29 results about "Robust optimization problem" patented technology

The invention belongs to the field of intelligent power grid dispatching, and particularly discloses an equivalent cluster aggregation method considering the running state of distributed resources. According to the method, firstly, the parameters of the distributed resources are obtained, constraint modeling is carried out on the basis of an equivalent cluster aggregation model, the constraint modeling comprises the running state of the distributed resources, and then robust optimization modeling is carried out on the constraint result to obtain an aggregation equation; then, a robust optimization problem is converted into a solvable single form by replacing Lagrange relaxation and dual operation, and finally, an aggregation feasible region is calculated through an operation layer, so that the defect that an existing equivalent clustering method lacks a robust optimization and calculation mechanism adaptive to a distributed resource running state and network constraint is overcome; the calculation deviation of the aggregation feasible region is avoided, and the solving complexity is simplified, so that the accurate description of the aggregation feasible region is realized, and the equivalent cluster can more reliably respond to the scheduling requirement of the smart power grid.

The invention belongs to the technical field of power grid dispatching, and particularly discloses a dual-port equivalent cluster aggregation method. According to the method, an independent aggregation operation interval is established for two ports by solving a specific robust optimization problem, an initial aggressive boundary is formed, and an initial high-dimensional polyhedral model is established. The key improvement is that the condition extreme value of the power of the other port is solved under the constraint of the operation interval of one port, so that the physical coupling relation is converted into a series of accurate boundary points, and meanwhile, the calculation complexity is also reduced. And finally, cutting the initial and aggressive high-dimensional model by using the boundaries, eliminating an infeasible space generated by neglecting coupling, and quickly generating a feasible dual-port aggregation interval. The problems of large model error and low aggregation efficiency caused by dual-port interconnection are systematically solved.

According to the virtual energy storage equipment dispatching method and device, the storage medium and the computer equipment provided by the invention, when the virtual energy storage equipment is dispatched, the virtual battery model of each piece of equipment is constructed according to the physical parameters of each piece of virtual energy storage equipment, and the model is represented by a convex polytope in a semi-plane form, so that a power grid is more stable when facing load fluctuation; then, boundary summation is carried out on all the models to obtain an outer approximation model with aggregation flexibility, and therefore the rapid calculation characteristic of the NBM is reserved; then, a three-stage robust optimization problem is determined through a price guidance mechanism, and the three-stage robust optimization problem is converted into a mixed integer linear programming model by adopting a KKT optimization condition, so that the complexity of the problem is simplified, and the decomposability is realized while the problem is efficiently solved. And finally, optimizing and solving the external approximation model by using the mixed integer linear programming model to obtain a scheduling signal so as to perform parameter scheduling on each device, thereby effectively managing the energy demand and supply in the power grid.

This invention relates to intelligent transportation technology and aims to provide a data-driven robust site selection method for heterogeneous risk management in drone-assisted accident handling. The method includes: cleaning and preprocessing historical traffic incident data, extracting accident tags and performing quantitative classification based on the U-I-A multidimensional evaluation framework to obtain a structured dataset; decomposing uncertain demand into basic normal demand and multi-source heterogeneous fluctuations, and establishing a structured uncertainty set based on the quantitatively classified structured dataset; modeling the robust site selection planning model as a minimax robust optimization problem, and then equivalently reconstructing it into a solvable single-stage mixed integer linear programming model; using a solver to solve the problem, obtaining the globally optimal hangar site selection scheme and the number of supporting drones, and outputting the results. This site selection method enhances the reliability and robustness of the drone emergency response system; it enables scientific planning of drone site deployment, achieving rapid and stable emergency response.

