A Multi-Scenario, Multi-Dimensional, and Multi-Level Resource Carrying Capacity Assessment Method for Distribution Networks
By constructing a multi-level structured evaluation system through multi-scenario Monte Carlo simulation and K-means clustering algorithm, and combining dynamic weights and multi-resource coordination strategies, the system solves the problems of insufficient multi-scenario coverage and resource coordination optimization in existing distribution network evaluation methods, and realizes accurate quantification and dynamic evaluation of distribution network carrying capacity.
Patent Information
- Application Number
- CN202610988830.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-03
- Publication Date
- 2026-11-10
AI Technical Summary
Existing power distribution network assessment methods cannot fully cover boundary conditions in multiple scenarios, lack a unified quantitative system across multiple dimensions, cannot adapt to random fluctuations in the output of distributed power sources, have conservative assessment results with poor adaptability, and have isolated improvement strategies, making it difficult to achieve multi-resource collaborative optimization, thus limiting the potential for capacity improvement.
Monte Carlo simulation is used to generate multiple scenarios. K-means clustering algorithm is used for dimensionality reduction to construct a multi-level structured index system. Subjective and objective weighting methods are used to calculate dynamic weights and construct a two-level optimization model. Combined with the synergistic improvement strategy of static var compensator, energy storage, network reconfiguration and demand response, dynamic evaluation is carried out through fuzzy comprehensive evaluation method.
It achieves dynamic coverage of all operating conditions, accurately quantifies the carrying capacity of the distribution network, forms a multi-resource collaborative improvement mechanism, accurately locates bottlenecks, dynamically adjusts weights, outputs comprehensive scores and levels, and supports scientific evaluation of high-proportion distributed resource access.
Smart Images

Figure CN123069837A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of power system technology, and in particular to a method for assessing the multi-dimensional and multi-level resource carrying capacity of distribution networks in multiple scenarios. Background Technology
[0002] Currently, a wide range of resources, including distributed photovoltaic (PV), distributed wind power, energy storage systems, flexible loads, and electric vehicles, are being integrated into the distribution network on a large scale. The traditional passive, unidirectional radial distribution network is rapidly transforming into a new type of distribution network that is active, bidirectional, and multi-source coordinated. The installed capacity of distributed power sources continues to climb, with some regions seeing new distributed PV installations exceeding those of centralized systems. The distribution network is shifting from partially accepting distributed power sources to globally absorbing them, placing disruptive demands on grid power balance, reactive power regulation, power quality, protection coordination, and dispatch operation. With the rapid implementation of new business models such as county-wide PV, industrial park distributed power sources, transformer area energy storage, and virtual power plants, the uncertainty on both the source and load sides of the distribution network has significantly increased. The intermittency and volatility of PV and wind power output, combined with the spatiotemporal randomness of loads, leads to frequent problems in the distribution network such as voltage exceeding limits, backflow of power, line overload, harmonic amplification, three-phase imbalance, and protection malfunctions. Traditional distribution network planning methods, operation control strategies, and evaluation systems are no longer adequate for the complex operating scenarios of high-proportion distributed resource integration. Current assessments of distribution network resource carrying capacity generally rely on static typical operating conditions, failing to cover multiple scenario boundary conditions such as light load, high photovoltaic power generation, low load, fault maintenance, and extreme weather, thus failing to reflect the true carrying capacity limit under source-load time-series fluctuations. Assessment indicators often focus on single safety constraints such as voltage and load rate, lacking a unified quantitative system encompassing multiple dimensions including safety, economy, flexibility, reliability, power quality, and low carbon emissions. The model hierarchy is unclear, lacking a multi-layered nested framework of target layer, criterion layer, and indicator layer, making it difficult to accurately locate weak nodes and bottleneck lines from overall carrying capacity assessment. Weight assignment often uses fixed subjective weights or static objective weights, unable to dynamically adjust with operating conditions, resulting in conservative, distorted, and poorly adaptable assessment results. Carrying capacity improvement methods often employ single measures such as reactive power compensation, voltage regulation, energy storage, and network reconfiguration, failing to achieve multi-device, multi-strategy, and multi-level collaborative optimization, resulting in limited improvement space and poor control effects. This severely restricts the orderly grid connection and efficient consumption of distributed power sources, posing significant challenges to the safe, reliable, economical, and low-carbon operation of the distribution network.
[0003] Currently, for assessing the carrying capacity of distribution networks containing distributed generation (DG) sources, mainstream domestic and international technical solutions mostly use the maximum connectable capacity of DG sources as the assessment target. They rely on traditional power flow calculations and security constraint verification for analysis. Some methods use analytic hierarchy process (AHP), entropy weighting method, etc., to assign weights to indicators and combine fuzzy comprehensive evaluation to score the distribution network's operational status. These methods typically first determine the distribution network topology parameters, load levels, and DG access schemes, setting static operating constraints such as voltage, current, and line power. Then, they use trial-and-error methods, sensitivity analysis, or conventional optimized power flow algorithms to calculate the distribution network's carrying capacity. Finally, they select indicators such as voltage qualification rate, network loss rate, and line load rate for weighted calculations to complete the comprehensive assessment and output results. These methods have found some application in engineering projects.
