Intelligent power network resilience enhancement energy configuration method and system for variable environment

Abstract

The application provides a kind of smart power network resilience enhancement energy configuration method and system for variable environment, which comprises the following steps: generating typical scenarios according to the uncertainty of wind power and photovoltaic power output;According to N-K criterion, construct power line fault model to simulate the extreme situation of multiple line faults;Considering each typical scenario and fault situation, an optimization model is established to minimize line loss;Based on the optimization model, a two-level solution structure of master problem and subproblem is built;By setting decision consistency constraints between the master problem and the subproblem, and based on the constraints, relaxation and updating are carried out;Iterative solution until the global optimal energy configuration scheme is obtained.The application can ensure long-term cost-effectiveness while flexibly responding to short-term extreme faults, thereby significantly improving the overall resilience and reliability of the power network.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of power networks, and particularly relates to a smart power network resilience enhancement energy configuration method and system for a variable environment. BACKGROUND

[0002] In modern power systems, the power distribution network, as a key link connecting the power generation side and the end users, is often the first to be affected and fail when subjected to extreme environmental impacts, such as strong winds, heavy rain, freezing, geological disasters, or network attacks. With the increase in various extreme weather and potential attack threats worldwide, the vulnerability of the power distribution network is increasingly apparent. Large-scale line failures not only cause power supply interruptions for some users, but also may trigger a chain reaction and expand the blackout range if not restored in time. In this context, how to improve the adaptability and recovery capability of the power grid in the face of a variable environment has become a key problem to be solved in the field of power system planning and operation.

[0003] For the resilience of the power distribution network, the existing technology mainly starts from the perspective of distributed energy configuration, and improves the anti-interference ability and self-healing ability of the network to external impacts by reasonably deploying renewable energy units and energy storage devices within the network. However, in practical applications, there are mainly two types of configuration methods:

[0004] Large-scale fixed unit configuration under long-term planning perspective: By planning large-capacity renewable energy generation devices or fixed energy storage systems, the power grid is provided with basic and stable energy support. However, this method mainly focuses on long-term average demand and lacks adaptability to short-term emergencies, making it difficult to respond to rapidly changing loads and fault conditions in a timely manner.

[0005] Small flexible device configuration under short-term scheduling perspective: In the event of a fault or emergency, flexible resources such as mobile energy storage and emergency power are added to meet instantaneous demand fluctuations. Although this method can respond quickly, its scale or capacity is often small, and if not combined with long-term planning, it is difficult to balance global economy and long-term reliability, leading to resource waste or inability to cope with cumulative demand.

[0006] When facing the two main uncertainties of renewable energy output fluctuation and natural disaster failure, the traditional method also has the following shortcomings:

[0007] Context-aware optimization: usually uses scenario enumeration or scenario sets to describe the random fluctuations of renewable energy sources such as wind power and photovoltaic power. Although multiple scenarios can be evaluated in long-term planning, when the number of scenarios is large, the calculation scale increases significantly, and improper coverage of scenarios can lead to results deviating from actual operating requirements.

[0008] Emergency dispatch optimization: By constructing extreme failure scenarios (such as multiple line failures at the same time), the system can ensure a certain defense capability, but it often has the defect of over-conservatism or only designs for the worst case, resulting in investment redundancy and lack of flexibility.

[0009] Therefore, the existing configuration strategy has limitations in considering long-term stability and short-term emergency flexibility, and it is difficult to fully improve the resilience of the power distribution network in a changing environment. SUMMARY

[0010] The purpose of the present application is to solve the problems in the above background art, and to provide a smart power network resilience enhancement energy configuration method and system for a changing environment, which combines situation-driven optimization and emergency response scheduling, and integrates fixed energy unit configuration and flexible energy unit deployment into a unified multi-stage optimization framework. By setting decision consistency constraints between the master problem and the sub-problem and using a collaborative solving algorithm, the short-term extreme failure can be flexibly responded to while ensuring long-term cost-effectiveness, thereby significantly improving the overall resilience and reliability of the power network.

[0011] The technical scheme adopted by the present application is: a smart power network resilience enhancement energy configuration method for a changing environment, comprising the following steps:

[0012] Generate typical scenarios according to the uncertainty of wind power and photovoltaic power output; construct a power line failure model according to the N-K criterion to simulate extreme situations of multiple line failures at the same time;

[0013] Consider each typical scenario and failure situation, and establish an optimization model with the objective of minimizing line loss; set binary decision variables to represent the configuration state of distributed energy units and the topology reconstruction of the power grid;

[0014] Based on the optimization model, build a two-level solving structure of master problem and sub-problem, take the fixed energy unit configuration and network topology decision as the master problem, and take the situation-driven optimization and emergency response scheduling as two sub-problems; the situation-driven optimization is used for long-term planning to determine the optimal configuration of fixed energy units, and the emergency response scheduling is used for short-term scheduling to determine the optimal deployment of flexible energy units;

[0015] By setting decision consistency constraints between the master problem and the sub-problem, and based on the constraints, relaxation and updating are carried out; iterative solution is carried out until the global optimal energy configuration scheme is obtained, thereby improving the resilience of the power network in an uncertain environment.