The invention discloses a non-Gaussian random process prediction optimization control method based on a Wasserstein fuzzy set, and belongs to the technical field of blast furnace ironmaking system control. Aiming at the problems of modeling uncertainty and insufficient robustness under external disturbance and non-Gaussian disturbance existing in a traditional method, the method comprises the following steps: setting a coal injection amount, an oxygen-enriched flow, a cold air flow, a molten iron temperature and a Si content by collecting a pressure difference of a blast furnace ironmaking system; a blast furnace iron-making system is used for building a blast furnace neural network prediction model, output in a period of time in the future is predicted based on the current blast furnace state and the neural network model, a predicted output error is regarded as a random variable, and a Wasserstein fuzzy set with empirical distribution as the center is built to describe distribution uncertainty of the Wasserstein fuzzy set; in an MPC rolling optimization process, Wasserstein distance constraint is introduced, a distribution robust optimization problem is converted into a solvable optimization problem, and a group of optimal control sequences are obtained, so that a system does not depend on a specific random distribution hypothesis, and modeling errors and non-Gaussian external disturbance can be effectively processed.

δThe invention discloses a virtual unit model construction method considering a power supply type distributed energy cluster space-time coupling relation, and relates to the field of electrical engineering, and the method comprises the following steps: carrying out the modeling of a virtual unit based on the dispatching power y of a superior power grid, obtaining a virtual unit parameter u, and enabling the constraint Uy to be smaller than or equal to u; the method comprises the following steps: modeling a variable x in a power supply type distributed energy cluster and a superior power grid dispatching power correction value delta based on a power distribution network framework to obtain a constraint psi x < = psi, setting target functions as the sum of all time periods and the weighted sum of unit parameters u, and decomposing the problem into a two-stage adaptive robust optimization problem; the sub-problem is set to solve the maximum power correction under the current virtual unit parameters, and the main problem is set to solve the current unit parameter weighted sum according to the maximum power error; and iteratively solving the main problem and the sub-problems until the solving result of the sub-problems is 0, and obtaining parameters of the virtual unit model. Experimental results prove that the method has the advantages of being high in solving speed, good in solving effect and suitable for participating in power system dispatching.

The invention discloses a multi-terminal flexible DC power grid distribution robust damping optimization method considering random fluctuation. The method comprises the following steps: firstly, building a second-order approximation model of a damping ratio about control parameters and wind power around a dominant oscillation mode; secondly, a Gaussian mixture model is adopted to describe statistical characteristics of wind power prediction errors, and a Wasserstein distance-based distribution uncertainty set is constructed around empirical distribution; expressing a'minimum damping lifting / constraint 'target as a distributed robust optimization problem, and converting the worst expected target and constraint into deterministic convex optimization by using GMM analysis statistics and quadratic form properties; and finally, the optimal parameters of the additional damper are obtained and issued for application. According to the method, robust setting is performed on any approximate distribution under the condition of not depending on accurate prior distribution, the critical modal damping can be remarkably improved, the probability border crossing risk can be inhibited, and the method has the advantages of being low in calculation overhead and robust in setting effect.

The invention belongs to the technical field of underwater positioning, and discloses an underwater robot positioning method and system based on a self-adaptive robust loss function. According to the method, based on an underwater acoustic signal two-way propagation time observation equation, the position and sound velocity of an underwater robot (AUV) are used as unknown quantities, and signal two-way propagation time is used as an error variable of an observed quantity; an adaptive robust loss function is introduced, a robust optimization problem taking the adaptive robust loss function as an optimization target is constructed, a progressive non-convexity method (GNC) is adopted to solve the robust optimization problem, a sequential updating method of the position and the sound velocity of the AUV is provided, sequential estimation of the position and the sound velocity of the AUV is carried out, and the position and the sound velocity of the AUV are estimated. And deriving a closed-form estimation result of the position of the AUV under the condition that the sound velocity is determined and a closed-form estimation result of the sound velocity under the condition that the position of the AUV is determined. According to the method, robust stable estimation of the position and the sound velocity of the AUV under the outlier interference condition is realized.