[0004] Existing assessment methods generally have significant limitations, primarily in their limited scenario coverage. Traditional assessments often employ static scenarios based on typical days or operating conditions, failing to cover multiple operational scenarios such as high load, light load, large-scale photovoltaic power generation, power flow reversal, voltage overruns, and fault maintenance. This makes it difficult to reflect the true load-bearing capacity boundaries under source-load time-series fluctuations and extreme conditions. Furthermore, they suffer from single-dimensionality and fragmented indicators. Existing methods often focus on single safety constraints such as voltage and load factor, failing to form a unified multi-dimensional assessment system encompassing safety, economy, flexibility, power quality, and reliability. This results in an inability to comprehensively characterize the overall load-bearing capacity level, and there are also issues of missing layers and insufficient coupling, lacking a multi-layered structured model of target layer—criteria layer—indicator layer. Achieving collaborative modeling across the planning, operation, and control layers is insufficient to support a full-chain analysis from overall assessment to local bottleneck identification. Furthermore, the static weights and poor adaptability of mainstream analytic hierarchy process (AHP) and entropy weight method, which employ fixed weights, fail to adapt to the dynamic changes in the importance of indicators caused by random fluctuations in distributed power output and time-varying loads. As a result, the assessment results are either conservative or distorted, and the improvement strategies are relatively isolated. Existing methods such as voltage regulation, reactive power compensation, energy storage configuration, and network reconfiguration are mostly applied separately, failing to form a collaborative improvement mechanism that integrates reactive power compensation, energy storage, network reconfiguration, and demand response. Consequently, the capacity improvement space is limited, and the overall assessment accuracy, adaptability, and improvement effect cannot meet the actual needs of the large-scale acceptance of distributed resources in new distribution networks. Summary of the Invention
[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method for evaluating the multi-dimensional and multi-level resource carrying capacity of distribution networks in multiple scenarios.
[0006] The objective of this invention is achieved through the following technical solution: a multi-dimensional and multi-level resource carrying capacity assessment method for multi-scenario distribution networks, which includes the following steps:
[0007] S1. Based on historical operation data of distribution network, output time series data of distributed power sources, load time series data, meteorological data and equipment maintenance plan, Monte Carlo simulation is used to generate source-load operation scenarios, and dimensionality is reduced by clustering algorithm to extract typical operation scenarios covering all operating conditions.
[0008] S2. Construct a three-level structured indicator system including the target layer, criterion layer, and indicator layer;
[0009] S3. Using a combination of subjective and objective weighting methods, calculate the dynamic comprehensive weight of each indicator under different operating scenarios and at different times;
[0010] S4. Construct a two-layer optimization model that includes an upper-layer model and a lower-layer model; the upper-layer model takes maximizing the capacity of distributed power source access and / or the capacity of energy storage configuration as decision variables, and aims to maximize the comprehensive carrying capacity of the distribution network; the lower-layer model takes reactive power compensation capacity, energy storage charging and discharging power and / or network reconfiguration switching status as decision variables, and aims to minimize voltage overruns and network losses.
[0011] S5. Based on the solution results of the two-layer optimization model, a multi-resource collaborative improvement strategy including static var compensator optimization configuration, energy storage optimization configuration, network reconfiguration and demand response is adopted to improve the carrying capacity of the distribution network.
[0012] S6. Introduce the calculated dynamic comprehensive weight into the fuzzy comprehensive evaluation method, combine it with the indicator data under each typical operation scenario, calculate the comprehensive score and level of the distribution network resource carrying capacity, and complete the evaluation.
[0013] Specifically, in S1, a photovoltaic power output scenario is generated using a Beta distribution, a wind power output scenario is generated using a Weibull distribution, a load scenario is generated using a normal distribution, and a source-load joint scenario is constructed using a Copula function;
[0014] The clustering algorithm uses the K-means clustering algorithm, which aims to minimize the sum of intra-cluster distances to reduce the dimensionality of the source load operation scenario and calculate the probability of each typical scenario.
[0015] Specifically, the criteria layer includes dimensions of safety, economy, flexibility, reliability, and power quality;
[0016] The indicators of the safety dimension include at least one of voltage qualification rate, line load rate, transformer load rate, and power flow backflow rate.
[0017] The indicators for the economic dimension include at least one of network loss rate, investment cost, and operation and maintenance cost;
[0018] The metrics for the flexibility dimension include at least one of adjustable capacity margin, response speed, and adjustment cost;
[0019] The reliability dimension indicators include at least one of power supply reliability rate, power outage time, and power supply capacity margin.
[0020] The power quality dimensions include at least one of harmonic distortion rate, three-phase imbalance, and voltage fluctuation and flicker.