[0016] In the technical scheme, the modeling of the output uncertainty of the wind power and the photovoltaic power generation adopts a scenario generation method combining Latin hypercube sampling with scenario reduction to obtain a plurality of representative wind speed and irradiance scenarios and their occurrence probabilities, wherein a large number of initial scenarios are generated from the probability distribution of the wind speed and the solar irradiance by Latin hypercube sampling, and the initial scenario set is compressed into a representative typical scenario set by using a scenario reduction algorithm.

[0017] In the technical scheme, the line fault model is established by introducing a binary variable representing the attack state of the power line, and a constraint condition that the number of attacked lines does not exceed K is set, which is used to represent the situation that at most K power lines are simultaneously faulted under a variable environment; when the binary variable of a power line indicates that the line is attacked, the line is regarded as a fault disconnection to limit the power flow distribution under the corresponding fault situation in the optimization model.

[0018] In the technical scheme, binary decision variables are set in the energy configuration optimization model to depict the distributed energy configuration and the grid topology reconstruction decision: the first type of binary variable is used to indicate whether a fixed energy unit or a flexible energy storage device is installed at each power node, and takes the value 1 if installed, otherwise takes the value 0; the second type of binary variable is used to indicate the operating state of the power line, and takes the value 1 if the line is put into operation, and takes the value 0 if the line is disconnected, so as to introduce the installation selection of the distributed energy unit and the grid reconstruction scheme into the optimization decision.

[0019] In the technical scheme, the energy configuration optimization model is a two-stage optimization model, the first stage configures the fixed energy unit, and the second stage configures the flexible energy unit and the grid reconstruction constraint; the objective function includes the configuration cost of the fixed energy unit, the line loss cost of the grid, the configuration cost of the flexible energy unit, and the load loss cost under the extreme fault situation; the constraint conditions of the optimization model include the power flow balance constraint of the power network, the power output limit of the generator and the distributed energy unit, the line transmission capacity limit, the upper and lower limit constraint of the node voltage, the configuration number constraint of the fixed energy unit, the configuration number constraint of the flexible energy unit, and the constraint that the number of damaged lines does not exceed K under the line fault model.

[0020] In the technical scheme, the main problem is to minimize the stable energy unit configuration cost, and the constraints include the total installation number limit of the fixed energy unit at all nodes, the total configurable number limit of the wind power device, and the total installation number limit of the photovoltaic power device at each node.

[0021] In the above technical solution, the scenario-driven optimization problem aims to minimize the line loss weighted by the probability of occurrence of each wind power and photovoltaic scenario. The constraints include the power flow equation and balance conditions of the power network under normal operation, as well as the operation and capacity limitations of the fixed energy units. A decomposition-based probabilistic optimization algorithm is used to solve the problem, thereby determining the optimal configuration plan for the fixed energy units at each power node.

[0022] In the above technical solution, the emergency response scheduling problem aims to minimize the load loss caused by the attacked line failure. The constraints include the power network's flow balance constraints under fault scenarios, the operation constraints of flexible energy units, and the grid topology reconstruction constraints. The deployment plan of flexible energy units is iteratively optimized through an incremental collaborative solution algorithm.

[0023] In the above technical solution, a decision-consistency coupling mechanism is established between the master problem and the slave problem, and the master problem and the slave problem are collaboratively solved by iteratively updating the Lagrange multiplier. By adding coupling constraints to the Lagrange relaxation and introducing corresponding Lagrange multipliers, a Lagrange relaxation model containing the master problem and the slave problem is constructed. Then, the master problem and the slave problem are solved alternately and the Lagrange multiplier is updated according to the deviation of the solution after each round of iteration. This cycle is repeated until convergence, so as to achieve incremental collaborative optimization between the sub-models and obtain the global optimal solution that satisfies all constraints.

[0024] The present invention also provides an intelligent power network resilience enhancement energy configuration system for a changing environment, which is used to implement the method described in the above technical solution, including:

[0025] Uncertainty modeling module, used to perform uncertainty analysis on wind power and photovoltaic power generation output and generate corresponding typical wind speed and irradiance scenario sets;

[0026] A fault scenario generation module is used to construct a power network line fault scenario of the NK criterion and determine the combination and quantity limit of the attacked lines in the fault scenario;

[0027] An optimization model building module is used to establish a distributed energy configuration optimization model with the goal of minimizing grid line losses, and to set binary decision variables representing the installation status of distributed energy units and the grid line reconstruction status in the model;

[0028] A model decomposition module builds a two-level solution structure of master and slave problems based on the optimization model, with fixed energy unit configuration and network topology decision-making as the master problem, and scenario-driven optimization and emergency response scheduling as the two slave problems. Scenario-driven optimization is used for long-term planning to determine the optimal configuration of fixed energy units, while emergency response scheduling is used for short-term scheduling to determine the optimal deployment of flexible energy units.

