AC power grid reliability rapid control method and device based on state similarity
By constructing a system of flow equations based on state similarity, the problem of low computational efficiency of massive states in the power system is solved, fast and accurate reliability control is achieved, and the risk response capability of the power grid is improved.
Patent Information
- Application Number
- CN202310082013.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-02-06
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2043-02-06
AI Technical Summary
With the expansion of the power system, traditional reliability control methods are inefficient when calculating massive system states, resulting in delays in risk warnings or missed reports, making it difficult to meet the needs of efficient reliability assessment, especially in the lack of efficient calculation methods in the AC OPF model. .
By constructing a system of current equations based on state similarity, decoupling and iteratively solve it, the state similarity characteristics are used to transform the nonlinear optimization problem into a system of nonlinear equations, and quickly calculate the reliability index of the AC power grid, thereby improving the computing efficiency.
It realizes rapid and accurate reliability control, can respond quickly during energy structure adjustment and new energy access, take risk suppression measures in advance, improve the grid's risk resistance, and significantly improve the computing efficiency and accuracy of traditional methods.
Smart Images

Figure CN116565872B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of power system reliability control, and in particular to a method and device for rapid control of AC power grid reliability based on state similarity. Background Art
[0002] In recent years, with the continuous expansion of the scale of power systems, new technologies such as renewable energy, electric vehicles, AC and DC have brought new uncertainties and risks to the safe and stable operation of power systems, and have put forward higher requirements for the reliability control of power systems.
[0003] There are two main types of reliability control [1] :Monte Carlo simulation (MCS) and state enumeration (SE). Both methods require the optimal power flow (OPF) calculation of the system state set, analyze the power flow distribution of the power system in each state, obtain the system load reduction caused by the fault, and then calculate the reliability index of the entire system. However, with the increasing scale of the power system, the number of system states has increased exponentially. The corresponding massive OPF calculations bring a heavy computational burden to the reliability assessment, making it difficult for traditional reliability control methods to give the latest risk level of the system in a timely manner, which is easy to cause serious problems such as late risk warning or even missed warning, which may bring huge social and economic losses.
[0004] In order to improve the efficiency of reliability control, the most commonly used method is to reduce the number of system states. Reference [2] proposed a state space reduction technique to reduce the number of samples. This technique can remove system states with no load loss in advance. Variance reduction techniques have also been used to speed up the convergence of the MCS method, including: importance sampling [3] , dual variable [4] and the cross entropy method [5] . Reference [6] proposed a state enumeration reliability evaluation algorithm based on quick sorting, which can quickly select system states in descending order of probability. Furthermore, reference [7] proposed a fault fast screening technology to search for the most serious fault state based on probability and fault capacity. In addition, reference [8] proposed a state enumeration method (IISE) based on influence increment, which can transfer part of the influence of high-order states to corresponding low-order states. However, due to the limitation of evaluation accuracy, the above methods can only reduce the number of states to a limited extent. Reliability control still needs to process a large-scale set of system states, which is difficult to meet the high efficiency requirements of practical applications.
[0005] Improving the speed of analysis and calculation of a single system state is also a way to improve the efficiency of reliability control. For a large set of system states, if the OPF calculation speed of each system state can be accelerated, the reliability control efficiency can be greatly improved. Reference [9] applies post-optimization analysis to the reliability control of power generation and transmission systems, and uses the base state to quickly derive the optimal load reduction for the remaining states. Reference
[10] proposes to use a multi-parameter planning method to establish a transmission line state dictionary set, thereby quickly calculating the massive states generated by load fluctuations and generator failures, avoiding a large number of optimization solution processes. However, the above studies are limited to the calculation of DC OPF in power systems and are difficult to adapt to AC OPF models.
[0006] References
[0007] [1] W. Li, Risk Assessment of Power Systems: Models Methods and Applications. Hoboken, NJ, USA: Wiley, 2014.
[0008] [2] C.Singh and J.Mitra, "Composite System Reliability Evaluation UsingState Space Pruning," IEEE Trans.Power Syst., vol.12, no.1, pp.471-479, Feb.1997.
[0009] [3] Q.Chen and L.Mili, "Composite Power System Vulnerability Evaluation to Cascading Failures Using Importance Sampling and Antithetic Variates," IEEETrans.Power Syst., vol.28, no.3, pp.2321-2330, Aug.2013.
[0010] [4]R.Billinton and A.Jonnavithula,“Composite System AdequacyAssessment Using Sequential Monte Carlo Simulation with Variance ReductionTechniques,”IET Proc.-Gener.Transm.Distrib.,vol.144,no.1,pp.1-6,Jan.1997.
