Main-distribution cooperative robust comprehensive reactive power optimization method considering conditional value-at-risk
Through the hierarchical reconstruction and comprehensive norm methods, combined with the conditional risk value measurement, the reactive power optimization problem of the main and distribution coordinated system was solved, the efficient, safe and economical operation of the system was achieved, the flexibility and reliability of the system were improved, and the dispatching cost was reduced.
Patent Information
- Application Number
- CN202510744292.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-05
- Publication Date
- 2025-09-12
AI Technical Summary
The existing main and distribution coordinated systems have safety and economic problems caused by global dynamic reconstruction strategies in reactive power optimization. The robust optimization methods have problems with overly conservative or inaccurate decisions. In addition, the reactive power optimization methods lack comprehensive consideration of the risk costs brought about by the flexibility, reliability, economy and safety of power system operation.
By reconstructing the hierarchical division, a hierarchical reconstruction model is established, the comprehensive norm is used to construct the uncertainty set of distributed PV and load, the conditional value at risk (CVaR) is introduced to measure the system operation risk, and a main and distribution coordinated robust integrated reactive power optimization model is established. The Gurobi solver is used to solve the model and output the optimal reactive power optimization scheduling scheme.
It achieves efficient, safe and economical operation of the main and distribution coordinated system, reduces dispatching costs, improves system flexibility, reliability and safety, balances economy and operational risks, and improves the conservatism of optimization results.
Smart Images
Figure CN120638527A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of reactive power optimization, and in particular relates to a main distribution coordinated robust comprehensive reactive power optimization method considering conditional risk value. Background Art
[0002] As the global energy structure accelerates its transition toward renewable energy, a large number of distributed power sources and regulation equipment are connected to the power system. While this promotes the development of clean energy, it also causes problems such as frequency fluctuations and power over-limits, making reactive power optimization technology a key to ensuring stable system operation. The main (substation level) and distribution (distribution line level) coordinated system integrates the resources and advantages of substations and distribution lines. Following the principles of layered zoning and local balancing, it can coordinate reactive power equipment at the substation and distribution line levels, avoiding the transmission of reactive power over long-distance lines or multi-stage transformers. This can effectively reduce the reactive power flowing on transmission lines, achieve the goal of reducing grid losses, and realize more efficient and coordinated operation of the power system. Therefore, studying reactive power optimization methods based on main and distribution coordination is of great significance.
[0003] Currently, there are three main problems that need to be solved in the reactive power optimization method based on the coordination of main and distribution systems: (1) The global dynamic reconstruction strategy currently used in the main and distribution coordination system has obvious problems in large-scale system applications. From a safety perspective, this strategy will disrupt the system power balance, trigger large-scale power flow transfer, lead to line overload, equipment damage, and even cascading failures; from an economic perspective, it involves a large number of equipment adjustments and complex scheduling, and the reconstruction and scheduling costs are high. Comprehensive considerations show that the global dynamic reconstruction strategy is not suitable for large-scale systems, and new solutions are urgently needed to ensure the efficient, safe, and economical operation of the main and distribution coordination system.
[0004] (2) Most existing power distribution coordination systems use mathematical sets to restrict the output range or scenario probability distribution of distributed photovoltaic (PV) power generation, and establish robust optimization or stochastic optimization methods for power distribution coordination systems. However, these two methods suffer from the problems of overly conservative decision-making or poor accuracy. Distributed robust optimization combines the advantages of robust optimization and stochastic optimization, providing an important reference for dealing with uncertainty factors in power distribution coordination systems.
[0005] (3) Existing reactive power optimization methods only consider economic factors, such as line loss costs and equipment operating costs, but lack comprehensive consideration of the risk costs brought about by the flexibility, reliability, economy and safety of power system operation.
[0006] In this context, the present invention proposes a robust integrated reactive power optimization method for main and distribution coordination considering conditional value at risk. The method aims to measure the operating risk of the system through CVaR (conditional value at risk), optimize resource allocation using the main and distribution coordination system, and combine distributed robust optimization to handle distributed PV and load uncertainties, so as to achieve a comprehensive improvement in the operational flexibility, reliability, economy and safety of the power system, and provide strong technical support for the stable and efficient operation of the new power system. Summary of the Invention
[0007] In order to overcome the problems in the prior art, the present invention proposes a robust integrated reactive power optimization method for main and distribution coordination considering conditional risk value.