[0029] The synergistic coordination module is used for setting a decision consistency constraint between the master problem and the slave problem, relaxing and updating based on the constraint, iteratively solving until a global optimal energy configuration scheme is obtained, thereby improving the resilience of the power network in an uncertain environment.

[0030] The present application has the advantages that: the present application considers long-term fixed energy configuration and short-term flexible resource deployment in a unified method process, and is particularly suitable for large-scale power grid systems; by combining renewable energy uncertainty and N-K situations of line faults, the anti-interference and recovery ability of the network in a variable environment can be more comprehensively evaluated and improved; taking line loss minimization as the target (which can be expanded or combined with other targets), the economy and reliability of power distribution are ensured, and investment and operation benefits are balanced.

[0031] Further, the present application can uniformly cover the probability distribution of wind speed and irradiance by Latin hypercube sampling, improve the sampling accuracy of random factors, and use scenario reduction algorithm to compress a large number of initial scenarios into a few representative scenarios, thereby reducing the dimension and calculation amount of the model solving process; the present application significantly reduces the solving complexity while keeping a relatively accurate approximation of the uncertainty distribution characteristics, facilitating application in large-scale systems.

[0032] Further, the present application considers the situation of multiple line faults under the N-K criterion, so that the system can better resist external shocks (natural disasters or man-made attacks); by introducing "attacked" into the binary variable, different line combination faults and their influence on power flow can be flexibly represented in the model; by limiting the number of simultaneous fault lines and evaluating their influence, the reliability and resilience can be targetedly improved in the planning and scheduling stage.

[0033] Further, the present application realizes the discretization expression of controllable elements of the distribution network by embedding the device configuration and line reconstruction decisions in the form of binary variables into the optimization model; different types of distributed energy units can be configured, and the lines can be switched or reconstructed according to the network state, which helps to adjust the operation scheme in time in the fault or high loss scenario; the introduction of binary variables makes the problem a mixed integer model, but it is compatible with existing numerical optimization or distributed algorithms, so that a feasible and efficient solution is obtained.

[0034] Further, the present application fully reflects the influence of different time scales on network planning by processing the configuration problem of fixed resources (long period) and flexible resources (short period or emergency stage) in stages; multiple costs such as investment cost (fixed / flexible energy unit), operation loss, and fault load loss are included in a unified framework to avoid local optimization leading to poor global efficiency; both normal operation (line loss, etc.) and safety and economy under extreme situations (load loss) are considered to improve overall resilience.

[0035] Further, the present application can effectively reduce excessive investment or resource mismatch in the long-term planning stage by minimizing the configuration cost of fixed energy units as the main problem target; the total installation quantity limit is set for wind power, photovoltaic, etc. to prevent blind expansion of the system and avoid unnecessary economic burden; while meeting the basic demand of renewable energy access, the investment scale and system layout are brought into a controllable range.

[0036] Further, the present application can more accurately measure the operation performance of the system under different renewable power levels by weighting the wind power and photovoltaic scenarios; the present application targets the minimum line loss with probability weighting, which helps to improve overall energy utilization efficiency and reduce power transmission loss; it is suitable for large-scale problems in multiple scenarios, more efficient in calculation, and can fully utilize the parallelism or distributed collaborative optimization method.

[0037] Further, in the scenario of multiple line faults or attacks, the present application maximizes the reduction of load loss and improves the safety of system power supply through flexible energy deployment and line reconstruction; the emergency dispatch can obtain a feasible solution in a short time and gradually adjust the configuration of flexible energy according to the current fault condition; in the case of partial system damage, the demand of key areas or key loads is prioritized, thereby improving the resilience of the distribution network in disaster or attack scenarios.

[0038] Further, the present application effectively decomposes multiple sub-problems that are closely coupled through the Lagrangian relaxation method, reducing the difficulty of directly solving large-scale mixed integer problems; the main and slave problems are solved alternately and the Lagrangian multiplier is updated to quickly narrow the solution space, ensure global consistency, and reduce the risk of the algorithm falling into local optimum; the Lagrangian relaxation or other distributed iterative algorithms are more adaptable when facing large-scale energy systems, while retaining the flexibility to handle different scenarios or faults. BRIEF DESCRIPTION OF DRAWINGS

[0039] Figure 1 The present application is a method flowchart;

[0040] Figure 2 The present application is a master-slave problem allocation diagram. DETAILED DESCRIPTION

[0041] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments to facilitate a clear understanding of the present invention, but they do not constitute a limitation to the present invention.

[0042] Example 1

[0043] like Figure 1 As shown, the present invention provides a method for energy configuration for enhancing resilience of a smart power network in a changing environment, comprising the following steps:

[0044] Generate typical scenarios based on the uncertainty of wind and photovoltaic power output; build a power line fault model according to the NK criterion to simulate the extreme scenario of simultaneous failure of multiple lines;

[0045] Considering various typical scenarios and fault conditions, an optimization model is established to minimize line losses. Binary decision variables are set to represent the configuration status of distributed energy units and the topological reconstruction of power grid lines.

