above process, the present invention comprehensively considers the coupling relationship between system reliability and frequency risk, and can comprehensively evaluate the power system reliability and frequency risk within the operation reliability time scale. Compared with the prior art, the present invention improves the accuracy of the operation reliability evaluation of the high-proportion wind power power system. Moreover, in order to solve the problem of scarce frequency regulation resources in the high-proportion wind power power system, the present invention proposes a frequency response model in which thermal power and wind power participate synergistically, which takes into account both the primary / secondary frequency modulation ramp rate limit and reserve limit of thermal power and the virtual inertia response process of wind turbines. Through this frequency response model, the frequency deterioration situation in the case of a high-proportion wind power system can be effectively improved.

[0175] To better understand the function of the present invention, a specific example is described below.

[0176] Taking the IEEE-RTS79 system as the test system; the rated capacity of the traditional units in this system is 3405 MW, and the peak load is 2850 MW. The reliability parameters and control model parameters of the units all adopt the fixed data of the test system. The wind power penetration rate interval is selected as 30 - 50%. The rated power of a single wind turbine is 1.5 MW. In this example, it is assumed that the load is the peak load, and the load regulation coefficient is D i = 75 (MW / Hz).

[0177] Figure 6 It shows the frequency fluctuation situation of the system under the same wind speed fluctuation situation and different wind power penetration rates. With the increase of the wind power penetration level, under the same wind speed fluctuation level, the system frequency fluctuation becomes larger. This shows that the frequency regulation resources of the high-proportion wind power system are more scarce and the ability to resist external power disturbances is worse.

[0178] Figure 7 It shows the influence of whether the wind turbines participate in the system frequency control or not on the system frequency deviation under the same power fluctuation condition. After the wind power participates in the system frequency control, the maximum system frequency deviation can be significantly reduced, and the possibility of frequency over-limit events can be reduced.

[0179] Figure 8It shows the impact on the calculation of the system reliability index EENS with or without considering the coupling relationship between reliability and frequency risk. When the wind power penetration rate is low, the system frequency fluctuation is not obvious, and considering the coupling relationship between reliability and frequency risk has little impact on the calculation of the reliability index. As the wind power penetration rate increases, the system frequency fluctuation intensifies. Ignoring the coupling relationship between reliability and frequency risk in the calculation of the reliability index will bring large deviations.

[0180] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than limiting the technical solutions. Those of ordinary skill in the art should understand that any modifications or equivalent replacements made to the technical solutions of the present invention without departing from the purpose and scope of the present technical solution shall be covered by the scope of the claims of the present invention.

Claims

1. A comprehensive evaluation method for the operation reliability and frequency of a high-proportion wind power power system, characterized in that, It includes the following steps: Step 1: Obtain the reliability parameters, electrical parameters of the generating units, the load data of the system, and the wind speed data; the generating units include thermal power generating units and wind power generating units; Step 2: Based on the reliability parameters, electrical parameters of the thermal power generating units, and the operating condition requirements, establish a frequency-sensitive reliability model for the thermal power generating units; Step 3: Enumerate the fault events of each order of the system to generate system scenarios; Step 4: Calculate the system parameters under each scenario, and establish a frequency response model considering the coordinated frequency regulation of thermal power generating units and wind power generating units; Step 5: Analyze the frequency dynamic response process and the unbalanced power change process of the system under each scenario, and solve the maximum frequency deviation, the frequency deviation after primary frequency regulation, the frequency deviation after secondary frequency regulation, the system primary frequency regulation time, and the system secondary frequency regulation time under each scenario; Step 6: Determine whether the frequency dynamic response process of each system scenario is fully simulated. If so, go to Step 7; if not, return to Step 4; Step 7: Based on the frequency-sensitive reliability model and the maximum frequency deviation under each scenario, correct the failure rate data of the generating units, and use the Markov equation to calculate the steady-state probability of each scenario; Step 8: Linearly process the frequency deviation curve and the unbalanced power curve of the system, and combine the probabilities of each scenario to calculate the comprehensive reliability index and the frequency deviation index of the system, and quantify the operating reliability level of the system; Among them, in step 2, the operating condition requirements include the frequency protection measures of the thermal power unit, and the frequency protection measures include the low-frequency protection threshold and the high-frequency protection threshold The frequency-sensitive reliability model is: Where U m is the statistical average value of the failure probability of the thermal power unit; is the lower bound of the frequency for the thermal power unit to maintain normal operation; is the upper bound of the frequency for the thermal power unit to maintain normal operation, and f is the actual value of the system frequency.