[0008] The technical solution of the present invention to solve the above technical problems is as follows: The present invention provides a method for robust integrated reactive power optimization of main and distribution systems taking into account conditional risk value, comprising the following steps: By dividing the reconstruction levels, a hierarchical reconstruction model of main and auxiliary equipment coordinated operation is established to determine the optimal reconstruction area; The comprehensive norm is used to construct the uncertainty set of distributed PV output and load, and a hierarchically reconstructed main and distribution coordinated distributed blue stick reactive power optimization model is established; Based on the main and distribution coordinated robust reactive power optimization model, the main and distribution coordinated operation performance measurement index is introduced to establish a main and distribution coordinated robust comprehensive reactive power optimization model; The main and distribution coordinated robust integrated reactive power optimization model is solved to output the optimal reactive power optimization scheduling scheme.
[0009] Furthermore, the hierarchical reconstruction model for the coordinated operation of the main and auxiliary systems is established by reconstructing the hierarchical division, including: The objective function is established with the goal of minimizing the reconstruction cost, and the constraints are established by considering multi-level reconstruction constraints, common power flow constraints, distributed PV constraints, reactive compensation constraints and energy storage constraints.
[0010] Furthermore, the comprehensive norm is used to construct the uncertainty set of distributed PV output and load, and the hierarchically reconstructed main distribution coordinated distributed blue stick reactive power optimization model is established, which previously included: With the goal of minimizing switching costs and network losses, a main distribution coordinated reactive power optimization model is established by considering distributed PV constraints, reactive compensation constraints, energy storage constraints, and the main distribution coordinated optimal power flow constraints. K-means scenario clustering is used to obtain representative feature scenarios and their probability distributions. The 1-norm and ∞-norm are used to limit the local and global errors of the probability distribution, and the distributed PV output and load uncertainty set based on the comprehensive norm is obtained.
[0011] Furthermore, the main and distribution coordinated operation performance measurement indicators include the flexibility index, reliability index, economic index and safety index of the main and distribution coordinated operation system measured by the CVaR theory.
[0012] Furthermore, the main power distribution coordinated robust integrated reactive power optimization model is solved to output an optimal reactive power optimization scheduling scheme, including: The Gurobi solver is used to solve the main distribution coordinated robust integrated reactive power optimization model using the C&CG algorithm. When the difference between the upper and lower bounds of the C&CG algorithm is less than the given convergence accuracy, the optimal reactive power optimization scheduling scheme is output.
[0013] Compared with the prior art, the present invention has the following technical effects: (1) This paper proposes a hierarchical reconstruction strategy for the coordinated operation of the main and distribution systems by dividing the reconstruction levels, which can effectively accelerate the reconstruction speed of the main and distribution coordinated system and reduce the scheduling cost.
[0014] (2) The CVaR-based robust integrated reactive power optimization model for main and distribution system coordination proposed in this paper takes into account the flexibility, reliability, economy and safety of the main and distribution system operation. It can help managers flexibly adjust the operation strategy of the main and distribution system and achieve a balance between economy and operation risks.
[0015] (3) The proposed main-distribution coordinated distributed reactive power optimization model based on comprehensive norm considers both 1-norm and ∞-norm to constrain the probability distribution of distributed PV and load characteristic scenarios, improves the conservatism of the optimization results under the single norm constraint, and achieves a balance between economy and robustness to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0017] Figure 1 This is a flow chart of the main and distribution coordinated robust integrated reactive power optimization method considering CVaR of the present invention; Figure 2 A topological diagram of the test system provided by the present invention; Figure 3 This is a graph showing changes in comprehensive reactive power optimization results under different risk preference coefficients provided by the present invention. DETAILED DESCRIPTION
[0018] To further illustrate the technical means and effects adopted by the present invention to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation methods, structures, features, and effects of the technical solutions proposed by the present invention. Specific features, structures, or characteristics in one or more embodiments may be combined in any suitable form. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.