[0011] [5]Y.Wang,V.Vittal,M.Abdi-Khorsand,and C.Singh,“ProbabilisticReliability Evaluation Including Adequacy and Dynamic Security Assessment,”IEEE Trans.Power Syst.,vol.35,no.1,pp.551-559,Jan.2020.
[0012] [6]H.Liu,Y.Sun,P.Wang,and L.Cheng,“A Novel States Selection Techniquefor Power System Reliability Evaluation,”Elect.Power Syst.Res.,vol.78,pp.1019-1027,2008.
[0013] [7]Y.Jia,P.Wang,X.Han,J.Tian,and C.Singh,“A Fast ContingencyScreening Technique for Generation System Reliability Evaluation,”IEEETrans.Power Syst.,vol.28,no.4,pp.4127-4133,Nov.2013.
[0014] [8]K.Hou,H.Jia,X.Li,X.Xu,Y.Mu,T.Jiang,and
[0015] [9] A.Safdarian, M.Fotuhi-Firuzabad, F.Aminifar, and MJGhorbany, "Composite Power System Adequacy Assessment Based on the Post OptimalAnalysis," Turk.J.Elec.Eng.Comput.Sci., vol.21, no.1, pp.90-106, Jan.2013.
[0016]
[10] P.Yong, N.Zhang, C.Kang, Q.Xia, and D.Lu, "MPLP-based Fast PowerSystem Reliability Evaluation Using Transmission Line Status Dictionary," IEEETrans.Power Syst., vol.34, no.2, pp.1630-1640, Mar.2019. Summary of the invention
[0017] The present invention provides a method and device for rapid control of AC power grid reliability based on state similarity. The present invention accelerates the calculation speed of AC OPF through the similarity between systems, and converts the nonlinear optimization problem into a nonlinear equation group based on the existing state similarity information, which greatly improves the efficiency of AC power grid reliability control, thereby meeting the high timeliness requirements for reliability calculation in the field of AC power grids. See the following description for details:
[0018] A method for rapid control of AC power grid reliability based on state similarity, the method comprising:
[0019] Construct an AC OPF model for each state in the power system state set, transform the AC OPF model into a power flow equation group based on the state similarity characteristics, decouple and iteratively solve the power flow equation group;
[0020] Verify whether the solution of the power flow equations is the optimal solution of the AC OPF model, including: feasibility verification and optimality verification;
[0021] For the system state that passes the verification, the optimal load reduction amount of the state is calculated using the obtained solution of the equation group. For the system state that fails the verification, the optimal load reduction amount of the state is solved using a nonlinear optimization algorithm.
[0022] The reliability index of the power system is calculated based on the optimal load reduction amount in each state. The risk level of the power system is set according to the reliability index. It is determined whether a safety warning is needed. The contribution of the power system state to the reliability index is counted, and measures are taken for the top 1% of the power system states.
[0023] The state similarity features include:
[0024] The branch where the power flow is limited, the first phase angle, the first voltage, the first active power reduction, the first reactive power reduction, the first active power output and the first reactive power output of the generator reaching the lower limit of the variable;
[0025] A second phase angle, a second voltage, a second active power reduction, a second reactive power reduction, a second active power output and a second reactive power output of the generator reaching the upper limit of the variables;
[0026] For two power system states, if the state similarity features of the two states are exactly the same, it indicates that there is state similarity between the two states, and the state similarity features at this time are saved in the state similarity database;
[0027] The AC OPF model is converted into a set of equations based on the state similarity features in the state similarity database.
[0028] Furthermore, the decoupling and iterative solution of the power flow equations is:
[0029] Decouple the variable into two parts X α and X β , X α Including phase angle variables and voltage variables, X β Including active power variables and reactive power variables, the power flow equations are divided into two parts, and X is updated alternately and iteratively. α and X β , until the maximum number of iterations N is reached max Or convergence accuracy tol, the final solution of the power flow equations is [X α ,X β ];
[0030] The feasibility check is used to check whether the obtained solution satisfies the inequality constraints of the AC OPF model.
[0031] Furthermore, the optimality check is to determine whether there is and So that:
[0032]
[0033]
[0034] Among them, G j (X) represents the jth constraint in the power flow balance constraint; H k (X) represents the kth constraint among the inequality constraints that touch the limit.