[0021] Specifically, the objective function of the upper-level model is:
[0022] ;
[0023] In the formula, t represents the objective function value of the comprehensive carrying capacity of the distribution network's multiple resources; t represents the operating time; T represents the set of all times within the evaluation period; i represents the distributed resource access node number; and D represents the set of all distributed resource access nodes. It provides power to the distributed power source of node i.
[0024] Specifically, the constraints of the upper-level model include:
[0025] Current constraints:
[0026] ;
[0027] ;
[0028] ;
[0029] In the formula, Inject active power into node m at time t; Inject reactive power into node m at time t; , Let K be the active and reactive power of branch line km at time t; Let be the voltage amplitude at node m at time t; Let be the voltage amplitude at node k at time t; , For line resistance and reactance; Branch current;
[0030] Voltage constraint:
[0031] ;
[0032] In the formula, Node voltage; , These are the upper and lower limits of the allowable node voltage;
[0033] SVC reactive power constraint:
[0034] ;
[0035] In the formula, This refers to the reactive power compensation amount of the SVC. This is the lower limit of reactive power compensation for SVC; This is the upper limit of reactive power compensation for SVC.
[0036] Specifically, the objective function of the lower-level model is:
[0037] ;
[0038] In the formula, Let be the per-unit voltage value of node i; , represents the upper and lower limits of the node voltage; N represents the total number of systems.
[0039] Specifically, the constraints of the lower-level model include:
[0040] Energy storage operation constraints:
[0041] ;
[0042] ;
[0043] In the formula, For energy storage unit k in the next time The amount of stored energy in each step; Let be the amount of electricity stored in energy storage unit k at time t; Let be the charging power of energy storage unit k at time t; Let k be the discharge power of energy storage unit k at time t; The charging efficiency and discharging efficiency of energy storage unit k; For time step; These represent the maximum charging power and maximum discharging power of the energy storage unit.
[0044] Power balance constraints:
[0045] ;
[0046] ;
[0047] In the formula, For active and reactive power of the power supply; For active and reactive power loads; The set of nodes directly connected to node i; The active and reactive power of the branch between node i and node j.
[0048] Specifically, the multi-resource collaborative enhancement strategy in S5 includes:
[0049] The optimal grid connection point and compensation capacity of the static var compensator are determined based on the minimum singularity sensitivity method.
[0050] Based on the goal of minimizing voltage over-limit, the location and capacity of energy storage are optimized, and a dynamic model of the state of charge of energy storage is established.
[0051] Distribution network reconfiguration is achieved by optimizing branch switch status, and flexible load demand response is used as an auxiliary adjustment method.
[0052] With the overall goal of synergistically improving the minimization of node voltage exceedance and total active power loss of the distribution network, the configuration and operation strategies of each resource are iteratively optimized.
[0053] Specifically, the fuzzy comprehensive evaluation method in S6 includes:
[0054] Construct a factor set with evaluation indicators as elements and a comment set with carrying capacity levels as elements:
[0055] ;
[0056] In the formula, The factor set is m; the total number of evaluation indicators is m. Let i be the evaluation index;
[0057] ;
[0058] In the formula, A collection of comments; There are 5 rating levels;
[0059] The membership degree of each indicator to each evaluation level is calculated using the membership function, forming a fuzzy evaluation matrix:
[0060] ;
[0061] In the formula, This is a fuzzy evaluation matrix;
[0062] ;
[0063] In the formula, Let be the membership degree of the i-th indicator to the j-th evaluation level;
[0064] The dynamic comprehensive weight vector and the fuzzy evaluation matrix are combined using a fuzzy synthesis operation to obtain the comprehensive evaluation vector:
[0065] ;
[0066] ;
[0067] In the formula, Let be the dynamic comprehensive weight of the i-th indicator at time t;
[0068] ;
[0069] ;
[0070] In the formula, This is a comprehensive evaluation vector; The comprehensive membership degree of the overall carrying capacity of the distribution network corresponding to the j-th evaluation level; For fuzzy synthesis operators;
[0071] The comprehensive evaluation vector and the scores corresponding to the comment set are weighted and summed to obtain the comprehensive score of the distribution network resource carrying capacity:
[0072] ;
[0073] In the formula, The comprehensive score of resource carrying capacity of the power distribution network in multiple scenarios; Let be the percentage score corresponding to the j-th evaluation level.
[0074] The present invention has the following advantages:
[0075] 1. This invention constructs a multi-scenario evaluation system covering all operating conditions, establishes a multi-dimensional integrated and multi-layered nested structured evaluation framework, realizes dynamic adaptive adjustment of indicator weights, forms a multi-resource collaborative capacity enhancement mechanism, accurately quantifies the distribution network's capacity to support diverse distributed resources, and provides scientific support for the planning, operation, grid connection approval, and optimization and transformation of distribution networks with high proportions of distributed power sources, energy storage, and flexible load access.