[0019] In one embodiment of the present invention, referring to Figure 1-Figure 3 , a robust integrated reactive power optimization method for main and distribution coordination considering conditional risk value is provided, which includes the following steps: By dividing the reconstruction levels, a hierarchical reconstruction model of main and auxiliary equipment coordinated operation is established to determine the optimal reconstruction area; The comprehensive norm is used to construct the uncertainty set of distributed PV output and load, and a hierarchically reconstructed main and distribution coordinated distributed blue stick reactive power optimization model is established; Based on the main and distribution coordinated robust reactive power optimization model, the main and distribution coordinated operation performance measurement index is introduced to establish a main and distribution coordinated robust comprehensive reactive power optimization model; The main and distribution coordinated robust integrated reactive power optimization model is solved to output the optimal reactive power optimization scheduling scheme.
[0020] The following is a detailed explanation of each of the above steps: Step 100: Initialization, input the parameters of each substation and corresponding distribution line, distributed PV output and load history data, distributed PV parameters, reactive compensation equipment parameters, energy storage parameters, and initial scenario probability distribution.
[0021] Step 200: By dividing the reconstruction levels, a hierarchical reconstruction model for coordinated operation of the main and distribution systems is established to determine the optimal reconstruction area; wherein, the hierarchical reconstruction model establishes an objective function with the goal of minimizing the reconstruction cost, and establishes constraint conditions by considering multi-level reconstruction constraints, general power flow constraints, distributed PV constraints, reactive power compensation constraints, and energy storage constraints.
[0022] The objective function is established with the goal of minimizing the reconstruction cost: ; Where, represents the total cost of hierarchical reconstruction; represents the reconfiguration cost coefficient of the feeder tie switch; represents the reconstruction cost coefficient of the transformer tie switch; represents the reconstruction cost coefficient of the substation tie switch; represents a collection of substations; Indicates substation The transformer collection in;T Indicates the total time of scheduling; express t Time Substation Lower transformer Feeder reconstruction regional state variables in the region; express t Time Substation The lower transformer reconstructs the regional state variables; express t Time Substation The reconstructed region state variables.
[0023] The constraint conditions established by considering multi-level reconstruction constraints, general power flow constraints, distributed PV constraints, reactive power compensation constraints, and energy storage constraints specifically include: The multi-level reconstruction constraints are: ; ; ; ; Where, express t Time Substation i The reconstructed region state variables; Indicates the number of transformers; express t Timeline l The switch status on Indicates a power distribution line; Represents the set of substation tie switches; represents the set of transformer tie switches; represents the feeder tie switch set; Represents a set of branch segment switches; Indicates the number of tie switches in the substation; Indicates the number of transformer tie switches; Indicates the number of feeder tie switches; Indicates the number of branch section switches.
[0024] The common power flow constraint is: ; ; Where, Representation node exist The merit of every moment, Representation node exist The reactive power of the moment; Represents a load node exist The merit of every moment; Represents a load node exist The reactive power of the moment; 、 For branch exist Active direction at the moment: express Active slave nodes at all times Flow Node , Represents a slave node Flow Node ; express Reactive slave node at all times Flow Node ; Indicates a gateway node exist The merit of every moment, Indicates a gateway node exist The reactive power of the moment; Representation node Distributed PV in The merit of every moment; Representation node Distributed PV in The reactive power of the moment; Representation node Energy storage in Charging power at the moment, Representation node Energy storage in Discharge power at the moment; Representation node Reactive power compensation device at The reactive power of the moment; Representation node The capacitor is The reactive power of the moment; Indicates the upper limit of the active power of the checkpoint. Indicates the reactive power upper limit of the checkpoint; Indicates a gateway node exist The merit of every moment, Indicates a gateway node exist The reactive power of the moment; Indicates a branch The switch state variable; Indicates a branch ij The lower limit of active power, Indicates a branch ij The lower limit of reactive power; Indicates a branch ij Active power upper limit; Indicates a branch ij The reactive power limit; and Decision nodes and A 0-1 auxiliary variable indicating whether it is a child node or a parent node; is the root node set; is a set of nodes; L A collection of branches.
[0025] The distributed PV constraints include: ; ; Where, Representation node i Distributed PV in t Reactive power generated at any moment; Representation node i Distributed PV in t Predicted power at the moment; Representation node i The capacity of the distributed PV.