The invention belongs to the technical field of micro-grid two-stage robust scheduling problems, and particularly relates to a micro-grid robust scheduling problem solving method without dual transformation. Comprising the following steps: converting a two-stage micro-grid robust scheduling model into a main model with the same constraint condition and decision variable set; constructing an auxiliary model of the deterministic optimization scheduling main model, wherein the sum of related Boolean variables representing uncertainty in the auxiliary model and the main model is equal to 1; based on the benefit conflict between the main model and the auxiliary model, searching an optimal scene through the auxiliary model in a leading edge iteration manner, and screening the worst scene from the corresponding severe scenes by the main model; the algorithm provided by the invention can effectively identify the worst operation scene of the robust optimization problem.

The invention relates to the technical field of electric power system operation and optimization, in particular to a wind-thermal bundling system adjustable margin prediction method adapting to uncertain working conditions, which comprises the following steps of: 1, establishing a power model; 2, parameterized expression is carried out; step 3, establishing a robust optimization problem; 4, determining candidate upper and lower bounds; step 5, solving a sub-problem; 6, determining an interface power boundary; 7, outputting a prediction result; according to the method, a multi-period power model of a delivery interface point is established for a wind power and thermal power bundled delivery scene, wind power volatility, thermal power climbing / minimum output and other adjustment constraints are uniformly introduced into an interface adjustable margin for description, and a unified aggregation model of wind power-thermal power-energy storage and delivery point interface power is established, so that the regulation constraints of the delivery interface point are optimized. The upper and lower limits of thermal power output, the climbing constraint and other mechanism conditions are explicitly incorporated, so that the predicted interface power adjustable interval has decomposable and executable scheduling significance, and meanwhile, the interface power adjustable interval is lighter and supports the reduction of control complexity.

A control apparatus according to an embodiment is a control apparatus that embeds a virtual network for implementing provision of a service on a physical network, the control apparatus including: a first acquisition unit that acquires a predicted value of an amount of traffic of the service and a predicted value of an amount of power including renewable energy usable by each of physical nodes constituting the physical network; a second acquisition unit that acquires information regarding the physical network; a solution calculation unit that calculates an optimal solution of a 2-stage robust optimization problem regarding allocation of virtual nodes constituting the virtual network to the physical nodes and route determination between the virtual nodes to minimize a link congestion rate representing a maximum value of a utilization rate of a link constituting the physical network and a node power congestion rate representing a maximum value of a power consumption rate of a node constituting the physical network on the basis of the predicted value of the amount of traffic, the predicted value of the amount of power, and the information regarding the physical network; and a control unit that controls the virtual network embedded in the physical network on the basis of the allocation of the virtual nodes and the route determination that are represented by the optimal solution.

The invention relates to a robust optimization algorithm-based power grid emergency material scheduling method under rain and snow freezing disasters, which comprises the following steps of: acquiring meteorological data and traffic facility information under the rain and snow freezing disasters, and constructing a line fault probability prediction model; monte Carlo simulation is adopted to generate disaster scenes, and a typical disaster scene set is obtained through a scene reduction technology; constructing a joint scheduling optimization mathematical model considering emergency material allocation and emergency equipment allocation, wherein the joint scheduling optimization mathematical model comprises a pre-disaster pre-deployment stage and an in-disaster dynamic response stage; solving a two-stage robust optimization problem by adopting an improved column and constraint generation algorithm, gradually converging to a robust optimal solution, outputting a scheduling scheme of an optimal emergency material and an emergency repair team, determining a repair target and an optimal driving path of the repair team, and generating a collaborative recovery strategy of a post-disaster power grid and a traffic system; and rapid recovery after disasters and reasonable configuration of emergency resources are realized, so that the emergency response speed and recovery capability of the power grid under rain, snow and freezing disasters are improved.