[0046] Based on the optimization model, a two-level solution structure of master and slave problems is established, with fixed energy unit configuration and network topology decision-making as the master problem, and scenario-driven optimization and emergency response scheduling as the two slave problems. Scenario-driven optimization is used for long-term planning to determine the optimal configuration of fixed energy units, while emergency response scheduling is used for short-term scheduling to determine the optimal deployment of flexible energy units.

[0047] By setting decision consistency constraints between the master problem and the slave problem, and relaxing and updating based on the constraints; iteratively solving until the global optimal energy configuration solution is obtained, the resilience of the power network in uncertain environments is improved.

[0048] Specifically, fixed energy units mainly include wind turbines, photovoltaic power generation units and fixed electrochemical energy storage devices, which are used to provide stable power generation / energy storage capabilities; the corresponding flexible energy units may include movable energy storage equipment or small distributed generators that can be easily installed / uninstalled.

[0049] This embodiment specifically includes the following steps:

[0050] Step 1: Establish the wind and solar uncertainty and NK model faced in the power network.

[0051] In actual projects, predicting the active power of wind turbines and photovoltaic power generation units is challenging, as their output power is primarily affected by the wind speed and irradiance at their location. By establishing relationships between wind speed and the output active power of wind turbines, and between irradiance and the output active power of photovoltaic units, the actual active power of wind turbines and photovoltaic units can be estimated. This embodiment provides a scenario-based description of the output uncertainty of wind and photovoltaic power generation to accurately assess the operating status of the power network under different renewable energy output levels. This method uses a function sampling method based on an uncertainty variable probability model to generate large-scale scenarios.

[0052] Specifically, a large number of initial planning scenarios for sources and loads are first generated by the Latin hypercube sampling method, and then the scenarios are compressed using the backward scenario reduction method to obtain a set of typical scenarios and their corresponding scenario probabilities, thereby obtaining the maximum output power of wind energy devices and photovoltaic units under different scenarios. Based on the generation of long-term and short-term wind speed and irradiance scenarios, the sets and , the corresponding scene probability is and .

[0053] Preferably, the scene generation method adopted in this embodiment specifically includes the following steps:

[0054] In a given wind speed range, the maximum output power of the wind energy device is determined based on its power-wind speed characteristic curve; the active power output of the photovoltaic power generation unit is regarded as a function of solar irradiance: the rated power of the photovoltaic unit under different irradiances is obtained through the correspondence between the characteristics of the photovoltaic module and the irradiance value. Statistical distributions are established for wind speed and irradiance respectively (such as probability density functions obtained based on historical data or predicted data) for subsequent sampling. Latin hypercube sampling is performed on the above probability distributions of wind speed and irradiance to generate a large number of initial scene samples; in this process, in order to take into account the analysis needs of different time scales, long-term and short-term wind speed and irradiance distributions can be sampled separately to obtain two types of initial scene sets: (long term) and (Short term). Considering that the number of initial scenarios in actual large-scale power systems may be very large, directly using all scenarios will lead to a sharp increase in the amount of calculation. To this end, this embodiment uses a backward scenario reduction algorithm to compress the initial scenario set; by measuring the similarity of different scenarios in wind speed and irradiance dimensions and controlling the total landscape deviation, redundant scenarios are gradually eliminated and similar scenarios are merged during the reduction process; finally, a typical scenario set is generated. and Corresponding to long-term and short-term analysis, each scenario or All have their own probability of occurrence and The maximum output power (or corresponding power range) of the wind power device and the photovoltaic unit is given for each typical scenario, so that the randomness of the wind and light output can be more accurately characterized.

[0055] By the above method, the embodiment can significantly reduce the number of scenarios, reduce the dimension of the subsequent optimization model, and improve the solving efficiency and scalability while retaining the uncertainty characteristics of wind and photovoltaic output.

[0056] Preferably, to cope with the possible large-scale failure or attack of power lines in a variable environment, the embodiment introduces the N-K criterion in the model, and the specific method is as follows:

[0057] Line attack binary variable setting record For line Binary variable of whether attacked: when , it means that the line is in an attacked state and is considered to be a fault disconnection; if , the line is in normal operation under the current scenario. The combination of different line attacks reflects the multiple failure modes that may occur in the system under extreme environmental or malicious damage.

[0058] Let , where K is the maximum number of lines that can fail (or be attacked) at the same time; this inequality is used to describe the N-K criterion: that is, under the most unfavorable circumstances, the system still needs to have the ability to cope with at most K line failures. Through this constraint, the impact of multi-line failure on network topology and power flow distribution can be fully considered in the optimization process.