[0026] The reactive power compensation constraints include: ; ; ; Where, Representation node i The reactive power compensation device can provide the minimum reactive power. Representation node i The reactive power compensation device can provide the maximum reactive power; Representation node i The upper capacitor t The number of compensation groups at the moment; Representation node i Single group compensation power of upper capacitor; Representation node i The number of groups of capacitors added; Representation node i The number of groups of upper capacitors is reduced; Indicates time T The maximum value of the change in the number of internal capacitor groups; Indicates the maximum number of compensation groups of capacitors.
[0027] The energy storage constraints include: ; ; ; ; Where, and Node i Energy storage in t The charge and discharge status at each moment; Representation node i At the upper limit of energy storage charging power, Representation node i The lower limit of the energy storage charging power; Representation node i At the upper limit of energy storage discharge power, Representation node i At the lower limit of energy storage discharge power; For nodes i Energy storage in t State of charge at the moment; For nodes i The capacity of the energy storage; Representation node i The upper limit of the state of charge of the energy storage, Representation node i The lower limit of the state of charge of the energy storage.
[0028] Step 300: Use the comprehensive norm to construct the uncertainty set of distributed PV output and load, and establish a hierarchically reconstructed main distribution coordinated distributed blue stick reactive power optimization model.
[0029] With the goal of minimizing switching operation cost and network loss, a main distribution coordinated reactive power optimization model is established by considering distributed PV constraints, reactive compensation constraints, energy storage constraints and main distribution coordinated optimal power flow constraints; the comprehensive norm is used to construct the uncertainty set of distributed PV output and load, and a main distribution coordinated distributed robust reactive power optimization model is established.
[0030] As an example, this step 300 includes the following steps: Step 310: With the goal of minimizing switching operation costs and network losses, a main distribution coordinated reactive power optimization model is established by considering distributed PV constraints, reactive power compensation constraints, energy storage constraints, and main distribution coordinated optimal power flow constraints.
[0031] The objective function is established with the goal of minimizing the switching action cost and network loss, specifically: ; ; ; Where, represents the cost of switching action; represents the network loss cost; represents the total cost of reactive power optimization; Indicates the operation cost coefficient of the substation tie switch; Indicates the operation cost coefficient of the transformer tie switch; Indicates the action cost coefficient of the feeder tie switch; Indicates the action cost coefficient of the section switch; is the network loss cost coefficient; for t Timeline ij The current value; For the line ij resistance value.
[0032] The distributed PV constraints, reactive power compensation constraints, and energy storage constraints are the same as those in step 200 and will not be described here. The improved main distribution coordinated optimal power flow constraints are: ; ; ; ;
[0033] ; ; Where, For the line ij The reactance value; for t Time Node i Voltage value; and are the lower and upper limits of voltage respectively; Auxiliary variables for establishing the relationship between voltage drop and line switch status; Auxiliary constant introduced for linearization.
[0034] Step 320: Use the comprehensive norm to construct the uncertainty set of distributed PV output and load, and establish the main distribution coordinated distributed blue rod reactive power optimization model.
[0035] Considering the high randomness of historical data on distributed PV and load, it is difficult to directly obtain an accurate scenario probability distribution from historical data. This paper first uses K-means scenario clustering to obtain representative feature scenarios and their probability distributions. The 1-norm and ∞-norm are then used to limit the local and global errors of the probability distribution, resulting in a distributed PV output and load uncertainty set based on the comprehensive norm: ; Where, Represents an uncertain set of feature scenarios constructed based on data-driven features; Indicates the number of feature scenes; Indicates the s The probability corresponding to each scenario; Indicates the s The positive offset state corresponding to the probability of each scene; Indicates the s The negative offset state corresponding to the scene probability; Indicates the s A positive offset for the probability of each scene; Indicates the s The negative offset of the probability of each scene; represents the maximum allowable deviation of probability under the 1-norm constraint; represents the maximum permissible deviation of probability under the ∞-norm constraint; Represents the clustering result s The initial probability of each feature scene; Indicates the number of historical data.
[0036] Based on the established distributed PV output and load uncertainty set, the charge and discharge state of the energy storage and the capacitor-related discrete variables are set as the first-stage variables according to the time scale of the equipment adjustment. , the remaining continuous variables are set as the second stage variables , establish the main distribution coordinated distributed blue rod reactive power optimization model: ; ; ; ; ; Where, is the set of variable constraints in Section 1; is the set of variable constraints in the second stage; 、 、 、 、 、 、 、 、 、 、 、 、 is a constant matrix; and Respectively for scenes s The first and second stage variables under ; It is the predicted output and load characteristic scenario data matrix of distributed PV.
