Running reliability evaluation and risk reduction method for asynchronous interconnected power grid based on flexible high-voltage direct-current power transmission system

Abstract

The invention discloses an operation reliability evaluation and risk reduction method for an asynchronous interconnected power grid based on a flexible high-voltage direct-current power transmission system. The operation reliability evaluation and risk reduction method mainly comprises the following steps: 1) reliability modeling; and 2) state selection and connectivity analysis. And 3) fault analysis. And 4) remedial measures. And 5) alternating current feasibility recovery. And 6) a criterion is converged. According to the method, the problem of mismatching between the time-varying reliability of the unrepairable element and the non-time-varying unavailability model of the repairable element is solved, so that state merging of different elements can be realized to simplify the reliability evaluation process; meanwhile, a power generation and transmission system state evaluation model considering the rescheduling process is provided, and the accuracy of operation reliability evaluation is improved; finally, an expected direct current transmission power (EDCTP) index and a direct current port outage probability (DCTOP) index are provided, and the indexes can better quantify the reliability of the flexible direct current system.

Description

Technical Field

[0001] The present invention specifically relates to an operation reliability assessment and risk reduction method for an asynchronous interconnected power grid based on a flexible high-voltage direct current transmission system. Background Art

[0002] Driven by energy transformation, flexible HVDC transmission technology has developed rapidly in the past few decades. Due to its flexibility in power flow regulation and voltage control, flexible HVDC systems play a vital role in renewable energy integration and asynchronous grid interconnection. However, the operating status of a large number of electronic devices has an important impact on the operating mode of the DC system, and due to the coupling between the AC system and the DC system, the operating status of the electronic equipment will further affect the asynchronous AC grid (AGs / FDCS) system interconnected by the flexible HVDC transmission system. Reliability assessment is an effective way to quantify the risk of AGs / FDCS systems and can provide a reference for power system operators to formulate dispatch strategies.

[0003] For the reliability assessment of AGs / FDCS systems, the first step is to build a reliability model for the equipment. Based on the series-parallel combination method, a multi-state reliability model of the voltage source converter (VSC) can be established. Different from the traditional VSC, the modular multilevel converter (MMC) adopts a sub-module cascade structure composed of power modules, and its state transition can be represented by a discrete state continuous time Markov chain.

[0004] At present, the multi-state equivalent model is widely used in point-to-point high-voltage DC systems with VSC access points. In addition, in the reliability assessment of the power generation and transmission system of the AC / DC hybrid power grid, the model can improve the efficiency of state sampling. However, for DC systems with a mesh topology, this method will lead to a significant increase in dimensions, thereby increasing the difficulty of modeling. The operating state of the DC system can be determined by the state of the converter station and the DC link, which is referred to as the DC unit model in this paper. In existing studies, in order to reduce the number of enumerated states, only first-order and second-order DC unit faults are usually considered.

[0005] In flexible HVDC systems, the failure probability of non-repairable components is time-varying, which means that DC units also have time-varying failure probabilities, which needs to be considered in reliability modeling to improve the assessment accuracy. Unreliability functions have been widely used to describe the time-varying reliability of non-repairable components. However, this model cannot directly consider the operating period because it involves cumulative probability distribution. In fact, when conducting operational risk assessment, a time period needs to be specified to simulate system operation and calculate the corresponding risks. Moreover, non-repairable components use a reliability model with time-varying failure probability, while other repairable components use an availability model with non-time-varying average failure probability. The inconsistency of the two reliability models makes it difficult to merge the states of flexible DC systems. In addition, previous studies have ignored the impact of the converter station topology of flexible HVDC systems on the operating mode.

[0006] In the existing operational reliability assessment considering the redispatch process, the assessment scope is limited to the power generation system and does not involve the power transmission system. There are the following challenges to achieve this goal: ① It is not possible to use the same method as in the power generation system reliability assessment to mitigate the expected indirect energy shortage (EIENS). In this regard, this paper proposes a state assessment model for the power transmission system considering the redispatch process, and the EIENS index is integrated into the model, but at the same time it will increase the nonlinearity of the model. ② Due to the strong coupling of active and reactive power between the flexible DC converter station and the AC power grid, in order to ensure the accuracy of the state assessment results, it is necessary to consider the configuration of voltage and reactive power at the same time in the modeling process considering the redispatch process, but the complexity and computational burden of the optimization problem modeling will inevitably increase. ③ How to quantify the reliability of the DC system embedded in the AC power grid and explore the remedial measures of the flexible high-voltage DC system in emergency conditions to mitigate the operational risks of the system still need further discussion. Summary of the invention

[0007] The purpose of the present invention is to solve the problems existing in the prior art.