[0035] Wherein, the method comprises:
[0036] When the verification fails, other state similarity features in the state similarity library are used for calculation until the solution passes the verification or there are no more features to be selected in the state similarity library;
[0037] If the solutions obtained based on the existing features in the state similarity database cannot pass the verification, it proves that the power system state has a new state similarity feature. The optimal load reduction amount of the power system state needs to be solved by a nonlinear optimization algorithm, and the state similarity feature information of the power system state is added to the state similarity database.
[0038] A device for rapid control of AC power grid reliability based on state similarity, the device comprising: a processor and a memory, the memory storing program instructions, the processor calling the program instructions stored in the memory to enable the device to execute any one of the method steps described.
[0039] The beneficial effects of the technical solution provided by the present invention are:
[0040] 1. The fast and accurate reliability control achieved by the present invention is crucial for the safe and reliable operation of the system. With the adjustment of energy structure and the large-scale access of new energy to the power grid, the reliability control of the power system focuses more on online short-term assessment of risk indicators. The present invention aims to improve the efficiency of traditional reliability assessment, quickly respond to the fluctuation of renewable energy in the system, accurately calculate the short-term reliability indicators of the power system in the future, and then take effective risk mitigation measures in advance to improve the risk resistance of the power grid;
[0041] 2. The present invention can quickly calculate the reliability index of the power system and monitor the operation risk level of the power system under the rapid change of source and load, so as to more intuitively understand the system operation status and formulate corresponding improvement measures;
[0042] 3. The present invention can calculate the optimal power flow model of the AC power grid based on state similarity information, adopt an equation solving algorithm based on state similarity to replace the optimization algorithm, and then quickly analyze and obtain the optimal power flow distribution of the system. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] Figure 1 It is a flow chart of a fast control method of AC power grid reliability based on state similarity;
[0044] Figure 2 It is a structural schematic diagram of an AC power grid reliability rapid control device based on state similarity;
[0045] Figure 3 Schematic diagram for comparing the computational efficiency of various methods. DETAILED DESCRIPTION
[0046] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention are described in further detail below.
[0047] In view of the problems existing in the background technology, an embodiment of the present invention proposes an AC power grid reliability control method based on state similarity. Combined with the power system flow characteristics, an AC power flow equation group based on state similarity is established, and the reliability index of the calculation system is quickly solved to construct a reliability rapid control device suitable for the AC power system.
[0048] Example 1
[0049] A reliability control method for AC power grid based on state similarity, see Figure 1 , the method comprises the following steps:
[0050] 101: Generate a system state set according to real-time system operation data, load and renewable energy data;
[0051] The system state s is a combination of the component state z and the load and renewable energy output level p, and the expression is as follows:
[0052] s=[zp] (1)
[0053] z=[z line z gen ] (2)
[0054] p=[S d S gmin S gmax ] (3)
[0055] Among them, z line and z gen is the state of the transmission line and generator; S dis the node load; S gmin and S gmax These are the upper and lower limits of generator output. Generators here include: conventional generators and distributed power sources such as wind power and photovoltaics.
[0056] 102: Based on the system state set, an AC OPF model is constructed to analyze and calculate the optimal load reduction of the power system under fault conditions. The model is as follows:
[0057] min f LC =∑P LC (4)
[0058] The constraints are as follows:
[0059] G(X)=[U]Y bus U * -S LC -C g S g +S d =0 (5)
[0060] F min ≤|Y ft U|≤F max (6)
[0061] U min ≤U≤U max (7)
[0062] S gmin ≤S g ≤S gmax (8)
[0063] 0≤S LC ≤S d (9)
[0064] Among them, f LC is the objective function; S LC and P LC are the complex power and active power of load reduction respectively; (5) represents the power flow balance constraint, G(X) is the power flow balance constraint; U and S g are node voltage and generator output respectively; U * represents the conjugate of U; Y bus is the node admittance matrix; C g is the generator connection matrix; (6) shows the branch flow constraints; Y ft is the branch admittance matrix; F min and F max are the lower and upper limits of the branch flow respectively; (7) represents the voltage constraint; V min and V maxare the lower and upper limits of voltage; (8) and (9) are the generator output constraint and load reduction constraint.
[0065] 103: Establish a power flow equation group based on state similarity, and then quickly calculate the optimal load reduction amount of the system state;
[0066] When calculating the optimal load reduction, the optimal solutions of different system states often have the same optimization characteristics, which can be called state similarity. State similarity characteristics include: branches with power flow limits, phase angles, voltages, active power reductions, reactive power reductions, generator active outputs and reactive outputs that reach the lower limit of variables, and phase angles, voltages, active power reductions, reactive power reductions, generator active outputs and reactive outputs that reach the upper limit of variables. For two system states, their component state parameters, loads and renewable energy output levels are different. If the state similarity characteristics of the two are exactly the same, it means that there is state similarity between the two system states. This optimization feature information can be saved in the state similarity database as a state similarity feature.