[0059] When , line does not participate in power flow transmission, which is equivalent to removing the line from the network; accordingly, the power flow equation needs to redistribute the active and reactive power balance of each node under the remaining topology structure to ensure the feasibility of system operation or minimize load loss under the fault scenario.

[0060] Through the above modeling method, the embodiment can systematically evaluate the line failure distribution of the power grid under adverse environments or external attacks in the optimization framework, and achieve higher safety margin and reliability in the planning and scheduling level combined with the N-K criterion.

[0061] Step 2: According to the theory of smart power network resilience enhancement strategy, an energy configuration model oriented to minimize power network loss is established. Based on the distributed collaborative optimization method, the energy configuration optimization model is divided into a situation-driven optimization model and an emergency response scheduling model at the model level; wherein the situation-driven optimization model is used for long-term planning to determine the optimal configuration of fixed energy units, and the emergency response scheduling model is used for short-term scheduling to determine the optimal deployment of flexible energy units;

[0062] Assume that the binary decision variable represents each power node whether to install a fixed energy unit (such as energy storage, wind turbine, photovoltaic device) or mobile flexible energy storage device. If the node is installed with the corresponding energy resource, the corresponding decision variable is 1, otherwise it is 0.

[0063] The energy configuration optimization model is a two-stage optimization model, the first stage configures fixed energy units such as wind energy, solar energy and fixed energy storage devices, etc.; the second stage configures flexible energy units and power grid reconstruction constraints such as mobile energy storage systems, etc.

[0064] The objective function of the embodiment includes: stable energy unit configuration cost , line loss cost in situation-driven optimization , flexible energy unit configuration cost , and load loss cost in emergency response scheduling . The energy configuration optimization model is as follows:

[0065] ;

[0066] Wherein, binary variables , , and respectively represent whether to configure fixed energy storage units (corresponding to superscript dg), wind turbine devices (corresponding to superscript wtg), photovoltaic power generation units (corresponding to superscript pvg) and flexible energy storage units (corresponding to superscript mps) these distributed resources at node , and the configuration is 1, otherwise it is 0; binary variable represents whether the line is attacked, and the attack is 0, otherwise it is 1; variables , , and respectively represent the rated power of different distributed resources configured at node ; variable l ij represents the current square variable in branch ij. denotes the scenario lower line square of the upper current; variable denotes the line impedance on; variable denotes the node active power demand of the load; variable denotes the scenario lower node active power of the cut-out load; variable denotes the node penalty coefficient of different load types; constant , and denote the unit power cost (yuan / kW) of different high-cost fixed distributed resources, respectively; constant denotes the unit power cost (yuan / kWh) of the flexible energy storage unit; constant and denote the unit cost coefficient (yuan / kW) of line loss and the unit cost coefficient (yuan / kWh) of load loss, respectively; and denote the probability set under different normal operation scenarios and extreme scenarios, respectively; set , , , and denote the set of configuring high-cost fixed distributed energy, line loss of power flow calculation, configuring flexible energy storage unit distributed energy, line attack and load loss of power flow calculation, respectively.

[0067] The constraint conditions are as follows:

[0068]

[0069]

[0070]

[0071] wherein, , and denote the maximum number of fixed energy storage units, wind power devices and photovoltaic power generation units, respectively. Constraints (2)-(4) represent the number limit of fixed high-cost distributed resources;

[0072]

[0073]

[0074]

[0075] where binary variable denotes the status of line ; variable denotes the power flow on line ; constant denotes the number of nodes; constant M is a large enough number to ensure the linearization of the constraints; sets and denote the input and output line sets of node , respectively; sets and denote the existing line and candidate line sets, respectively. Constraint (5) indicates that there are n-1 active lines in the system, which guarantees the connectivity of the distribution network; constraint (6) indicates the current limit on line ; and constraint (7) indicates the power flow limit on line

[0076]

[0077]

[0078] where superscript s1 indicates that all variables are changed in the long-term regular operation scenario ; variables , and denote the actual active power output of the different fixed high-cost distributed resources configured at node ; variables , and denote the actual reactive power output of the corresponding distributed resources; variables and denote the active power and reactive power output of the generator configured at node ; variables and denote the active power and reactive power on line ; variable denotes the reactive power cut off at node under scenario ; and variable denotes the line reactance. Constraints (8) and (9) represent the voltage balance constraints at nodes active and reactive power balance constraints.

[0079]

[0080]

[0081] where variables and represent the first and last voltage values on line ; constraints (10) and (11) represent the voltage balance constraints at nodes .

[0082]

[0083]

[0084] where variables and represent the upper limits of active and reactive power on line ; constraints (12) and (13) limit the active and reactive power on line and ensure network connectivity.

[0085]

[0086]

[0087] where variables and represent the upper limits of active and reactive power of the conventional generator at node 1; constraints (14) and (15) represent the active and reactive power limits of the generator at node 1.