[0076] 2. This invention constructs a source-load uncertainty model based on Beta distribution, Weibull distribution, normal distribution and Copula function, and achieves dimensionality reduction of massive operating scenarios through K-means clustering, forming a typical scenario set covering all operating conditions such as light load, large photovoltaic power generation, power flow reversal, voltage over-limit, and fault repair, thus solving the problem of result distortion caused by traditional assessment relying only on static typical days;
[0077] 3. This invention constructs a two-layer planning model with the upper-layer objective of maximizing the distributed power supply access capacity and the lower-layer objective of minimizing the node voltage exceedance. It combines the Distflow power flow equation to transform the non-convex problem into a convex optimization problem, thereby achieving accurate quantification of carrying capacity and efficient global solution.
[0078] 4. This invention determines the optimal grid connection point of SVC based on minimum singularity sensitivity, optimizes energy storage location and capacity based on minimizing voltage over-limit, and combines network reconfiguration and demand response to form a synergistic improvement mechanism of reactive power compensation + energy storage + network reconfiguration + demand response, systematically alleviating operational bottlenecks such as voltage over-limit, power flow reversal, and line overload.
[0079] 5. This invention introduces dynamic comprehensive weights into the fuzzy comprehensive evaluation system, and completes the multi-index fusion calculation through the membership matrix and weighted average operator to output the comprehensive score and grade evaluation of the distribution network carrying capacity, thereby realizing the dynamic quantitative evaluation of the distribution network resource carrying capacity in all scenarios and throughout the entire process. Attached Figure Description
[0080] Figure 1 This is a schematic diagram of the evaluation method of the present invention. Detailed Implementation
[0081] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention; that is, the described embodiments are merely some embodiments of the invention, and not all embodiments. The components of the embodiments of the invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.
[0082] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.
[0083] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0084] The present invention will be further described below with reference to the accompanying drawings, but the scope of protection of the present invention is not limited to the following description.
[0085] like Figure 1 As shown, a multi-scenario, multi-dimensional, and multi-level resource carrying capacity assessment method for distribution networks is presented. This method includes the following steps:
[0086] S1. Based on historical operation data of distribution network, output time series data of distributed power sources, load time series data, meteorological data and equipment maintenance plan, Monte Carlo simulation is used to generate source-load operation scenarios, and dimensionality is reduced by clustering algorithm to extract typical operation scenarios covering all operating conditions.
[0087] In this embodiment, the source-load operation scenarios cover typical operating conditions such as normal operation, high load, light load, large photovoltaic power generation, power flow reversal, voltage over-limit, fault repair, and extreme weather. The massive number of scenarios are reduced in dimensionality by using the K-means clustering algorithm to extract representative typical operating scenarios and eliminate redundant and similar scenarios. This reduces the computational complexity while ensuring the accuracy of the assessment, and achieves dynamic coverage of the entire operating conditions of the distribution network, solving the problem of large deviations between traditional static scenario assessments and actual operation.
[0088] In this embodiment, a Beta distribution is used to generate the photovoltaic power output scenario:
[0089] ;
[0090] In the formula, Solar irradiance; Maximum irradiance; The shape parameter of the Beta distribution; It is a gamma function;
[0091] Photovoltaic power output model:
[0092] ;
[0093] In the formula, Let t be the active power output of the photovoltaic system. Rated power of photovoltaic power; Standard test irradiance (1000 W / m²); The temperature coefficient of photovoltaic power; The temperature of the photovoltaic cell; The standard test temperature is 25℃.
[0094] Wind power output scenarios are generated using Weibull distribution:
[0095] ;
[0096] In the formula, Wind speed; The shape parameter of the Weibull distribution; The scaling parameter of the Weibull distribution;
[0097] Wind turbine output model:
[0098] ;
[0099] In the formula, Contribute to the wind turbine; This refers to the rated power of the fan. To cut in wind speed; Rated wind speed; To cut off the wind speed;
[0100] Load scenarios are generated using a normal distribution:
[0101] ;
[0102] In the formula, The active power of the load; This is the average load. The standard deviation of the load;
[0103] And a source-load joint scenario is constructed using Copula functions:
[0104] ;
[0105] In the formula, The probability of the source / charge edge distribution; It is an n-dimensional normal distribution function; It is an inverse normal distribution function; The correlation coefficient.
[0106] The clustering algorithm used is K-means clustering, which aims to minimize the sum of intra-cluster distances to reduce the dimensionality of the source load operation scenarios and calculate the probability of each typical scenario.
[0107] ;
[0108] ;
[0109] In the formula, The sum of intra-cluster distances (objective function); Number of typical scenarios; For the set of scenarios of type k; For a single running scenario; It is the cluster center of the k-th class; Let k be the number of scenarios of type k.
[0110] ;
[0111] In the formula, This represents the probability of the k-th typical scenario. This represents the total number of scenes generated.
[0112] S2. Construct a three-level structured indicator system including the target layer, criterion layer, and indicator layer;
[0113] The criteria layer includes dimensions of safety, economy, flexibility, reliability, and power quality.