[0037] Step 400: Based on the main distribution coordinated robust reactive power optimization model, the main distribution coordinated operation performance measurement index is introduced to establish a main distribution coordinated robust comprehensive reactive power optimization model considering CVaR.
[0038] As an example, this step 400 includes: Step 410: Based on the CVaR theory, the flexibility, reliability, economy and safety insufficiency comprehensive risk of the main and distribution coordinated operation system is measured, and flexibility index, reliability index, economy index and safety index are introduced.
[0039] Flexibility index: This invention uses the absorption rate of distributed PV to characterize the flexibility of the main and distribution coordinated operation system: ; Where, Indicates the distributed PV absorption rate of the main and distribution coordinated operation system.
[0040] Reliability index: The present invention uses the voltage deviation rate to characterize the reliability of the main and distribution coordinated operation system: ; Where, Indicates the voltage deviation rate of the main and distribution coordinated operation system; Indicates the rated voltage of the main and distribution coordinated operation system.
[0041] Economic performance index: This invention uses line loss rate to characterize the economic performance of the main and distribution coordinated operation system: ; Where, Indicates the line loss rate of the main and distribution coordinated operation system.
[0042] Safety index: This invention uses voltage fluctuation rate to characterize the safety of the main and distribution coordinated operation system: ; Where, Indicates the voltage fluctuation rate of the main and distribution coordinated operation system.
[0043] Step 420: Based on the flexibility index, reliability index, economic index and safety index, a comprehensive measurement index of the main and distribution coordinated operation system is established, and a main and distribution coordinated robust comprehensive reactive power optimization model considering CVaR is established.
[0044] Among the established indicators, the flexibility index is a positive indicator, and the larger its value, the better the flexibility; the other indicators are negative indicators, and the smaller the value, the better. In this regard, the present invention establishes a comprehensive measurement index of the main and distribution coordinated operation system based on the efficiency coefficient method: ; Where, Represents the comprehensive measurement index of the main and distribution coordinated operation system; represents the weight coefficient of the flexibility index, represents the weight coefficient of the reliability index, represents the weight coefficient of economic indicators, Represents the weight coefficient of the security index.
[0045] The present invention uses CVaR to measure the risk loss of the main and distribution coordinated operation system when the comprehensive performance of flexibility, reliability, economy and safety is insufficient, which is expressed as: ; in, Indicates risk of loss; Indicates boundary value; Indicates confidence.
[0046] ; The main distribution coordinated distributed robust reactive power optimization model established in step 300 is transformed into a main distribution coordinated comprehensive robust reactive power optimization model considering CVaR: ; ; ; Where, is the risk preference coefficient.
[0047] Step 500: Solve the main power distribution coordinated robust integrated reactive power optimization model and output the optimal reactive power optimization scheduling solution.
[0048] The main distribution coordinated robust integrated reactive power optimization model considering CVaR proposed in this invention is built in MATLAB, and the Gurobi solver is used to solve the model using the C&CG algorithm. When the difference between the upper and lower bounds of the C&CG algorithm is less than the given convergence accuracy, the optimal reactive power optimization scheduling scheme is output.
[0049] The C&CG algorithm decomposes the main and distribution coordinated robust integrated reactive power optimization model considering CVaR into a main problem and a sub-problem for iterative solution.
[0050] The main problem is to determine the optimal first-stage optimization variables when the probability distribution of the feature scene is known. , and provides a convergence lower bound for the C&CG algorithm: ; ; ; ; Where, k is the number of iterations of the algorithm; represents the objective function of the main problem; Indicates the k Iteration times scenario s The second stage variables below.
[0051] The subproblem is the optimization variable provided in the first stage of the main problem. Under the constraints of uncertainty set, we find the worst probability distribution of feature scene and provide the upper bound of convergence for C&CG algorithm. : ; ; In the above formula, Auxiliary variables representing the objective function of the subproblem; represents the objective function of the subproblem; Indicates the k+ 1 iteration s The probability corresponding to each scenario.