[0088]

[0089]

[0090] where variables and represent the upper and lower limits of voltage magnitude at node ; variable represents the upper limit of current square on line ; constraint (16) represents the voltage balance constraint at node the voltage magnitude on line is limited; constraints (17) represent limits on the current on line

[0091]

[0092]

[0093] where variables and represent the active and reactive power demand at node , respectively. Constraints (18) and (19) represent the active and reactive power limits of the switched load at node , respectively.

[0094]

[0095]

[0096]

[0097]

[0098]

[0099]

[0100]

[0101]

[0102]

[0103] where variables and represent the power factor angles of the configured wind and photovoltaic devices. Constraints (20)-(22) represent the active and reactive power output limits of the fixed energy storage; constraints (23)-(25) represent the active and reactive power output limits of the wind; and constraints (26)-(28) represent the active and reactive power output limits of the photovoltaic.

[0104]

[0105]

[0106]

[0107]

[0108]

[0109] where variables , , and are introduced auxiliary variables, since the second order cone relaxation constraints contain quadratic equality constraints, which are inconvenient for subsequent calculation, auxiliary variables are introduced to convert them into second order cones, and constraints (29)-(32) represent the converted second order cone relaxation constraints.

[0110]

[0111]

[0112]

[0113]

[0114] where, represents the maximum number of mobile energy storage devices. Constraint (34) represents the number limit of mobile energy storage devices. Constraint (35) represents that there are n-1 effective lines in the system, which guarantees the connectivity of the power distribution network; constraint (36) represents the current limit on line ; constraint (37) represents the power flow limit on line .

[0115]

[0116]

[0117]

[0118] where the superscript s2 indicates that all variables change under short-term extreme scenarios , and the variables in the brackets after each constraint represent the dual variables of the constraint. Variables and denote the lower node the actual injected active power of the mobile energy storage device at the upper configuration; the constant K denotes the number of attacking lines. The constraint (38) represents the limit on the number of attacking lines; the constraints (39) and (40) represent the active and reactive power balance constraints of node i, respectively.

[0119]

[0120]

[0121]

[0122]

[0123]

[0124]

[0125]

[0126]

[0127]

[0128]

[0129]

[0130]

[0131]

[0132]

[0133]

[0134] where the variable in the bracket after each constraint denotes the dual variable of the constraint. The constraints (41) and (42) represent the active and reactive power balance constraints of node voltage balancing constraints; constraints (43) and (44) limit the active and reactive power of line and ensure network connectivity; constraints (45) and (46) represent the active and reactive power limits of the generator at node 1; constraint (47) represents the limit on the voltage magnitude at node ; constraint (48) represents the limit on the current on line ; constraints (49) and (50) represent the active and reactive power limits of the load shedding at node ; constraints (51)-(55) represent the second-order cone relaxation constraints after introducing auxiliary variables.

[0135] The scenario-driven optimization model is used to configure high-cost stable energy units under the long-term planning framework to reduce the impact of wind and photovoltaic volatility on power network line loss, and the model is:

[0136]

[0137] ;

[0138] The emergency response scheduling model is used to configure low-cost flexible energy units under the short-term planning framework to reduce the load loss caused by the attack on the power line, and the model is:

[0139]

[0140] .

[0141] Step 4, based on the optimization model, a two-level solution structure of the master problem and the slave problem is built, and the optimal allocation scheme is obtained, as shown in Figure 2 .

[0142] Three groups of variables and 、 and 、 and are introduced, which respectively represent the installation location of the distributed energy unit according to the scenario-driven optimization model and the emergency response scheduling model, and three constraints 、 and are further introduced, which represent the solution based on the scenario-driven optimization model, the long-term planning model and the short-term emergency model. According to the distributed collaborative optimization algorithm, based on the scenario-driven optimization model, the problem can be divided into a master problem and two slave problems, wherein the master problem is:

[0143]

[0144]

[0145]

[0146] In the above formulae, are newly introduced variables, , , and , , respectively represent the dual variables corresponding to the constraints, respectively represent the optimal solutions obtained based on the situation-driven optimization model and the emergency response scheduling model. In this problem, the variables include , , and .

[0147] The master problem of the situation-driven optimization model is:

[0148]

[0149]

[0150]

[0151]

[0152] s.t. (5)-(33);

[0153] The master problem of the emergency response scheduling model is:

[0154]

[0155]

[0156]

[0157]

[0158] s.t. (38)-(55).

[0159] The embodiment establishes a coupling mechanism of decision consistency between the situation-driven optimization model and the emergency response scheduling model, and makes the distributed energy unit configuration decision of each power node consistent in the situation-driven optimization model and the emergency response scheduling model by introducing a constraint condition; the coupling constraint ensures that the fixed energy unit configuration scheme obtained by the situation-driven optimization and the flexible energy unit deployment scheme obtained by the emergency response scheduling have the same value on the overlapping decision variables, thereby ensuring that the optimization decisions of the sub-models in different stages are coordinated. The master problem and the slave problem are solved in a way of iterative updating of the Lagrange multiplier: by adding the coupling constraint to the Lagrange relaxation and introducing the corresponding Lagrange multiplier, a Lagrange relaxation model including the master problem and the slave problem is constructed, and then the master problem and the slave problem are solved alternately and the Lagrange multiplier is updated according to the deviation of the solution after each iteration, and the cycle is repeated until convergence, so as to realize incremental collaborative optimization and solution between the master problem and the slave problem, and obtain an approximate global optimal solution that meets all constraints, until 、 With .