2. The comprehensive evaluation method for the operation reliability and frequency of a high-proportion wind power power system according to claim 1, characterized in that: In step 3, each order of system fault accidents is enumerated by the state enumeration method to generate system scenarios, and the number of system scenarios generated by the enumeration is where N all is the total number of units in the system.

3. The comprehensive evaluation method for the operation reliability and frequency of a high-proportion wind power power system according to claim 2, characterized in that: In step 4, the system parameters include the equivalent ramp rate K of the system, the spinning reserve capacity SR of the system, the equivalent inertia constant H of the system i , the equivalent governor speed regulation coefficient R i , the system frequency deviation factor B i , and the unbalanced power ΔP of the system i ; Where: where Δf is the frequency deviation of the system; N all is the total number of units in the system; K i is the ramp rate of the i-th unit; state i is the state of the i-th unit, 1 represents the normal operating state of the unit, and 0 represents the unit in a fault state; P load is the load power level, which is used as the reference value for the per-unit system power; where P i is the actual output of the i-th unit, and P i,min is the minimum allowable output of the i-th unit; where, H i is the inertia coefficient of the i-th unit; Cap i is the rated capacity of the i-th unit; Wherein, is the speed regulation coefficient of the i-th unit; f rate is the rated frequency of the system; B i = 1 / R i + D i ; where D i is the system load regulation coefficient; ΔP i = ΔP Wind + ΔP G ; where, ΔP Wind is the system power fluctuation caused by wind speed fluctuation; ΔP G is the system power fluctuation caused by unit failure.

4. The comprehensive evaluation method for the operation reliability and frequency of a high-proportion wind power power system according to claim 3, characterized in that: The wind power generating unit is a doubly-fed wind turbine; in Step 4, in the frequency response model, the virtual inertia control operation mode is adopted for the wind power generating unit, and the rotor kinetic energy is lifted to participate in the system frequency response process, and the following relationship is satisfied: where f kopt is the virtual inertia factor, ω r0 is the sampled value of the rotor speed of the doubly-fed wind turbine before the disturbance, k vir is the virtual inertia coefficient of the doubly-fed wind turbine, Δf is the system frequency deviation, and p is the number of pole pairs of the doubly-fed wind turbine; Where, ΔP WT is the active power change of the doubly-fed wind turbine, ω s is the synchronous rotor speed of the power grid, H eq is the equivalent inertia time constant of the doubly-fed wind turbine, f s is the system frequency; where Δf kopt is the power adjustment factor, K opt is the maximum power tracking coefficient, ω g is the rotor speed, ω max is the speed when entering the constant power region, ω min is the starting speed of the rotor; K OPPT = K opt (f kopt - Δf kopt ); Where K OPPT is the power tracking coefficient of the fan under the virtual inertia control mode; In the formula, P OPPT is the output power of the wind turbine in the virtual inertia control mode, ω0 is the rotor speed when entering the maximum power region, ω1 is the rotor speed when entering the constant speed region, and P max is the maximum output power of the doubly-fed wind turbine.