[0114] The indicators of the safety dimension include at least one of voltage qualification rate, voltage deviation index, maximum line load rate, transformer load rate, and power flow backflow rate.
[0115] The voltage qualification rate is:
[0116] ;
[0117] In the formula, This represents the number of voltage-qualified nodes. This represents the total number of nodes in the system.
[0118] Voltage offset index:
[0119] ;
[0120] In the formula, The actual voltage of node n at time t; The node's rated voltage;
[0121] Maximum line load rate:
[0122] ;
[0123] In the formula, Let t be the active power of branch l. The maximum allowable transmission power for branch l; This represents the total number of branches in the system.
[0124] Transformer load rate:
[0125] ;
[0126] In the formula, Let r be the actual load capacity of the r-th transformer at time t; R represents the rated capacity or maximum allowable capacity of the r-th transformer; R is the total number of transformers in the system.
[0127] The trend reversal rate is:
[0128] ;
[0129] In the formula, For the trend of reverse transmission rate; Let be the active power of branch l at time t, and define the positive direction of the active power of the branch as flowing from the upstream power grid to the load side; when When <0, it indicates that the branch is experiencing backflow; L is the set of branches in the system.
[0130] The economic indicators include at least one of the following: grid loss rate, investment cost, operation and maintenance cost, and the proportion of renewable energy generation; wherein, the grid loss rate is:
[0131] ;
[0132] In the formula, For the resistance of branch l; Let be the active and reactive power of branch l at time t; The voltage at the end node of branch l.
[0133] Percentage of renewable energy generation:
[0134] ;
[0135] In the formula, The output of the distributed power source at node n at time t; Let n be the load power at node n at time t; For DG access node set;
[0136] The investment cost is:
[0137] ;
[0138] In the formula, 1. Annualized investment cost; A. Set of equipment to be configured; Let A be the unit investment cost of equipment A; The configuration capacity or number of devices a; Let be the annualized factor for device a; The discount rate; The service life of device a;
[0139] The operation and maintenance costs are:
[0140] ;
[0141] In the formula, For maintenance costs The unit fixed maintenance cost for device a; The unit variable maintenance cost for device a; Let be the output of device a at time t; For time step.
[0142] The indicators for the flexibility dimension include at least one of adjustable capacity margin, response speed, adjustment cost, and net load volatility; wherein, net load volatility is:
[0143] ;
[0144] In the formula, This provides adjustable capacity margin. Let A be the capacity that adjustable resource a can be adjusted upwards at time t; A is the set of adjustable resources. Let t be the total system load at time t;
[0145] The response speed is:
[0146] ;
[0147] In the formula, This refers to the overall system response time. The response time of adjustable resource a; Let be the power of adjustable resource a that participates in the adjustment at time t;
[0148] The adjustment cost is:
[0149] ;
[0150] In the formula, To adjust costs; The unit adjustment cost of adjustable resource a; Let be the adjustable power of adjustable resource a at time t; For time step;
[0151] Net load volatility is:
[0152] ;
[0153] In the formula, The net loads at time t and t-1 are respectively, where the net load is the difference between the system load and the output of the distributed power source.
[0154] The reliability dimension indicators include at least one of power supply reliability rate, power outage time, and power supply capacity margin.
[0155] The power supply reliability rate is:
[0156] ;
[0157] In the formula, For power supply reliability; Let n be the load power of node n; To evaluate the power outage time of node nnn within the evaluation period; The total duration of the evaluation period; N is the set of system nodes;
[0158] The power outage time is:
[0159] ;
[0160] In the formula, The system load-weighted average fault outage time Let n be the load power of node n; The fault outage time of node n during the evaluation period;
[0161] The power supply capacity margin is:
[0162] ;
[0163] In the formula, For power supply capacity margin; Let t be the maximum power supply capacity of the distribution network at time t; Let t be the total load of the system at time t.
[0164] The power quality dimensions include at least one of harmonic distortion rate, three-phase imbalance, and voltage fluctuation and flicker.
[0165] The harmonic distortion rate is:
[0166] ;
[0167] The three-phase imbalance is:
[0168] ;
[0169] In the formula, Let be the three-phase voltage imbalance at node n at time t; It is a negative sequence voltage component; It is the positive sequence voltage component;
[0170] Voltage fluctuation rate:
[0171] ;
[0172] In the formula, Let be the voltage fluctuation rate of node n at time t; and These represent the maximum and minimum voltages of node n within the statistical time window, respectively. Let n be the rated voltage of node n;
[0173] Long-term flicker value:
[0174] ;
[0175] In the formula, For the long-term flicker value of node n; Let M be the short-time flicker value of node n within the q-th short-time statistical window; M is the number of short-time statistical windows.
[0176] The constructed indicators are then positiveized and dimensionlessized:
[0177] Positive indicators (the higher the better):
[0178] ;
[0179] Negative indicators (the smaller the better):
[0180] ;
[0181] In the formula, This represents the original value of the j-th indicator for the i-th object; These are the maximum and minimum values of the indicator; The dimensionless index (0~1).