[0008] The technical solution adopted to achieve the purpose of the present invention is as follows: an operation reliability assessment and risk reduction method of an asynchronous interconnected power grid based on a flexible high-voltage direct current transmission system, mainly comprising the following steps:

[0009] 1) Reliability modeling.

[0010] To express the unavailability model of the DC unit, the historical service time, subsequent operation time, redundancy and converter station topology of the redundant system are predefined.

[0011] 1.1) Reliability model of irreparable systems

[0012] from Figure 3It can be seen that the DC unit contains a large number of electronic power devices, for example, the upper bridge arm (ArmU) and the lower bridge arm (ArmL) in the converter valve (CV) are formed by the cascade of IGBT-based half-bridge submodules (I-HBSM); the hybrid DC circuit breaker (HDCCB) contains an ultra-fast disconnect (UFD) submodule, an IGBT-based full-bridge submodule (I-FBSM) (in the main circuit (MC) of the load commutation switch (LCS)) and a diode-based full-bridge submodule (D-FBSM) (in the main circuit breaker (MB)). These irreparable submodule systems usually adopt redundant designs, and their reliability functions can be expressed by the k-out-of-n model. Take one bridge arm of the converter valve as an example:

[0013]

[0014] Among them, t represents the historical service time; represents the failure rate of the submodule in the converter valve; η represents the redundancy. It should be noted that due to the high reliability of the arm inductance (AI), it is considered to be completely reliable. Replacing the failure rate and redundancy in equation (1) can obtain the reliability function of other irreparable systems.

[0015] 1.2) Unavailability model of irreparable systems

[0016] According to the definition of reliability function and conditional probability, the probability of an irreparable failure of a component within time t after T years of normal operation is expressed as:

[0017]

[0018] Where f(τ) is the failure probability density function; Q(T) is the unreliability function. The relationship between f(τ), R(T) and Q(T) satisfies the following equation:

[0019]

[0020] The mathematical expectation of the duration of unavailability depends on the time x when the fault occurs and the probability density of the component failing at time x after T years of normal operation:

[0021]

[0022] Where UD is the unavailability duration in t. Dividing equation (4) by t, we get the unavailability rate model due to unrepairable failures:

[0023]

[0024] 2) State selection and connectivity analysis.

[0025] Firstly, the continuous and discrete variables in the AGs / FDCS model are sampled by non-sequential Monte Carlo simulation, where the continuous variables include system load and wind power output, and the discrete variables include the states of traditional generators, AC branches, wind turbines, and DC units. Then, the connectivity of the entire power grid is analyzed, and the impact of the interaction between the component state and the converter station operation mode is considered.

[0026] 2.1) Operation modes of flexible HVDC systems under different converter station topologies

[0027] Figure 2 A point-to-point HVDC system with a symmetrical unipolar configuration is demonstrated, where the positive and negative poles share the same converter station, and the positive and negative DC voltages have the same amplitude but opposite directions. Under this converter station configuration, damage to the main components (referring to the converter valve unit and the DC link unit) will disrupt the power transmission of the entire channel. Therefore, the flexible HVDC system with a symmetrical unipolar topology has two operating modes: outage or bipolar operation.

[0028] For the bipolar topology, the positive and negative poles with opposite voltage amplitudes share the same grounding pole and are connected to two independently operated converters. During normal operation, the positive and negative poles operate symmetrically, and when the main component of one pole fails, the remaining normal pole can operate in the form of an asymmetrical unipolar pole, which means that the system will only lose half of its transmission capacity. Therefore, there are three operating modes for the flexible high-voltage direct current system using the bipolar topology: outage, unipolar operation or bipolar operation.