[0067] Based on the state similarity information in the state similarity database, the AC OPF model of the system state can be converted into a set of equations. For example, if the branch power flow constraint (6) is limited, the corresponding inequality constraint should be converted into an equality constraint. If the variable in constraint (7), (8) or (9) is limited, the variable becomes a constant. Therefore, the power flow equation set F(X) based on state similarity is obtained:
[0068]
[0069]
[0070] in, and represents the real part and the imaginary part; V and θ are the voltage amplitude and phase angle; P LC and Q LC are active and reactive load reduction respectively; P g and Q g is the active and reactive power of the generator; n b ,n g , and n l are the number of nodes, generators and lines respectively; a l is the number of limit-breaking circuits; a1, a2, a3, a4, a5 are the number of limit-breaking variables, a x =a1+a2+a3+a4+a5.
[0071] But since (4n b +2n g -a x )>>(2n b +a l), F(X) is an underdetermined system of equations with infinite solutions. In order to solve the system of equations, the embodiment of the present invention proposes an alternating iterative solution method for the state similarity power flow equations. First, the variables are decoupled into two parts X α and X β , X α Including phase angle variables and voltage variables, X β Including active power variables and reactive power variables, the equation group (10) is split into two parts:
[0072]
[0073] Among them, G * represents the rest of the tidal current equilibrium equation; S b =S LC +C g S g .
[0074] One part is the nonlinear system of equations F α (X α ):
[0075]
[0076]
[0077] in, and are the forward and reverse limit branch flow constraints respectively; a α =a1+a2. If equation (13) is an overdetermined system of equations, the least squares method is used to solve the system of equations.
[0078] The other part is the linear equation system F β (X β ):
[0079]
[0080]
[0081] Among them, P b =P LC +C g P g ;Q b =Q LC +C g Q g ; a7 and a8 are the number of limit variables; a β =a7+a8.
[0082] Further alternate iterations to update the two solutions X α and X β :
[0083]
[0084]
[0085] in, represents the i-th iteration X β The value of represents the i-th iteration X β The updated value of ; E represents the identity matrix; represents the i-th iteration of the equation system F β X β The Jacobian matrix of .
[0086] Likewise,
[0087]
[0088]
[0089] in, represents the i-th iteration X α The value of represents the i-th iteration X α The updated value of represents the i-th iteration of the equation system F α X α The Jacobian matrix of G θ and G V represents the derivative of the tidal current equilibrium equation G with respect to V and θ; and H V f The forward branch power flow equation H f Derivatives with respect to θ and V; and Represents the reverse branch power flow equation H t The derivatives with respect to θ and V.
[0090] Formula (17) and (19) are iterated alternately until the maximum number of iterations N is reached. max or convergence accuracy tol,
[0091] i>N max (twenty one)
[0092]
[0093] The iteration ends and the solution is obtained
[0094] 104: Verify whether the obtained solution X is the optimal solution of the AC OPF model, including: feasibility verification and optimality verification.
[0095] The feasibility check is to check whether the obtained solution satisfies the inequality constraints of the AC OPF model, namely, equations (6), (7), (8), and (9).
[0096] The optimality check is to determine whether there is and So that:
[0097]
[0098]
[0099] Among them, G j (X) represents the jth constraint in the power flow balance constraint; H k (X) represents the kth constraint among the inequality constraints that touch the limit.
[0100] For optimality checking, the following conditions are used to determine whether the obtained solution meets the optimality condition (23):
[0101]
[0102] Where H(X) represents the inequality constraint and rank represents the rank of the matrix.
[0103] If the verification passes, the obtained solution is the optimal solution of the AC OPF model, and the optimal load reduction amount of the corresponding system state can be obtained through the optimal solution; otherwise, the verification fails, and the obtained solution is not the optimal solution of the AC OPF model.
[0104] Failure to pass the verification means that the state does not meet the state similarity feature, and other features in the state similarity library need to be used for calculation until the solution passes the verification or there are no more features to be selected in the state similarity library. If all solutions based on the existing features in the state similarity library cannot pass the verification, it proves that the system state has a new state similarity feature. The optimal load reduction amount of this state needs to be solved by the optimization algorithm, and the state similarity feature information of this state is added to the state similarity database.