The power system real-time scheduling method and system considering energy storage differentiated response comprise: updating wind power and photovoltaic output and load prediction data every preset time interval in a scheduling period; a robust optimization model containing new energy prediction error uncertainty and energy storage differentiated response is established, including constructing an objective function, robust operation constraints of different types of energy storage power stations, system power balance constraints and line transmission capacity constraints; a robust dual method is used to convert the robust optimization problem containing interval uncertainty into a deterministic mixed integer linear programming problem; according to the wind power and photovoltaic output and load prediction data in each scheduling period, the robust optimization model of the power system real-time scheduling is solved every other scheduling period, and the real-time power instructions of two types of energy storage and conventional units are output. The present application can fully exert the regulation flexibility and comprehensive benefits of different types of energy storage power stations, can adapt to different regional new energy fluctuation characteristics, and can improve the new energy consumption capacity.

The invention belongs to the technical field of UAV-assisted MEC. The invention provides a joint optimization method for auxiliary edge calculation of an anti-interference unmanned aerial vehicle under energy constraint, and the method comprises the steps: constructing a system model comprising a ground terminal, the unmanned aerial vehicle and an interference source, and determining the position of each node in a time slot and a trajectory variable of the unmanned aerial vehicle; calculating a channel gain estimation value based on a line-of-sight propagation model, and introducing an imperfect channel state information error boundary; the method comprises the following steps: constructing a joint optimization problem taking maximization of an uplink average unloading rate as a target by combining propulsion energy consumption of an unmanned aerial vehicle and an interference source energy constraint, converting the problem into a robust optimization problem through a worst case criterion, and decomposing the problem into two sub-problems of scheduling power optimization and trajectory optimization; and carrying out iterative solution by utilizing an alternating optimization strategy and a continuous convex approximation technology until the optimal unloading scheduling, transmitting power and flight path are obtained through convergence. According to the method, the joint optimization problem of resource limitation in a complex interference environment is effectively solved, and the system communication performance is remarkably improved.

The invention relates to the technical field of power system planning, and discloses a distributed robust optimization-based source-grid-load-storage collaborative planning method, which comprises the following steps of: constructing a probability distribution fuzzy set based on a Wasserstein distance to describe the uncertainty of wind power output, photovoltaic output and load demand in a power system; based on the probability distribution fuzzy set based on the Wasserstein distance, a source network load storage collaborative planning model is constructed, and the source network load storage collaborative planning model takes minimization of the sum of investment cost and expected operation cost as a target function and comprises a plurality of system operation constraints; converting the source network load storage collaborative programming model from a two-stage distribution robust optimization problem into a mixed integer linear programming problem by using a duality theory; and solving the mixed integer linear programming problem to obtain an optimal scheme of power grid capacity expansion, energy storage configuration and demand response configuration. According to the distributed robust optimization-based source network load storage integrated planning method, collaborative optimization of each link and effective processing of uncertainty are realized.

The invention belongs to the technical field of wireless communication and edge computing, and discloses an unmanned aerial vehicle mobile edge computing system time delay optimization method based on reconfigurable intelligent surface assistance in a non-ideal environment, and the system comprises a UAV carrying an MEC server, an RIS equipped with a plurality of passive reflection units, and a plurality of Internet of Things (IoT) devices. And a wireless information and energy simultaneous transfer (SWIPT) technology is integrated to support dynamic switching between a power division (PS) mode and a time switching (TS) mode. Aiming at multiple reality constraints such as non-ideal channel state information (CSI), transceiver hardware damage, RIS discrete phase offset and nonlinear energy collection, the method provided by the invention takes minimization of total task processing delay of scheduled IoT equipment as a target to establish a worst-case robust optimization problem; the method comprises the following steps: jointly optimizing an unmanned aerial vehicle receiving beam forming vector, an RIS phase shift matrix, an SWIPT mode parameter and a task division ratio through an alternating optimization (AO) framework, and dynamically selecting a PS or TS mode as an optimal working mode based on an optimization result.