5. The comprehensive evaluation method for the operation reliability and frequency of a high - proportion wind - power power system according to claim 4, wherein: In step 5, the maximum frequency deviation Δf of the system max is calculated by the following formula: Δf max = -max{abs(Δf(t k ))}; where Δf(t k ) represents the sampled value of the system frequency deviation at time t k ; The calculation formula for the frequency deviation Δf1 after primary frequency regulation is: The calculation formula for the frequency deviation Δf2 after secondary frequency regulation is:

6. The comprehensive evaluation method for the operation reliability and frequency of a high - proportion wind - power power system according to claim 5, wherein: In Step 5, the calculation process of the system primary frequency regulation time t1 and the system secondary frequency regulation time t2 includes: Set the initial sampling time t = 0, the iteration step size is Δt, and the convergence accuracy is ε; calculate the primary frequency regulation steady-state deviation Δf1 and the secondary frequency regulation steady-state deviation Δf2 of the system in this state; For the k-th iteration loop, the sampling time is t k = t k-1 + Δt; determine whether it satisfies the convergence condition |Δf(t k ) - Δf(t k-1 )| ≤ ε, |Δf(t k ) - Δf1| ≤ ε; if the condition is not satisfied, enter the (k + 1)-th iteration loop; if satisfied, output the corresponding sampling time t k as the primary frequency regulation time t1 of the system: Meanwhile, for the sampling time t of the k-th iteration loop k = t k-1 + Δt; It also determines whether it satisfies the convergence condition |Δf(t k ) - Δf(t k-1 )| ≤ ε, |Δf(t k ) - Δf2| ≤ ε; If the condition is not satisfied, it enters the (k + 1)-th iteration loop; If it is satisfied, the corresponding sampling time t k is output as the secondary frequency regulation time t2 of the system:

7. The comprehensive evaluation method for the operation reliability and frequency of a high - proportion wind - power power system according to claim 6, wherein: Step 7 includes: S71. Based on the frequency-sensitive reliability model and the maximum frequency deviation under each scenario, correct the fault probability parameters of the generating units to make Where U new is the corrected probability of unit failure; λ new is the corrected failure rate of the unit; μ is the average repair rate of the unit; S72. Based on the corrected generator failure rate and the Markov process, calculate the steady-state probability of each scenario; Among them, S72 includes: S721. Construct a state space diagram according to the component state transition; S722. Construct a state transition matrix T according to the state space diagram; S723. Calculate the probability vector P of each state according to the Markov equation to obtain the steady-state probability of each scenario: P(T - I) = 0; In the formula, I is the identity matrix.

8. The comprehensive evaluation method for the operation reliability and frequency of a high - proportion wind - power power system according to claim 7, wherein: In Step 8, the system comprehensive reliability index includes the expected energy not supplied EENS of the system and the expected indirect energy not supplied EIENS of the system. The expected energy not supplied EENS is used to describe the power loss risk; the expected indirect energy not supplied EIENS is used to describe the energy not supplied during the frequency regulation process; where: where N is the number of system states to be considered, Pr i is the probability of system state i, and t3 is the duration of load shedding in system state i; IENS i,1 = ΔP i × t1; where IENS i,1 is the expected indirect power shortage during the primary frequency regulation process, and IENS i,2 is the expected indirect power shortage during the secondary frequency regulation process.

9. The method for comprehensively evaluating the operation reliability and frequency of a high - proportion wind - power power system according to claim 8, wherein: In S8, the frequency deviation indicators include the expected number of times of system low-frequency overlimit ENUF, the expected system low-frequency duration EUFD, the probability Pr of the system having a low-frequency event LFE and the probability Pr of the system having a low-frequency event and recovering RLFE ; Where: where NUF i is the number of low-frequency over-limit occurrences in the i-th state. If there is no frequency over-limit in this state, then NUF i = 0; if the system frequency finally returns to the normal range after primary frequency regulation or secondary frequency regulation, then NUF i = 1; otherwise NUF i = 2; where UFD i is the duration that the frequency of the i-th state is lower than the lower limit of the normal range; where L i is a binary logic variable used to determine whether state i meets the condition. When the maximum system frequency deviation of state i exceeds the lower limit, L i = 1; otherwise L i = 0; where Δf m is the steady-state frequency deviation of the system; R vi is a binary logic variable used to determine whether the steady-state frequency deviation of state i meets the condition. If the deviation is within the allowable range, R vi = 1; otherwise, R vi = 0.

Images