[0029] 2.2) DC unit unavailability model considering converter station topology

[0030] from Figure 3 It can be seen that a converter unit (CU) contains an AC circuit breaker (ACCB), a converter transformer (CT), a converter valve (CV), a high-voltage DC circuit breaker (HDCCB), and a smoothing inductor (SI). In addition to the irreparable system, the converter valve also includes a pair of parallel control and protection systems (CPS) and cooling systems (CS). The DC link unit (DCLU) includes a DC line (DCL) and DC circuit breakers on the two ports of the branch. Based on the basic concepts of series and parallel connection, the availability of the redundant unit can be expressed as:

[0031]

[0032] Among them, U and A represent the unavailability rate and availability rate respectively. The unavailability rate of the DC unit is obtained by further state merging.

[0033]

[0034] As mentioned above, a DC system consisting of symmetrical monopole cannot use only one pole to transmit power, so the two poles sharing the same converter can be regarded as a series unit (the DC line capacity is twice the original and the availability is the square of the original). According to the above processing, the reliability modeling process takes into account the influence of the converter station topology and simplifies the state sampling process.

[0035] 2.3) Modeling of the redispatching process in the reliability assessment of power generation and transmission system operation

[0036] The operational reliability in this paper refers to the potential risk of the power system in the subsequent operation time t after T years of operation. In order to describe the time-related uncertainty sources in operation (such as system load and wind power output), the time scale of state assessment can be divided into several equal time periods. Figure 4 As shown, the length of each time window TW is determined by the changing step size of the uncertainty source.

[0037] Assume that in time period TW τ At the beginning, the power imbalance ΔP τ is a positive value. In traditional reliability assessment, the active power of traditional generators steps in adjacent intervals. From an operational perspective, this change is time-dependent and limited by the ramp rate. When considering the redispatch process, P G (t) is a monotonically increasing curve related to the spinning reserve and ramp rate of traditional generators, and is also related to other regulation measures in AGs / FDCS, such as rapid wind power curtailment and DC modulation.

[0038] In view of the reliability of system operation, load reduction should be minimized during the redispatch process, and the system should be able to recover quickly to avoid power shortage during the redispatch process. The power shortage during the redispatch process can be quantified by the indirect energy shortage (IENS) in the reliability assessment of traditional power generation systems. In order to extend this indicator to the operational reliability assessment of power generation and transmission systems with grid constraints, the present invention makes the following two assumptions:

[0039] ① The power modulation of the converter can be considered instantaneous, which means that the exchange power between the AC grid is not limited by the transmission capacity of the DC system. Therefore, the indirect energy shortage (IENS) can be traced back to the ramping behavior of the conventional generator.

[0040] ② The active output of the generator changes linearly within the adjustable range, so for the generator g, its indirect power is insufficient iens g It is equal to the area of ​​the triangle formed by the parallel lines of the coordinate axes and the generator output equation.

[0041] Based on the above assumptions, the whole system The operating reliability index IENS during the period can be expressed as the IENS of each generator put into operation. g sum.

[0042]

[0043] Among them, r G,g is the ramp rate of generator g; is the active power change of generator g in the interval [τ-1,τ]; Ω is a set of variables, whose subscripts 't' and 'G' refer to the subsequent operation time and the traditional generator respectively; τ and g represent the time in the subsequent operation time and the sequence number of the traditional generator respectively. The quantitative model proposed in the present invention can be further used to realize the state assessment model of risk reduction in the redispatch process.

[0044] 3) Fault analysis. After the power flow calculation, check the operating constraints (18), (20), (23), (24) and (25). If the power flow calculation result exceeds the limit, go to step 4) and use the linearized optimal power flow model to perform fault state assessment; otherwise, go to step 6) to check the convergence of the reliability assessment.

[0045] 4) Remedial measures. Linearized optimal power flow is used to form remedial measures with the lowest risk and total operating cost. The optimal power flow is iteratively solved based on the latest state variables until the approximate error meets the convergence criterion.

[0046] 4.1) State assessment model considering redispatching process and coordination management measures

[0047] To ensure the accuracy of the operational reliability evaluation results, AC-OPF is often used to model the state evaluation. The operational reliability evaluation time scale considered in this invention is one day. Since it is necessary to consider the system load and wind power conditions every hour, T W is set to one hour. Set operating cost as the optimization target and convert risk into economic benefit.

[0048]

[0049] Where RC and OC are risk cost and operating cost respectively; p is the penalty term; x is the set of state variables; V / θ is the union of voltage amplitude / voltage phase angle, subscripts 'AC', 'C' and 'DC' refer to the bus, converter station and DC system in the AC system respectively; P and Q represent the union of active power and reactive power respectively; subscripts 'D' and 'W' refer to load and wind turbine respectively.