[0182] S3. Using a combination of subjective and objective weighting methods, calculate the dynamic comprehensive weight of each indicator under different operating scenarios and at different times;
[0183] S4. Construct a two-layer optimization model that includes an upper-layer model and a lower-layer model; the upper-layer model takes maximizing the capacity of distributed power source access and / or the capacity of energy storage configuration as decision variables, and aims to maximize the comprehensive carrying capacity of the distribution network; the lower-layer model takes reactive power compensation capacity, energy storage charging and discharging power and / or network reconfiguration switching status as decision variables, and aims to minimize voltage overruns and network losses.
[0184] In this embodiment, the objective function of the upper-level model is:
[0185] ;
[0186] In the formula, t represents the objective function value of the comprehensive carrying capacity of the distribution network's multiple resources; t represents the operating time; T represents the set of all times within the evaluation period; i represents the distributed resource access node number; and D represents the set of all distributed resource access nodes. It provides power to the distributed power source of node i.
[0187] The constraints of the upper-level model include:
[0188] Current constraints:
[0189] ;
[0190] ;
[0191] ;
[0192] In the formula, Inject active power into node m at time t; Inject reactive power into node m at time t; , Let K be the active and reactive power of branch line km at time t; Let be the voltage amplitude at node m at time t; Let be the voltage amplitude at node k at time t; , For line resistance and reactance; Branch current;
[0193] Voltage constraint:
[0194] ;
[0195] In the formula, Node voltage; , These are the upper and lower limits of the allowable node voltage;
[0196] SVC reactive power constraint:
[0197] ;
[0198] In the formula, This refers to the reactive power compensation amount of the SVC. This is the lower limit of reactive power compensation for SVC; This is the upper limit of reactive power compensation for SVC.
[0199] The objective function of the lower-level model is:
[0200] ;
[0201] In the formula, Let be the per-unit voltage value of node i; , represents the upper and lower limits of the node voltage; N represents the total number of systems.
[0202] The constraints of the lower-level model include:
[0203] Energy storage operation constraints:
[0204] ;
[0205] ;
[0206] In the formula, For energy storage unit k in the next time The amount of stored energy in each step; Let be the amount of electricity stored in energy storage unit k at time t; Let be the charging power of energy storage unit k at time t; Let k be the discharge power of energy storage unit k at time t; The charging efficiency and discharging efficiency of energy storage unit k; For time step; These represent the maximum charging power and maximum discharging power of the energy storage unit.
[0207] Power balance constraints:
[0208] ;
[0209] ;
[0210] In the formula, For active and reactive power of the power supply; For active and reactive power loads; The set of nodes directly connected to node i; The active and reactive power of the branch between node i and node j.
[0211] S5. Based on the solution results of the two-layer optimization model, a multi-resource collaborative improvement strategy including static var compensator (SVC) optimization configuration, energy storage optimization configuration, network reconfiguration, and demand response is adopted to improve the carrying capacity of the distribution network. This embodiment can optimize the grid connection point and capacity of SVC for voltage over-limit and line overload bottlenecks identified in the assessment, carry out network reconfiguration optimization, and tap the demand response potential of flexible loads.
[0212] In this embodiment, the multi-resource collaborative enhancement strategy includes:
[0213] The optimal grid connection point and compensation capacity of the static var compensator are determined based on the minimum singularity sensitivity method.
[0214] ;
[0215] In the formula, This is the minimum singular value (voltage stability margin). For the trend of Jacobi matrix; These are singular vectors on the left and right; The reactive power of node i;
[0216] SVC operational constraints:
[0217] ;
[0218] In the formula, This refers to the real-time reactive power compensation amount for the SVC. This represents the minimum reactive power compensation. This represents the maximum reactive power compensation.
[0219] Based on the objective of minimizing voltage exceedance, the location and capacity determination of energy storage are optimized, and a dynamic model of the energy storage state of charge is established; specifically as follows:
[0220] Energy storage power constraints:
[0221] ;
[0222] Energy Storage SOC Dynamic Model:
[0223] ;
[0224] In the formula, The energy stored at time t is the amount of electricity. This refers to the charging and discharging power. For charging and discharging efficiency; ;
[0225] Distribution network reconfiguration is achieved by optimizing branch switch states, and flexible load demand response is used as an auxiliary adjustment method; the details are as follows:
[0226] Branch switch status:
[0227] ;
[0228] In the formula, For branch closure; The branch is disconnected; This is a collection of all network branches; This represents the number of nodes.
[0229] With the overall goal of synergistically improving both node voltage exceedance and total active power loss in the distribution network, the allocation and operation strategies of various resources are iteratively optimized. Specifically:
[0230] ;
[0231] In the formula, For node voltage exceeding the limit; This represents the total active power loss of the distribution network.