[0160] Embodiment 2

[0161] The application provides a multi-variable environment-oriented intelligent power network resilience enhancement energy configuration system for realizing the method in the above technical scheme, comprising:

[0162] An uncertainty modeling module is configured to perform uncertainty analysis on the output of wind power and photovoltaic power generation and generate a corresponding typical wind speed and irradiance scenario set;

[0163] A fault situation generation module is configured to construct a N-K criterion power network line fault situation and determine the combination and number limit of the attacked line in the fault situation;

[0164] An optimization model construction module is configured to establish a distributed energy configuration optimization model with the minimum power grid line loss as the target, and set binary decision variables representing the installation state of the distributed energy unit and the reconstruction state of the power grid line in the model;

[0165] A model decomposition module is configured to build a two-level solving structure of the master problem and the slave problem based on the optimization model, take the fixed energy unit configuration and network topology decision as the master problem, and take the situation-driven optimization and the emergency response scheduling as two slave problems; wherein the situation-driven optimization is used for long-term planning to determine the optimal configuration of the fixed energy unit, and the emergency response scheduling is used for short-term scheduling to determine the optimal deployment of the flexible energy unit;

[0166] A coordination module is configured to set a decision consistency constraint between the master problem and the slave problem, and perform relaxation and update based on the constraint; and iteratively solve until a global optimal energy configuration scheme is obtained, thereby improving the resilience of the power network in an uncertain environment.

[0167] Embodiment 3

[0168] The application provides a computer readable storage medium, which has a computer program stored thereon, the computer program being executed by a processor to implement the intelligent power network resilience enhancing energy configuration method for a variable environment.

[0169] Embodiment 4

[0170] The application provides an electronic device, which comprises a memory and a processor, the memory and the processor being in communication connection with each other, the memory storing computer instructions, and the processor executing the computer instructions to implement the intelligent power network resilience enhancing energy configuration method for a variable environment.

[0171] Those skilled in the art should understand that the embodiments of the application can be provided as a method, a system, or a computer program product. Therefore, the application can be in the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the application can be in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0172] The application is described with reference to flowcharts and / or block diagrams of the method, device (system), and computer program product according to the embodiments of the application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device that implements the flowcharts and / or block diagrams. Figure 1 The function of the device specified in one flow or multiple flows and / or blocks Figure 1 The function of the device specified in one flow or multiple flows and / or blocks

[0173] These computer program instructions can also be stored in a computer readable storage medium capable of guiding a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable storage medium produce a product comprising instruction devices, which implement the flowcharts and / or block diagrams. Figure 1 The function of the device specified in one flow or multiple flows and / or blocksFigure 1 the function specified in the one or more blocks.

[0174] These computer program instructions can also be loaded into computer or other programmable data processing devices, so that a series of operational steps are performed on the computer or other programmable data processing devices to generate computer-implemented processes, thus the instructions executed on the computer or other programmable data processing devices provide the function implemented in the flow Figure 1 the flow or flows and / or blocks Figure 1 the function specified in the one or more blocks.

[0175] The embodiments of the present application described above are only illustrative, not limiting, and the above specific embodiments are only illustrative, not limiting, and those of ordinary skill in the art can make many forms under the inspiration of the present application without departing from the purpose of the present application and the scope protected by the claims, which all belong to the protection of the present application.

[0176] The contents not described in detail in the specification belong to the prior art known to those skilled in the art.

Claims

1. A method for energy configuration to enhance the resilience of a smart power network in a changing environment, characterized by: The following steps are involved: Generate typical scenarios based on the uncertainty of wind and photovoltaic power output; build a power line fault model according to the NK criterion to simulate the extreme scenario of simultaneous failure of multiple lines; Considering typical scenarios and fault conditions, an optimization model is established to minimize line loss. Setting binary decision variables to represent the configuration status of distributed energy units and the topological reconstruction of power grid lines; Based on the optimization model, a two-level solution structure of master and slave problems is established, with fixed energy unit configuration and network topology decision-making as the master problem, and scenario-driven optimization and emergency response scheduling as the two slave problems. Scenario-driven optimization is used for long-term planning to determine the optimal configuration of fixed energy units, while emergency response scheduling is used for short-term scheduling to determine the optimal deployment of flexible energy units. By setting decision consistency constraints between the master problem and the slave problem, and relaxing and updating based on the constraints; iteratively solving until the global optimal energy configuration solution is obtained, the resilience of the power network in uncertain environments is improved.