[0050] The penalty factor is added to convert the indirect power shortage into risk cost, and the operation risk in the AGs / FDCS redispatch process is reduced by coordinating multiple adjustment measures in the state assessment model. In addition, in order to reasonably utilize wind energy, wind energy unutilized (WENU) is also incorporated into RC.

[0051]

[0052] Where C is the penalty factor. The superscript 'N' refers to the rated value of the variable. The operating cost of the system is modeled using a quadratic function.

[0053]

[0054] Among them, a g 、b g and c g is the operating cost coefficient of generator g. The constraints of the optimization model are shown below. For simplicity, the superscript ‘τ’ is omitted for the constraints that are not related to timing.

[0055] ① Power balance constraints of AC and DC buses

[0056]

[0057]

[0058]

[0059] Among them, P S ,c and Q S , c are the active power and reactive power injected from the DC bus of converter station c to the point of common coupling (PCC) node; Pij and Qij in formula (13) are calculated by the branch power flow model (BFM); Gij and Bij are the real and imaginary parts of the elements in the i-th row and j-th column of the network admittance matrix; P CDC , i is the active power injected into the DC system by converter station c.

[0060] ② Power balance constraints of converter stations

[0061] like Figure 5 As shown in the figure, the steady-state model of the converter station can be equivalent to the DC side bus C and the converter station loss P C loss. The power balance constraints of converter station c include:

[0062]

[0063] Where c is the number of the converter station; the branch powers P and Q are calculated by the branch power flow model in equation (13), and their directions are given by Figure 5The subscripts marked in are defined. The converter station loss is approximated by the quadratic function of the converter valve phase current, where ac, bc and cc are loss coefficients; I C , c represents the phase current.

[0064] ③ Voltage constraints of the main converter station

[0065] Vi=1;i∈Ω DC (Converter station using constant DC voltage control) (16)

[0066] ④ Power constraints of node loads

[0067] The node load d is limited by the following constraints:

[0068]

[0069] in, It represents the power factor angle at the node d. The present invention assumes that the power factor of the node load is constant.

[0070] ⑤ Voltage amplitude constraint

[0071]

[0072] The superscripts 'max' and 'min' refer to the upper and lower bounds of the corresponding variables, respectively.

[0073] ⑥ Voltage phase angle constraints at the balance node

[0074] θi=0;i∈Ω AC (Balance Node) (19)

[0075] ⑦ Generator constraints

[0076]

[0077]

[0078]

[0079] Among them, formula (21) is used to ensure the reactive power balance between generators, and the penalty term and The objective function is introduced to ensure that the solution of the optimal power flow has sufficient voltage and reactive power reserve margin.

[0080] ⑧Capacity constraints of AC and DC lines

[0081]

[0082] Among them, Φ AC,DC is the union of the branches in AGs / FDCS; Sij Refers to the apparent power from bus i to bus j (branch ij).

[0083] ⑨Converter station capacity constraints

[0084]

[0085] Among them, S c is the apparent power of converter c.

[0086] Operating constraints of conventional generators

[0087]

[0088] ⑩Constraints on converter station operation

[0089] The flexible high voltage direct current system of the present invention adopts a master-slave control mode. In order to avoid excessive modulation of the converter station within a time period, a converter station operation constraint (26) is introduced.

[0090]

[0091] Among them, the penalty term ε is introduced in the objective function + and ε - To limit the adjustment range of the converter station; the subscripts 'APC', 'RPC' and 'VC' represent converters using fixed active power control, fixed reactive power control and fixed AC voltage control, respectively.

[0092] 4.2) Linear approximation of the state assessment model

[0093] In order to improve the efficiency of reliability assessment, the present invention adopts a variety of linearization methods. Due to space limitations, the relationship between linearization methods and nonlinear models is briefly introduced: ① The constrained cost variable method is used to linearly approximate the quadratic scalar in the objective function, including operating costs and EIENS. ② The continuous linear approximation (SLA) method is used to linearly approximate the branch power flow equation and the converter loss equation, and the linearization error convergence standard of the branch power is set to 0.01pu. ③ The line capacity and converter station capacity are second-order cone constraints, which are approximated by a polygon-based piecewise linearization method.