[0232] S6. Introduce the calculated dynamic comprehensive weight into the fuzzy comprehensive evaluation method, combine it with the indicator data under each typical operation scenario, calculate the comprehensive score and level of the distribution network resource carrying capacity, and complete the evaluation.
[0233] This embodiment uses the fuzzy comprehensive evaluation method to quantitatively evaluate the resource carrying capacity of the distribution network. It combines dimensionless index data, dynamic adaptive weights, and fuzzy membership functions to complete the determination of the distribution network carrying capacity level and the calculation of the comprehensive score. The specific formulas and physical quantities are as follows:
[0234] Construct a factor set with evaluation indicators as elements and a comment set with carrying capacity levels as elements:
[0235] ;
[0236] In the formula, The factor set is m; the total number of evaluation indicators is m. Let i be the evaluation index;
[0237] ;
[0238] In the formula, A collection of comments; There are 5 rating levels;
[0239] The membership degree of each indicator to each evaluation level is calculated using the membership function, forming a fuzzy evaluation matrix:
[0240] ;
[0241] In the formula, This is a fuzzy evaluation matrix;
[0242] ;
[0243] In the formula, Let be the membership degree of the i-th indicator to the j-th evaluation level;
[0244] The dynamic comprehensive weight vector and the fuzzy evaluation matrix are combined using a fuzzy synthesis operation to obtain the comprehensive evaluation vector:
[0245] ;
[0246] ;
[0247] In the formula, Let be the dynamic comprehensive weight of the i-th indicator at time t; satisfy .
[0248] ;
[0249] Expanded to:
[0250] ;
[0251] In the formula, This is a comprehensive evaluation vector; The comprehensive membership degree of the overall carrying capacity of the distribution network corresponding to the j-th evaluation level; For fuzzy synthesis operators;
[0252] The comprehensive evaluation vector and the scores corresponding to the comment set are weighted and summed to obtain the comprehensive score of the distribution network resource carrying capacity:
[0253] ;
[0254] In the formula, The comprehensive score of resource carrying capacity of the power distribution network in multiple scenarios; The percentage score (100, 80, 60, 40, 20) corresponds to the j-th evaluation level.
[0255] The above description is merely a preferred embodiment of the present invention and does not constitute any limitation on the present invention. Any person skilled in the art can make many possible variations and modifications to the technical solution of the present invention, or modify it into equivalent embodiments, without departing from the scope of the present invention. Therefore, any modifications, equivalent changes, and alterations made to the above embodiments based on the technology of the present invention without departing from the scope of the present invention are within the protection scope of the present invention.
Claims
1. A method for assessing the resource carrying capacity of distribution networks across multiple scenarios and at multiple levels, characterized by: The method includes the following steps: S1. Based on historical operation data of distribution network, output time series data of distributed power sources, load time series data, meteorological data and equipment maintenance plan, Monte Carlo simulation is used to generate source-load operation scenarios, and dimensionality is reduced by clustering algorithm to extract typical operation scenarios covering all operating conditions. S2. Construct a three-level structured indicator system including the target layer, criterion layer, and indicator layer; S3. Using a combination of subjective and objective weighting methods, calculate the dynamic comprehensive weight of each indicator under different operating scenarios and at different times; S4. Construct a two-layer optimization model that includes an upper-layer model and a lower-layer model; the upper-layer model takes maximizing the capacity of distributed power source access and / or the capacity of energy storage configuration as decision variables, and aims to maximize the comprehensive carrying capacity of the distribution network; the lower-layer model takes reactive power compensation capacity, energy storage charging and discharging power and / or network reconfiguration switching status as decision variables, and aims to minimize voltage overruns and network losses. S5. Based on the solution results of the two-layer optimization model, a multi-resource collaborative improvement strategy including static var compensator optimization configuration, energy storage optimization configuration, network reconfiguration and demand response is adopted to improve the carrying capacity of the distribution network. S6. Introduce the calculated dynamic comprehensive weight into the fuzzy comprehensive evaluation method, combine it with the indicator data under each typical operation scenario, calculate the comprehensive score and level of the distribution network resource carrying capacity, and complete the evaluation.
2. The method for assessing the multi-dimensional and multi-level resource carrying capacity of a distribution network based on multiple scenarios as described in claim 1, characterized in that: In S1, a photovoltaic power output scenario is generated using a Beta distribution, a wind power output scenario is generated using a Weibull distribution, a load scenario is generated using a normal distribution, and a source-load joint scenario is constructed using a Copula function. The clustering algorithm uses the K-means clustering algorithm, which aims to minimize the sum of intra-cluster distances to reduce the dimensionality of the source load operation scenario and calculate the probability of each typical scenario.