[0105] 105: Calculate the system reliability index based on the optimal load reduction amount in each state, set the system risk level based on the reliability index, and let the operation and dispatching personnel determine whether a safety warning is needed. Count the contribution of each system state to the reliability index, and take priority measures for the system states that rank in the top 1% in terms of reliability index contribution.
[0106] If a system load state or renewable energy output state contributes significantly to the reliability index, the operation and dispatching personnel need to arrange the dispatching plan in advance and use dispatchable flexible resources to increase the power supply or reduce the power load in advance to ensure safe and reliable power supply to the system.
[0107] If the fault state of a system component contributes greatly to the reliability index, it means that the component is the weak link of the entire system. The operation and dispatching personnel should increase the inspection and monitoring of the component, replace the aging components in time, and ensure the healthy operation of the component. If the component is a generator, a new generator set can be put into production to increase the power generation reserve capacity. If the component is a line and the importance of load reduction is high, a new power supply line can be added to improve power supply redundancy. If the state of common failure of multiple system components contributes greatly to the reliability index, the operation and dispatching personnel should promptly repair the system components and take measures to increase the line transmission capacity and increase the generator output capacity to reduce the load reduction of the state.
[0108] In addition, after each reliability assessment control, the state similarity database calculated this time is saved and used for the next reliability control.
[0109] Example 2
[0110] A rapid assessment device for AC power grid reliability based on state similarity, see Figure 2 , the device includes: a processor, a memory, an input interface, an output interface, a network adapter, a power module and a bus.
[0111] The memory stores program instructions and calculation data, and the processor calls the program instructions and calculation data in the memory to execute the steps of the method for rapid assessment of AC power grid reliability based on state similarity. The program instructions stored in the memory include a state rapid analysis module, a result verification module and an index calculation module. The state rapid analysis module rapidly calculates and analyzes the optimal load reduction amount of the system state based on state similarity information. The result verification module is used to verify the feasibility and optimality of the obtained solution, so as to ensure the accuracy of the method of the present invention. The index calculation module calculates the reliability index of the system based on the optimal load reduction amount and occurrence probability of each system state. The device can communicate with the power system through a wired or wireless network, and can also be integrated into the power system.
[0112] The framework of the method proposed in the embodiment of the present invention is as follows Figure 1 The detailed description is as follows:
[0113] After inputting the system operation data, load and renewable energy data, the system state set is generated through state enumeration and Monte Carlo simulation methods to simulate each power system fault scenario.
[0114] Then, based on the system state, the power flow optimization objective function and constraints are constructed, and the optimal load reduction of the system state is quickly solved using the state similarity information. Further determine whether the result meets the feasibility and optimality. If so, the obtained solution is the optimal solution; if not, it is necessary to continue to traverse the state similarity database to find out whether there is a solution that can pass the verification. If it still does not exist, the impact of the system state is optimized and calculated, and the state-related information is stored in the state similarity database.
[0115] Finally, after analyzing all system states, the system reliability index is calculated based on the probability of all states and the load reduction. Taking the expected value of power shortage (EENS) as the reliability index, it can be calculated by the following formula:
[0116]
[0117] Where T is the system evaluation time; P(s) is the probability of state s; I(s) is the optimal load reduction amount in state s.
[0118] Example 3
[0119] The embodiment of the present invention is experimentally tested on the IEEE-RTS79 system. The system includes 24 nodes, 32 generator sets, 38 branches, and the peak load is 2850MW. R2020b platform, the computer is equipped with dual Platinum 8180CPU(ES) 28×1.8GHz and 128GB RAM.
[0120] The state similarity method is combined with the Monte Carlo method, the influence increment method and the state enumeration method. The calculation results are shown in Table 1. Figure 3 As shown:
[0121] Table 1 Reliability evaluation results of RTS-79 system
[0122]
[0123]
[0124] As shown in Table 1, the state similarity method is nearly 20 times faster than the traditional method. Only a small part of the system state is solved by the OPF optimization algorithm, and the rest of the state is quickly solved by the state similarity equation group. In addition, the error of the state similarity method is within 1%, and the accuracy of the calculation result is verified. Therefore, the rapid evaluation method based on state similarity in the embodiment of the present invention can greatly improve the efficiency of AC power grid reliability control.
[0125] Unless otherwise specified, the models of the components in the embodiments of the present invention are not limited, and any device that can perform the above functions may be used.