[0094] 5) AC feasibility recovery. If the linearized optimal power flow converges, the power flow calculation is verified to ensure that the solution to the optimal power flow strictly satisfies the power flow equation. If the solution to the power flow calculation satisfies the constraints, the iterative linearized optimal power flow is terminated and the process goes to step 6). Otherwise, the feasible region of the cross-line constraint is tightened by the scalar α, and then the linearized optimal power flow is re-solved using the latest solution and the modified constraints.

[0095] 6) Convergence criterion. Expected short supply (EENS) is the slowest to converge among various indicators, so its variance is used as the convergence criterion for reliability evaluation. When the variance of EENS is less than 0.05, the iteration is terminated and the reliability index is calculated, otherwise, return to step 2) for the next state evaluation.

[0096] Finally, EENS and expected wind energy unutilized (EWENU) are introduced for reliability assessment, and some indicators related to the redispatch process and DC system are redefined.

[0097] 6.1) Expected Indirect Energy Deficit (EIENS)

[0098] According to the definition of indirect electric energy shortage (IENS), EIENS is expressed as a risk-based reliability indicator to quantify the average energy shortage of AGs / FDCS during the redispatch process.

[0099]

[0100] Among them, Ω S represents the fault state space; e is the index of the fault event; N S The present invention adopts non-sequential Monte Carlo simulation (NMCS) for state sampling.

[0101] 6.2) Expected DC Transfer Power (EDCTP)

[0102] The cross-sectional transmission power between AC and DC systems can reflect the power transmission contribution of DC system in AGs / FDCS. Based on this, the indicator EDCTP is defined to quantify the reliability of flexible HVDC system.

[0103]

[0104] Among them, Ω TI is the set of transmission sections; Ω C,l is the set of converter stations associated with transmission section l (TIl); It refers to the active power injected from converter station c to the common coupling node under the accident e at time τ. Since the converter station has a bidirectional regulation characteristic, the absolute value of EDCTP is taken.

[0105] 6.3) DC port outage probability (DCTOP)

[0106] The number of DC ports in operation can reflect the integrity of the DC system. Therefore, a probability-based indicator DCTOP is defined as shown in Equation (29).

[0107]

[0108] Among them, n-DCTOP represents the probability that n ports in the DC system stop operating; is the number of ports that stop operating under accident e; N DCT is the total number of DC ports; I(·) represents a logic function, if the expression in the brackets is true, I=1, otherwise I=0.

[0109] The technical effect of the present invention is unquestionable. This patent proposes an analytical model for unavailability caused by irreparable faults, which characterizes the time-varying characteristics of component unavailability by considering the historical service time and subsequent operation time of the component. In addition, the model solves the mismatch between the time-varying reliability of irreparable components and the non-time-varying unavailability model of repairable components, thereby enabling the merging of the states of different components to simplify the reliability assessment process. This patent also proposes a state assessment model for the power generation and transmission system considering the rescheduling process, which extends the indirect expected energy shortage (EIENS) caused by the rescheduling process to the power generation and transmission system, reduces the operating risk of the system by optimizing the control of the system's dispatchable resources, and can consider grid constraints in the state assessment model and assess the risks caused by line failures, thereby improving the accuracy of the operational reliability assessment. Finally, this patent proposes the expected DC transfer power (EDCTP) and DC port outage probability (DCTOP) indicators, which can better quantify the reliability of the flexible DC system. BRIEF DESCRIPTION OF THE DRAWINGS

[0110] Figure 1 It is a flow chart of an operation reliability assessment method of an asynchronous interconnected power grid based on a flexible high-voltage direct current transmission system according to the present invention;

[0111] Figure 2 The present invention provides a point-to-point high voltage direct current system under different converter station topology structures;

[0112] Figure 3 is a reliability block diagram of the DC unit of the present invention;

[0113] Figure 4 is the active power curve of the conventional generator of the present invention under different state evaluation models;

[0114] Figure 5 It is a schematic diagram of the steady-state equivalent model of the MMC type converter station of the present invention;

[0115] Figure 6 is the port outage probability (DCTOP) of the DC system under different time conditions in the present invention DETAILED DESCRIPTION

[0116] The present invention is further described below in conjunction with the embodiments, but it should not be understood that the above subject matter of the present invention is limited to the following embodiments. Without departing from the above technical ideas of the present invention, various substitutions and changes are made according to the common technical knowledge and customary means in the art, which should all be included in the protection scope of the present invention.