2. The method according to claim 1, characterized in that The modeling of wind power and photovoltaic power generation output uncertainty adopts a scenario generation method that combines Latin hypercube sampling with scenario reduction to obtain several representative wind speed and irradiance scenarios and their occurrence probabilities. A large number of initial scenarios are generated from the probability distribution of wind speed and solar irradiance through Latin hypercube sampling, and then the scenario reduction algorithm is used to compress the initial scenarios into a representative set of typical scenarios.

3. The method according to claim 1, characterized in that The power line fault model is established by introducing a binary variable representing the power line attack state and setting a constraint that the number of attacked lines does not exceed K, which is used to characterize the situation where a maximum of K power lines fail simultaneously under a variable environment; When the binary variable of a power line indicates that the line is under attack, the line is considered to be disconnected due to a fault, so as to limit the power flow distribution under the corresponding fault scenario in the optimization model.

4. The method according to claim 1, wherein In the optimization model, binary decision variables are set to characterize the distributed energy unit configuration and grid topology reconstruction decisions: the first type of binary variable is used to indicate whether each power node is installed with a fixed energy unit or a flexible energy unit. If it is installed at the corresponding node, the value is 1, otherwise it is 0; The second type of binary variable is used to indicate the operating status of the power line. It takes the value 1 when the line is in operation and the value 0 when the line is disconnected. In this way, the installation selection of distributed energy units and the grid reconstruction plan are introduced into the optimization decision.

5. The method according to claim 1, wherein The optimization model is a two-stage optimization model. The first stage configures fixed energy units, and the second stage configures flexible energy units and grid reconstruction constraints. Its objective function includes the configuration cost of fixed energy units, grid line loss cost, configuration cost of flexible energy units, and load loss cost under extreme fault scenarios. The constraints of the optimization model include the power network's flow balance constraints, power output limitations of generators and distributed energy units, line transmission capacity limitations, node voltage upper and lower limit constraints, fixed energy unit configuration quantity constraints, flexible energy unit configuration quantity constraints, and the constraint that the number of damaged lines does not exceed K under the power line fault model.

6. The method according to claim 1, characterized in that The main problem aims to minimize the configuration cost of fixed energy units, and the constraints include the total installed number limit of fixed energy units at all nodes, the total configurable number limit of wind power devices, and the total installed number limit of photovoltaic power generation devices at each node.

7. The method according to claim 1, characterized in that The scenario-driven optimization problem aims to minimize the line loss weighted by the probability of occurrence of each wind power and photovoltaic scenario. The constraints include the power flow equation and balance conditions of the power network under normal operation, as well as the operation and capacity limitations of the fixed energy units. A decomposition-based probabilistic optimization algorithm is used to solve the problem, thereby determining the optimal configuration plan for the fixed energy units at each power node.

8. The method according to claim 1, characterized in that The emergency response scheduling problem aims to minimize the load loss caused by the attacked line failure. The constraints include the power network's flow balance constraints under fault scenarios, the operation constraints of flexible energy units, and the grid topology reconstruction constraints. The deployment plan of flexible energy units is iteratively optimized through an incremental collaborative solution algorithm.

9. The method according to claim 1, characterized in that A decision-consistency coupling mechanism is established between the master problem and the slave problem, and the master problem and the slave problem are collaboratively solved by adopting an iterative update method of Lagrange multipliers; By adding coupling constraints to Lagrangian relaxation and introducing corresponding Lagrangian multipliers, a Lagrangian relaxation model consisting of a master problem and a slave problem is constructed. The master problem and the slave problem are then solved alternately, and the Lagrangian multipliers are updated according to the deviation of the solution after each iteration. This cycle is repeated until convergence, thereby achieving incremental collaborative optimization among the sub-models and obtaining the global optimal solution that satisfies all constraints.

10. An intelligent power network resilience enhancement energy configuration system for a changing environment, characterized in that: Used to perform the method according to any one of claims 1 to 9, comprising: Uncertainty modeling module, used to perform uncertainty analysis on wind power and photovoltaic power generation output and generate corresponding typical wind speed and irradiance scenario sets; A fault scenario generation module is used to construct a power network line fault scenario of the NK criterion and determine the combination and quantity limit of the attacked lines in the fault scenario; An optimization model building module is used to establish an optimization model with the goal of minimizing grid line losses, and to set binary decision variables representing the installation status of distributed energy units and the reconstruction status of grid lines in the model; A model decomposition module builds a two-level solution structure of master and slave problems based on the optimization model, with fixed energy unit configuration and network topology decision-making as the master problem, and scenario-driven optimization and emergency response scheduling as the two slave problems. Scenario-driven optimization is used for long-term planning to determine the optimal configuration of fixed energy units, while emergency response scheduling is used for short-term scheduling to determine the optimal deployment of flexible energy units. The collaborative coordination module is used to set decision consistency constraints between the master problem and the slave problem, and relax and update them based on the constraints; it iteratively solves until the global optimal energy configuration solution is obtained, thereby improving the resilience of the power network in an uncertain environment.

Images