The invention discloses an automatic driving behavior decision robust optimization method guided by a long-period historical track, and belongs to the technical field of automatic driving. The method comprises the following steps: setting the number of historical decision trajectory windows, storing historical decision trajectory points of an automatic driving vehicle, and constructing a historical decision trajectory probability distribution model; obtaining a track point corresponding to the current decision, performing coordinate transformation on the track point corresponding to the current decision to obtain a space coordinate under a unified coordinate system, and constructing a probability distribution model of the track point corresponding to the current decision; calculating decision similarity measurement guided by historical decisions; and designing a behavior decision objective function according to decision similarity measurement to solve a robust optimization problem. According to the method, the problems of decision jitter and strategy abrupt change caused by dependence on single-frame observation or short-term trajectory information in a traditional method are avoided, the decision robustness, stability and reliability of an automatic driving system in a dynamic complex traffic environment are improved, and the risk of collision with surrounding traffic participants is reduced.

This application belongs to the field of smart grid dispatching, and specifically discloses an equivalent cluster aggregation method that takes into account the operating state of distributed resources. This application first obtains the parameters of the distributed resources and performs constraint modeling based on the equivalent cluster aggregation model, including the operating state of the distributed resources. Then, it performs robust optimization modeling on the constraint results to obtain its aggregation formula. Next, by replacing Lagrange relaxation and dual operations, the robust optimization problem is transformed into a solvable single form. Finally, the aggregated feasible region is calculated through a computational layer. This solves the shortcomings of existing equivalent cluster methods that lack robust optimization and solution mechanisms adapted to the operating state of distributed resources and network constraints, avoids calculation bias in the aggregated feasible region, and simplifies the solution complexity, thereby achieving an accurate characterization of the aggregated feasible region and enabling the equivalent cluster to respond more reliably to the dispatching needs of the smart grid.

The invention relates to the technical field of power systems, and discloses a zero-carbon building robust planning method considering a plant carbon sink mechanism, which comprises the following steps of: establishing a mathematical model, and quantifying the energy conversion efficiency, the carbon absorption capacity and the dynamic operation constraint of the mathematical model; based on the historical load, the photovoltaic data and the mathematical model, constructing a deterministic planning model, and coupling electricity-hydrogen-carbon balance constraints; introducing an interval uncertainty set to describe the random fluctuation of photovoltaic output, extending a deterministic planning model into a robust optimization problem, and controlling the conservative degree by adjusting a parameter gamma to form robust planning schemes with different anti-disturbance capabilities; and solving a robust planning scheme by using GUROBI in an MTALAB environment, and calculating configuration results under different gamma values. According to the zero-carbon building robust planning method considering the plant carbon sink mechanism, the reliability and economical efficiency of the system are balanced by constructing the interval uncertainty set, and the carbon neutralization measurement and calculation precision is improved by introducing the plant photosynthesis dynamic model, so that the building micro-grid has more comprehensive carbon neutralization capability.

The application belongs to the technical field of power grid dispatching, and specifically discloses a multi-port power grid aggregation method based on a cut plane projection. The application obtains constraints by modeling a multi-port power grid based on distributed energy and solves a robust optimization problem to obtain each port independent aggregation interval, which is integrated into an initial aggregation space. Then, a cut plane problem is solved to identify an infeasible region and obtain a cut plane constraint and a worst dispatch instruction. Subsequently, a feasible dispatch instruction closest to the worst dispatch instruction in the current aggregation space is found according to the cut plane constraint, and the core function of the feasible dispatch instruction is to provide a reference point for each time interval projection. The cut plane constraint is projected to each time interval space with the reference point, the aggregation space is cut and updated, and the infeasible region is gradually cut off through iteration of the process. This gradually and accurately depicts the interconnection characteristics between multi-ports, solves the aggregation difficulty problem caused by the lack of accurate description means in the prior art, and improves the aggregation reliability compared with the prior art.