[0117] Embodiment 1:

[0118] See also Figure 6 The DC unit model established by the present invention can describe the time-varying characteristics of the component failure probability, and the reliability of the flexible high-voltage DC system is quantified based on the model and the proposed DCTOP index. The DCTOP time-varying characteristics in the figure show that the reliability of the flexible DC system using a bipolar topology is higher than that of a symmetrical unipolar topology, and the reliability of the system under the symmetrical unipolar configuration shows a more obvious vulnerability as the service time and subsequent operation increase.

Claims

1. A method for evaluating the operational reliability and reducing the risk of an asynchronous interconnected power grid based on a flexible high-voltage direct current transmission system is characterized by mainly comprising the following steps: 1) Reliability modeling is performed. To express the unavailability model of the DC unit, the historical service time, subsequent operation time, redundancy and converter station topology of the redundant system are predefined. 1.1) Reliability model of irreparable systems As can be seen from Figure 3, the DC unit contains a large number of electronic power devices. These irreparable submodule systems usually adopt redundant design, and their reliability function can be expressed by the k-out-of-n model. Take one arm of the converter valve as an example: in, t represents the historical service time; represents the failure rate of the submodule in the converter valve; η represents the redundancy. Replacing the failure rate and redundancy in equation (1) can obtain the reliability function of other irreparable systems. 1.2) Unavailability model of irreparable systems According to the definition of reliability function and conditional probability, the probability of an irreparable failure of a component within time t after T years of normal operation is expressed as: Where f(τ) is the failure probability density function; Q(T) is the unreliability function. The relationship between f(τ), R(T) and Q(T) satisfies the following equation: The mathematical expectation of the duration of unavailability depends on the time x when the fault occurs and the probability density of the component failing at time x after T years of normal operation: Where UD is the unavailability duration in t. Dividing equation (4) by t, we get the unavailability rate model due to unrepairable failures: 2) State selection and connectivity analysis. Firstly, the continuous and discrete variables in the AGs / FDCS model are sampled by non-sequential Monte Carlo simulation, where the continuous variables include system load and wind power output, and the discrete variables include the states of traditional generators, AC branches, wind turbines, and DC units. Then, the connectivity of the entire power grid is analyzed, and the impact of the interaction between the component state and the converter station operation mode is considered. 2.1) Operation modes of flexible HVDC systems under different converter station topologies Figure 2 shows a point-to-point HVDC system with a symmetrical unipolar configuration, where the positive and negative poles share the same converter station, and the positive and negative DC voltages have the same amplitude but opposite directions. Under this converter station configuration, damage to the main components (referring to the converter valve unit and the DC link unit) will disrupt the power transmission of the entire channel. Therefore, the flexible HVDC system with a symmetrical unipolar topology has two operating modes: outage or bipolar operation. For the bipolar topology, the positive and negative poles with opposite voltage amplitudes share the same grounding pole and are connected to two independently operated converters. During normal operation, the positive and negative poles operate symmetrically, and when the main component of one pole fails, the remaining normal pole can operate in the form of an asymmetrical unipolar pole, which means that the system will only lose half of its transmission capacity. Therefore, there are three operating modes for the flexible high-voltage direct current system using the bipolar topology: outage, unipolar operation or bipolar operation. 2.2) DC unit unavailability model considering converter station topology Based on the structure diagram of the DC unit in Figure 3 and the basic concepts of series and parallel connection, the availability of the redundant unit can be expressed as: in, U and A represent the unavailability rate and availability rate respectively. The unavailability rate of the DC unit is obtained by further state merging. As mentioned above, a DC system consisting of symmetrical monopole cannot use only one pole to transmit power, so the two poles sharing the same converter can be regarded as a series unit (the DC line capacity is twice the original and the availability is the square of the original). According to the above processing, the reliability modeling process takes into account the influence of the converter station topology and simplifies the state sampling process. 2.3) Modeling of the redispatching process in the reliability assessment of power generation and transmission system operation The operational reliability in this paper refers to the potential risk of the power system in the subsequent operation time t after T years of operation. In order to describe the time-related uncertainty sources in operation (such as system load and wind power output), the time scale of state assessment can be divided into several equal time periods. As shown in Figure 4, the length of each time window TW is determined by the change step size of the uncertainty source. Assume that in time period TW τ At the beginning, the power imbalance ΔP τ is a positive value. In traditional reliability assessment, the active power of traditional generators steps in adjacent intervals. From an operational perspective, this change is time-dependent and limited by the ramp rate. When considering the redispatch process, P G (t) is a monotonically increasing curve related to the spinning reserve and ramp rate of traditional generators, and is also related to other regulation measures in AGs / FDCS, such as rapid wind power curtailment and DC modulation. In view of the reliability of system operation, load reduction should be minimized during the redispatch process, and the system should be able to recover quickly to avoid power shortage during the redispatch process. The power shortage during the redispatch process can be quantified by the indirect energy shortage (IENS) in the reliability assessment of traditional power generation systems. In order to extend this indicator to the operational reliability assessment of power generation and transmission systems with grid constraints, the present invention makes the following two assumptions: ① The power modulation of the converter can be considered instantaneous, which means that the exchange power between the AC grid is not limited by the transmission capacity of the DC system. Therefore, the indirect energy shortage (IENS) can be traced back to the ramping behavior of the conventional generator. ② The active output of the generator changes linearly within the adjustable range, so for the generator g, its indirect power is insufficient iens g It is equal to the area of ​​the triangle formed by the parallel lines of the coordinate axes and the generator output equation. Based on the above assumptions, the entire system The operating reliability index IENS during the period can be expressed as the IENS of each generator put into operation. g sum. in, r G,g is the ramp rate of generator g; is the active power change of generator g in the interval [τ-1,τ]; Ω is a set of variables, whose subscripts 't' and 'G' refer to the subsequent operation time and the traditional generator respectively; τ and g represent the time in the subsequent operation time and the sequence number of the traditional generator respectively. The quantitative model proposed in the present invention can be further used to realize the state assessment model of risk reduction in the redispatch process. 3) Fault analysis. After the power flow calculation, check the operating constraints (18), (20), (23), (24) and (25). If the power flow calculation result exceeds the limit, go to step 4) and use the linearized optimal power flow model to perform fault state assessment; Otherwise, go to step 6) to check the convergence of the reliability evaluation. 4) Remedial measures. Linearized optimal power flow is used to form remedial measures with the lowest risk and total operating cost. The optimal power flow is iteratively solved based on the latest state variables until the approximate error meets the convergence criterion. 5) AC feasibility recovery. If the linearized optimal power flow converges, the power flow calculation is verified to ensure that the solution to the optimal power flow strictly satisfies the power flow equation. If the solution to the power flow calculation satisfies the constraints, the iterative linearized optimal power flow is terminated and the process goes to step 6). Otherwise, the feasible region of the cross-line constraint is tightened by the scalar α, and then the linearized optimal power flow is re-solved using the latest solution and the modified constraints. 6) Convergence criterion. Expected short supply (EENS) is the slowest to converge among various indicators, so its variance is used as the convergence criterion for reliability evaluation. When the variance of EENS is less than 0.05, the iteration is terminated and the reliability index is calculated, otherwise, return to step 2) for the next state evaluation.