[0052] To verify the effectiveness of the hierarchical reconstruction strategy in the robust integrated reactive power optimization method for main and distribution coordination considering CVaR described in the present invention, Table 1 shows a comparison of the reactive power optimization results under the global reconstruction strategy and the hierarchical reconstruction strategy. It can be seen that compared with the global reconstruction strategy, the hierarchical reconstruction strategy proposed in the present invention reduces the reactive power optimization cost by 3.35% and speeds up the reconstruction by 21.78%.
[0053] Figure 3The relationship between the changes in the comprehensive reactive power optimization results under different risk preference coefficients is presented. It can be seen that as the risk preference coefficient increases, the operating cost of the main and distribution coordinated system continues to increase, while the CVaR value continues to decrease. This is because when the risk preference coefficient is small, the main and distribution coordinated system operates according to a risk-averse scheduling strategy, at which point managers expect high returns at the cost of high risk; when the risk preference coefficient is large, the system has higher operating costs and lower CVaR values, at which point managers expect a conservative operation and scheduling strategy. Therefore, by adjusting the risk preference coefficient, the scheduling strategy of the main and distribution coordinated system can be flexibly adjusted, guiding managers to better balance the system's operational risks and economic efficiency.
[0054] As can be seen from Tables 2 and 3, compared to establishing fuzzy sets using either the ∞-norm or the 1-norm alone, the combined norm constraint results in lower total operating costs and improved conservatism. Adjusting the confidence level of the ∞-norm and the 1-norm allows for flexible adjustment of decision conservatism, guiding managers to better balance the economy and robustness of system operation strategies.
[0055] Table 1
[0056] Table 2
[0057] Table 3
[0058] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.
Claims
1. A robust integrated reactive power optimization method for main and distribution systems considering conditional risk value, characterized in that: The following steps are involved: By dividing the reconstruction levels, a hierarchical reconstruction model of main and auxiliary equipment coordinated operation is established to determine the optimal reconstruction area; The uncertainty set of distributed PV output and load is constructed by using comprehensive norm, and a hierarchically reconstructed main and distribution coordinated distributed blue stick reactive power optimization model is established. Based on the main and distribution coordinated robust reactive power optimization model, the main and distribution coordinated operation performance measurement index is introduced to establish a main and distribution coordinated robust comprehensive reactive power optimization model; The main and distribution coordinated robust integrated reactive power optimization model is solved to output the optimal reactive power optimization scheduling scheme.
2. The robust integrated reactive power optimization method for main and distribution systems considering conditional risk value according to claim 1 is characterized in that: The hierarchical reconstruction model for the coordinated operation of the main and auxiliary components is established by reconstructing the hierarchical division, including: The objective function is established with the goal of minimizing the reconstruction cost, and the constraints are established by considering multi-level reconstruction constraints, common power flow constraints, distributed PV constraints, reactive compensation constraints and energy storage constraints.
3. The method for robust integrated reactive power optimization of main and distribution systems considering conditional risk value according to claim 2 is characterized in that: The comprehensive norm is used to construct the uncertainty set of distributed PV output and load, and a hierarchically reconstructed main and distribution coordinated distributed blue stick reactive power optimization model is established. The previous model includes: With the goal of minimizing switching costs and network losses, a main distribution coordinated reactive power optimization model is established by considering distributed PV constraints, reactive compensation constraints, energy storage constraints, and the main distribution coordinated optimal power flow constraints. K-means scenario clustering is used to obtain representative feature scenarios and their probability distributions. The 1-norm and ∞-norm are used to limit the local and global errors of the probability distribution, and the distributed PV output and load uncertainty set based on the comprehensive norm is obtained.
4. The robust integrated reactive power optimization method for main and distribution systems considering conditional risk value according to claim 2 is characterized in that: The main and auxiliary equipment coordinated operation performance measurement indicators include: flexibility indicators, reliability indicators, economic indicators and safety indicators of the main and auxiliary equipment coordinated operation system measured by CVaR theory.
5. The method for robust integrated reactive power optimization of main and distribution systems considering conditional risk value according to claim 2 is characterized in that: Solve the main power distribution coordinated robust integrated reactive power optimization model and output the optimal reactive power optimization scheduling scheme, including: The Gurobi solver is used to solve the main distribution coordinated robust integrated reactive power optimization model using the C&CG algorithm. When the difference between the upper and lower bounds of the C&CG algorithm is less than the given convergence accuracy, the optimal reactive power optimization scheduling scheme is output.