The invention discloses a passive body area network communication perception energy transfer resource allocation method based on a DRO algorithm, and relates to the field of passive body area networks, and the method comprises the steps: constructing a passive body area network system which comprises a hybrid access node, user receiving equipment and a plurality of passive label nodes which are connected with one another; constructing a communication perception energy transfer fusion model based on a passive body area network system; establishing a distributed robust optimization problem model based on the fusion model by taking maximization of communication, perception and energy transfer indexes as targets; utilizing a conditional risk value algorithm to convert opportunity constraint conditions in the distributed robust optimization problem model; according to the CVaR distributed robust optimization method, through the CVaR robust optimization algorithm and the alternating optimization algorithm, the resource allocation problem of the passive body area network under channel uncertainty is solved, and cooperative guarantee of multiple performance indexes and improvement of the overall efficiency of the system are achieved.

The invention provides an incomplete information constraint uncertain game strategy optimization method and system, and the method comprises the steps: building a nonlinear uncertain differential game process model according to the incomplete information characteristics of a nonlinear uncertain game dynamic process in a strong confrontation scene, and enabling the strategy amplitudes and game states of two parties to be constrained; taking an opponent game strategy and an unknown uncertain item as a disturbance item, introducing a non-quadratic form energy function, a control barrier function and a disturbance function upper bound on the basis of describing a nonlinear disturbance system of a nonlinear uncertain game dynamic process, and constructing a performance index function; converting a safety robust game problem of a nonlinear uncertain dynamic game process into a nonlinear nominal system constraint robust optimization problem; and introducing an event-triggered security robust reinforcement learning technology, and completing approximate solution of an HJB equation by evaluating an online approximation optimal performance index function of a neural network to obtain an own optimal game strategy. According to the method, the uncertain game strategy can be optimized.

This invention discloses a secure covert communication method based on cognitive radio, investigating a delay-tolerant relay covert unmanned aerial vehicle (UAV) and a cognitive radio framework involving multiple colluding eavesdroppers and listeners. In this framework, a legitimate UAV acts as an airborne relay, enabling communication when the direct link between the ground transmitter and receiver is blocked. Subsequently, considering the uncertainty of the locations of multiple eavesdropping and listening nodes, a robust optimization problem is constructed. By jointly optimizing the UAV's trajectory and power, and the transmitter's transmission power, a tractable version is constructed using Pinsk inequalities, Jensen inequalities, and a binary search method for extremely complex covert constraints. Following this, an algorithm based on alternating optimization is proposed to solve the optimization problem. Achieving low complexity, algorithms based on primal-dual search and continuous convex approximation are designed for each subproblem. Numerical results demonstrate the effectiveness of our proposed algorithm.

The invention discloses a robust energy-saving method for auxiliary simultaneous information and energy transfer of an active intelligent surface. The method comprises the following steps: establishing a nonlinear energy collection model and a statistical channel error model of an active intelligent surface link; constructing a robust optimization problem including an outage probability constraint and an active intelligent surface power constraint by taking minimization of base station transmitting power as a target; an improved optimization problem is established; an alternative optimization framework is adopted, and the improved optimization problem is decomposed into two sub-problems of base station transmitted beam forming optimization and active intelligent surface coefficient optimization; and alternately updating the two sub-optimization problems until the objective function converges to obtain a solution of the optimization problem. According to the method, random channel state errors and nonlinear circuit characteristics in an actual system are comprehensively considered, on the premise that the achievable rate of an information decoder, the outage probability of the receiving power of an energy receiver and the active intelligent surface power are met, the transmitting power of a base station is remarkably reduced, the network energy efficiency is improved, and good industrial practicability is achieved.

This application belongs to the field of power grid dispatching technology, specifically disclosing a two-port equivalent cluster aggregation method. This application establishes independent aggregation operating intervals for the two ports by solving a specific robust optimization problem, forming initial aggressive boundaries and building an initial high-dimensional polyhedral model. The key improvement lies in subsequently solving the conditional extremum of the power of the other port under the constraints of the operating interval of one port, thereby transforming the physical coupling relationship into a series of precise boundary points and reducing computational complexity. Finally, these boundaries are used to cut the initial, aggressive high-dimensional model, eliminating infeasible spaces caused by neglecting coupling, and quickly generating feasible two-port aggregation intervals. This systematically solves the problems of large model errors and low aggregation efficiency caused by two-port interconnection.