2. The method for evaluating the operation reliability and reducing the risk of an asynchronous interconnected power grid based on a flexible high-voltage direct current transmission system according to claim 1 is characterized in that: In the state assessment model considering the redispatch process and coordinated management measures, the AC optimal power flow (AC-OPF) is used to model the state assessment. The operation reliability assessment time scale considered in the present invention is one day. Since it is necessary to consider the system load and wind power conditions every hour, T W is set to one hour. Set operating cost as the optimization target and convert risk into economic benefit. Where RC and OC are risk cost and operating cost respectively; p is the penalty term; x is the set of state variables; V / θ is the union of voltage amplitude / voltage phase angle, subscripts 'AC', 'C' and 'DC' refer to the bus, converter station and DC system in the AC system respectively; P and Q represent the union of active power and reactive power respectively; subscripts 'D' and 'W' refer to load and wind turbine respectively. The penalty factor is added to convert the indirect power shortage into risk cost, and the operation risk in the AGs / FDCS redispatch process is reduced by coordinating multiple adjustment measures in the state assessment model. In addition, in order to reasonably utilize wind energy, wind energy unutilized (WENU) is also incorporated into RC. Where C is the penalty factor. The superscript 'N' refers to the rated value of the variable. The operating cost of the system is modeled using a quadratic function. Among them, a g , b g and c g is the operating cost coefficient of generator g. The constraints of the optimization model are shown below. For simplicity, the superscript ‘τ’ is omitted for the constraints that are not related to timing. ① Power balance constraints of AC and DC buses Among them, P S,c and Q S,c are the active power and reactive power injected from the DC bus of converter station c to the point of common coupling (PCC) node; P in formula (13) ij and Q ij Calculated by the branch flow model (BFM); G ij and B ij is the real and imaginary part of the element in the i-th row and j-th column of the network admittance matrix; P CDC,i It is the active power injected into the DC system by converter station c. ②Power balance constraints of converter stations As shown in Figure 5, the steady-state model of the converter station can be equivalent to the DC side bus C and the converter station loss P Closs The power balance constraints of converter station c include: Where c is the number of the converter station; the branch powers P and Q are calculated by the branch power flow model in equation (13), and their directions are defined by the subscripts marked in Figure 5. The converter station loss is approximated by the quadratic function of the converter valve phase current, where a c , b c and c c is the loss coefficient; I C,c Indicates the phase current. ③ Voltage constraints of the main converter station V i =1; i∈Ω DC (Converter station using constant DC voltage control) (16) ④Power constraints of node loads The node load d is limited by the following constraints: in, It represents the power factor angle at the node d. The present invention assumes that the power factor of the node load is constant. ⑤ Voltage amplitude constraint In i min ≤V i ≤V i max ;i∈Ω AC,C,DC (18) The superscripts 'max' and 'min' refer to the upper and lower bounds of the corresponding variables, respectively. ⑥ Voltage phase angle constraints at the balance node θ i =0; i∈Ω AC (Balance Node) (19) ⑦ Generator constraints Among them, formula (21) is used to ensure the reactive power balance between generators, and the penalty term and The objective function is introduced to ensure that the solution of the optimal power flow has sufficient voltage and reactive power reserve margin. ⑧Capacity constraints of AC and DC lines Among them, Φ AC,DC is the union of the branches in AGs / FDCS; S ij Refers to the apparent power from bus i to bus j (branch ij). ⑨Converter station capacity constraints Among them, S c is the apparent power of converter c. ⑩Operational constraints of traditional generators Operation constraints of converter stations The flexible high voltage direct current system of the present invention adopts a master-slave control mode. In order to avoid over-modulation of the converter station within a time period, a converter station operation constraint (26) is introduced. in, Introducing the penalty term ε into the objective function + and ε - To limit the adjustment range of the converter station; the subscripts 'APC', 'RPC' and 'VC' represent converters using fixed active power control, fixed reactive power control and fixed AC voltage control, respectively.