3. The method for assessing the multi-dimensional and multi-level resource carrying capacity of a distribution network based on multiple scenarios as described in claim 1, characterized in that: The criteria layer includes dimensions of safety, economy, flexibility, reliability, and power quality. The indicators of the safety dimension include at least one of voltage qualification rate, line load rate, transformer load rate, and power flow backflow rate. The indicators for the economic dimension include at least one of network loss rate, investment cost, and operation and maintenance cost; The metrics for the flexibility dimension include at least one of adjustable capacity margin, response speed, and adjustment cost; The reliability dimension indicators include at least one of power supply reliability rate, power outage time, and power supply capacity margin. The power quality dimensions include at least one of harmonic distortion rate, three-phase imbalance, and voltage fluctuation and flicker.
4. The method for assessing the multi-dimensional and multi-level resource carrying capacity of a distribution network based on multiple scenarios, as described in claim 1, is characterized in that: The objective function of the upper-level model is: ; In the formula, The objective function value for the comprehensive carrying capacity of the distribution network's diverse resources; t represents the running time; T represents the set of all times within the evaluation period; i represents the distributed resource access node number; D represents the set of all distributed resource access nodes; It provides power to the distributed power source of node i.
5. The method for assessing the multi-dimensional and multi-level resource carrying capacity of a distribution network based on multiple scenarios, as described in claim 4, is characterized in that: The constraints of the upper-level model include: Current constraints: ; ; ; In the formula, Inject active power into node m at time t; Inject reactive power into node m at time t; , Let K be the active and reactive power of branch line km at time t; Let be the voltage amplitude at node m at time t; Let be the voltage amplitude at node k at time t; , For line resistance and reactance; Branch current; Voltage constraint: ; In the formula, Node voltage; , These are the upper and lower limits of the allowable node voltage; SVC reactive power constraint: ; In the formula, This refers to the reactive power compensation amount of the SVC. This is the lower limit of reactive power compensation for SVC; This is the upper limit of reactive power compensation for SVC.
6. The method for assessing the multi-dimensional and multi-level resource carrying capacity of a distribution network based on multiple scenarios, as described in claim 1, is characterized in that: The objective function of the lower-level model is: ; In the formula, Let be the per-unit voltage value of node i; , represents the upper and lower limits of the node voltage; N represents the total number of systems.
7. The method for assessing the multi-dimensional and multi-level resource carrying capacity of a distribution network based on multiple scenarios, as described in claim 6, is characterized in that: The constraints of the lower-level model include: Energy storage operation constraints: ; ; In the formula, For energy storage unit k in the next time The amount of stored energy in each step; Let be the amount of electricity stored in energy storage unit k at time t; Let be the charging power of energy storage unit k at time t; Let k be the discharge power of energy storage unit k at time t; Let K be the charging efficiency and discharging efficiency of energy storage unit k. For time step; These represent the maximum charging power and maximum discharging power of the energy storage unit. Power balance constraints: ; ; In the formula, For active and reactive power sources; For active and reactive power of the load; The set of nodes directly connected to node i; The active and reactive power of the branch between node i and node j.
8. The method for assessing the multi-dimensional and multi-level resource carrying capacity of a distribution network based on multiple scenarios, as described in claim 1, is characterized in that: The multi-resource collaborative enhancement strategy in S5 includes: The optimal grid connection point and compensation capacity of the static var compensator are determined based on the minimum singularity sensitivity method. Based on the goal of minimizing voltage over-limit, the location and capacity of energy storage are optimized, and a dynamic model of the state of charge of energy storage is established. Distribution network reconfiguration is achieved by optimizing branch switch status, and flexible load demand response is used as an auxiliary adjustment method. With the overall goal of synergistically improving the minimization of node voltage exceedance and total active power loss of the distribution network, the configuration and operation strategies of each resource are iteratively optimized.
9. The method for assessing the multi-dimensional and multi-level resource carrying capacity of a distribution network based on multiple scenarios, as described in claim 4, is characterized in that: The fuzzy comprehensive evaluation method in S6 includes: Construct a factor set with evaluation indicators as elements and a comment set with carrying capacity levels as elements: ; In the formula, The factor set is m; the total number of evaluation indicators is m. Let i be the evaluation index; ; In the formula, A collection of comments; There are 5 rating levels; The membership degree of each indicator to each evaluation level is calculated using the membership function, forming a fuzzy evaluation matrix: ; In the formula, This is a fuzzy evaluation matrix; ; In the formula, Let be the membership degree of the i-th indicator to the j-th evaluation level; The dynamic comprehensive weight vector and the fuzzy evaluation matrix are combined using a fuzzy synthesis operation to obtain the comprehensive evaluation vector: ; ; In the formula, Let be the dynamic comprehensive weight of the i-th indicator at time t; ; ; In the formula, This is a comprehensive evaluation vector; The comprehensive membership degree of the overall carrying capacity of the distribution network corresponding to the j-th evaluation level; For fuzzy synthesis operators; The comprehensive evaluation vector and the scores corresponding to the comment set are weighted and summed to obtain the comprehensive score of the distribution network resource carrying capacity: ; In the formula, The comprehensive score of resource carrying capacity of the power distribution network in multiple scenarios; This represents the percentage score corresponding to the j-th evaluation level.