The invention relates to the technical field of supply chain management and artificial intelligence, and discloses a supply chain intelligent goods sorting method based on deep learning, comprising the following steps: S1, acquiring supply chain data, processing the supply chain data, and inputting the processed supply chain data into a preset deep learning model; s2, the supply chain data is processed by the deep learning model to predict and output a risk perception potential vector, and the risk perception potential vector comprises an expected marginal value corresponding to the distribution of a preset commodity to a target location and prediction uncertainty; s3, constructing and solving a robust optimization problem based on the output risk perception potential vector and a preset business constraint, so as to generate a goods distribution scheme; and S4, determining system loss according to a business result generated by the goods distribution scheme. According to the invention, end-to-end learning and robust optimization technologies are adopted, bottleneck diagnosis is carried out by using gradients, and the adaptability and accurate insight of goods sorting decision making are improved.

The invention discloses a data-driven virtual power plant resource aggregation method fusing physical information. The method comprises the following steps: establishing a mathematical model of distributed resources of a virtual power plant; training a physical information generation diffusion model through a data driving and physical rule fusion mode to obtain a new energy processing mean value and variance, and establishing a deterministic linear constraint of new energy output; the calculation complexity of high-dimensional distributed resource feasible region aggregation is considered, efficient aggregation is carried out on various resource feasible regions on the basis of the Cinonol polyhedron theory, and an aggregated polyhedron is equivalently converted into a semi-space constraint form; taking the deterministic boundary as a support set of a distributed robust optimization model, and constructing an optimization scheduling model for the virtual power plant to participate in system cooperative operation; and converting the distribution robust optimization problem into a solvable single-layer deterministic mathematical model, and solving the solvable single-layer deterministic mathematical model. According to the method, the new energy output rationality is ensured, the aggregation precision and the calculation efficiency are effectively balanced, and the accuracy and solvability of virtual power plant resource aggregation are improved.

The application discloses a hydrogen-containing mixed energy supply infrastructure planning method, and relates to the field of hydrogen energy, and comprises the following steps: step 1, three-network coupling modeling of an electric network, a hydrogen network and a traffic network: coupling network physical scene construction and optimization target establishment, traffic network constraint construction, hydrogen network constraint construction, electric network constraint construction and coupling constraint construction; step 2, decision dependence distribution robust optimization model construction: hydrogen demand decision dependence fuzzy set construction and distribution robust optimization framework construction; step 3, distribution robust optimization problem reconstruction and solving method: decision dependence scene probability construction, norm item linearization and opportunity constraint reconstruction; and step 4, three-network coupling system example simulation analysis. The application effectively solves the problem of whether a hydrogen vehicle or a hydrogen refueling station is first developed when hydrogen fuel vehicles have not been popularized, realizes reasonable planning of hydrogen infrastructure, reduces investment and operation costs, improves new energy consumption of an electric network and effectively avoids energy waste in the early development of hydrogen vehicles.

The invention belongs to the technical field of power grid dispatching, and particularly discloses a multi-port power grid aggregation method based on cut plane projection. According to the method, the constraint is obtained by modeling the multi-port power grid based on the distributed energy, the robust optimization problem is solved to obtain the independent aggregation interval of each port, and the independent aggregation interval is integrated into the initial aggregation space. And solving a cutting plane problem, identifying an infeasible region, and obtaining a cutting plane constraint and a worst scheduling instruction. Then, according to these, a feasible scheduling instruction closest to the worst scheduling instruction in the current aggregation space is found, and the core function of the feasible scheduling instruction is to provide a reference point for the projection of each time period; and projecting a cutting plane constraint to each time period space by using the reference point, cutting and updating an aggregation space, and gradually cutting off an infeasible region by iterating the process. The interconnection characteristics among multiple ports are gradually and accurately depicted, the problem of difficult aggregation caused by lack of accurate depicting means in the prior art is solved, and compared with the prior art, the aggregation reliability is improved.