[0126] Those skilled in the art will appreciate that the accompanying drawing is only a schematic diagram of a preferred embodiment, and the serial numbers of the embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.
[0127] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A rapid control method for AC power grid reliability based on state similarity, characterized in that: The method comprises: Construct an AC OPF model for each state in the power system state set, transform the AC OPF model into a power flow equation group based on the state similarity characteristics, decouple and iteratively solve the power flow equation group; Verify whether the solution of the power flow equations is the optimal solution of the AC OPF model, including: feasibility verification and optimality verification; For the system state that passes the verification, the optimal load reduction amount of the state is calculated using the obtained solution of the equation group. For the system state that fails the verification, the optimal load reduction amount of the state is solved using a nonlinear optimization algorithm. Calculate the reliability index of the power system based on the optimal load reduction amount in each state, set the risk level of the power system according to the reliability index, determine whether a safety warning is needed, calculate the contribution of the power system state to the reliability index, and take measures for the top 1% of the power system states; Among them, min f LC =∑P LC The constraints are as follows: G(X)=[U]Y bus U * -S LC -C g S g +S d =0 F min ≤|Y ft U|≤F max IN min ≤U≤U max S gmin ≤S g ≤S gmax 0≤S LC ≤S d Among them, f LC is the objective function; S LC and P LC are the complex power and active power of load reduction respectively; G(X) is the power flow balance constraint; U and S g are node voltage and generator output respectively; U * represents the conjugate of U; Y bus is the node admittance matrix; C g is the generator connection matrix; Y ft is the branch admittance matrix; F min and F max are the lower and upper limits of branch power flow respectively; V min and V max are the lower and upper limits of voltage; The power flow equations F(X) based on state similarity are: in, and represents the real part and the imaginary part; V and θ are the voltage amplitude and phase angle; P LC and Q LC are active and reactive load reduction respectively; P g and Q g is the active and reactive power of the generator; n b ,n g , and n l are the number of nodes, generators and lines respectively; a l is the number of limit-breaking circuits; a1, a2, a3, a4, a5 are the number of limit-breaking variables, a x =a1+a2+a3+a4+a5.
2. The method for rapid control of AC power grid reliability based on state similarity according to claim 1 is characterized in that: The state similarity features include: The branch where the power flow is limited, the first phase angle, the first voltage, the first active power reduction, the first reactive power reduction, the first active power output and the first reactive power output of the generator reaching the lower limit of the variable; A second phase angle, a second voltage, a second active power reduction, a second reactive power reduction, a second active power output and a second reactive power output of the generator reaching the upper limit of the variables; For two power system states, if the state similarity features of the two states are exactly the same, it indicates that there is state similarity between the two states, and the state similarity features at this time are saved in the state similarity database; The AC OPF model is converted into a set of equations based on the state similarity features in the state similarity database.
3. The AC power grid reliability rapid control method based on state similarity according to claim 1 is characterized in that: The decoupling and iterative solution of the power flow equations is: Decouple the variable into two parts X α and X β , X α Including phase angle variables and voltage variables, X β Including active power variables and reactive power variables, the power flow equations are divided into two parts, and X is updated alternately and iteratively. α and X β , until the maximum number of iterations N is reached max Or convergence accuracy tol, the final solution of the power flow equations is [X α ,X β ].
4. The method for rapid control of AC power grid reliability based on state similarity according to claim 1, characterized in that: The feasibility check is used to check whether the obtained solution satisfies the inequality constraints of the AC OPF model.
5. The method for rapid control of AC power grid reliability based on state similarity according to claim 1, characterized in that: The optimality check is to determine whether there is and So that: Among them, G j (X) represents the jth constraint in the power flow balance constraint; H k (X) represents the kth constraint among the inequality constraints that touch the limit.
6. The method for rapid control of AC power grid reliability based on state similarity according to claim 1, characterized in that: The method comprises: When the verification fails, other state similarity features in the state similarity library are used for calculation until the solution passes the verification or there are no more features to be selected in the state similarity library; If the solutions obtained based on the existing features in the state similarity database cannot pass the verification, it proves that the power system state has a new state similarity feature. The optimal load reduction amount of the power system state needs to be solved by a nonlinear optimization algorithm, and the state similarity feature information of the power system state is added to the state similarity database.
7. A rapid control device for AC power grid reliability based on state similarity, characterized in that: The device comprises: a processor and a memory, wherein program instructions are stored in the memory, and the processor calls the program instructions stored in the memory to enable the device to execute any one of the methods of claims 1-6.