3. The method for evaluating the operation reliability and reducing the risk of an asynchronous interconnected power grid based on a flexible high-voltage direct current transmission system according to claim 1 is characterized in that: EENS and expected wind energy unutilized (EWENU) are introduced for reliability assessment, and some indicators related to the redispatch process and DC system are redefined. 1) Expected Indirect Energy Deficit (EIENS) According to the definition of indirect electric energy shortage (IENS), EIENS is expressed as a risk-based reliability indicator to quantify the average energy shortage of AGs / FDCS during the redispatch process. Among them, Ω S represents the fault state space; e is the index of the fault event; N S The present invention adopts non-sequential Monte Carlo simulation (NMCS) for state sampling. 2) Expected DC Transfer Power (EDCTP) The cross-sectional transmission power between AC and DC systems can reflect the power transmission contribution of DC system in AGs / FDCS. Based on this, the indicator EDCTP is defined to quantify the reliability of flexible HVDC system. Among them, Ω TI is the set of transmission sections; Ω C,l is the set of converter stations associated with transmission section l (TIl); It refers to the active power injected from converter station c to the common coupling node under the accident e at time τ. Since the converter station has a bidirectional regulation characteristic, the absolute value of EDCTP is taken. 3) DC port outage probability (DCTOP) The number of DC ports in operation can reflect the integrity of the DC system. Therefore, a probability-based indicator DCTOP is defined as shown in Equation (29). Among them, n-DCTOP represents the probability that n ports in the DC system stop operating; is the number of ports that stop operating under accident e; N DCT is the total number of DC ports; I(·) represents a logic function, if the expression in the brackets is true, I=1, otherwise I=0.

Images