A robust optimization-based intelligent charging and discharging method and system for electric vehicles

Abstract

This invention relates to the field of electric vehicle charging control technology, and discloses a robust optimization-based intelligent charging and discharging method and system for electric vehicles. It aims to solve the problems of traditional deterministic optimization methods struggling to cope with photovoltaic output, load demand, and electricity price fluctuations, while also failing to consider battery life and economic efficiency. The method includes: establishing a deterministic charging optimization model with the objective of minimizing grid power purchase costs and battery depreciation costs, incorporating constraints such as battery state of charge dynamics and operating modes; constructing a robust optimization model based on information gap decision theory, defining robust boundaries for load, photovoltaic power, and electricity price, and constructing a worst-case scenario; obtaining the Pareto optimal solution set using a multi-objective particle swarm optimization algorithm; and selecting the solution with the lowest satisfaction level and the highest satisfaction level as the final strategy through fuzzy satisfaction assessment. This invention can achieve economical, reliable, and battery-friendly charging and discharging scheduling under uncertain environments, adapting to smart microgrids and V2G scenarios, and improving overall operational performance.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle charging control technology, specifically to a robust optimization-based intelligent charging and discharging method and system for electric vehicles. Background Technology

[0002] With the popularization of electric vehicles and the development of Vehicle-to-Grid (V2G) technology, electric vehicles can not only serve as electrical loads but also participate in grid dispatch as distributed energy storage units. However, the charging behavior of electric vehicles is affected by multiple factors such as user driving habits, fluctuations in photovoltaic power generation, changes in grid load, and uncertainty in electricity prices. Traditional deterministic optimization methods are difficult to guarantee the economy and robustness of the system under uncertain environments.

[0003] Existing charging strategies are mostly based on deterministic assumptions, ignoring the random fluctuations in photovoltaic output, load demand, and electricity prices during actual operation. This can lead to the failure of optimization results or a surge in costs in practical applications. In addition, frequent charge-discharge switching accelerates battery aging and increases depreciation costs, while existing methods often fail to consider battery lifespan and economic efficiency in a unified manner.

[0004] Therefore, there is an urgent need for an intelligent charging and discharging method for electric vehicles that can balance economy, equipment lifespan, and operational robustness in uncertain environments. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a robust optimization-based intelligent charging and discharging method and system for electric vehicles, aiming to solve the problems in the background technology.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a robust optimization-based intelligent charging and discharging method for electric vehicles, comprising the following steps:

[0007] Step S1: Establish a deterministic charging optimization model: with the goal of minimizing grid power purchase costs and battery depreciation costs, while incorporating dynamic constraints on battery state of charge, operating mode constraints, user-specified state of charge boundary constraints, equipment safety operation constraints, and power balance constraints; the equipment safety operation constraints include charging and discharging power limits.

[0008] Step S2: Based on the deterministic charging optimization model, construct a robust optimization model of information gap decision theory: set the load, photovoltaic, and electricity price as uncertain variables and define robust boundaries, construct the worst scenario, and substitute it into the constraint update in step S1;

[0009] Step S3: Solve the robust optimization model using the multi-objective particle swarm optimization algorithm: initialize particles according to the charging and discharging power limit and time window, evaluate the robustness of particles in the worst scenario during iteration, and obtain the Pareto optimal solution set by screening non-dominated solutions.

[0010] Step S4: Select a strategy based on the optimal solution set using fuzzy satisfaction: Calculate the satisfaction of each solution under the three robust objectives, take the lowest satisfaction of each solution, and select the solution with the largest lowest satisfaction as the final charging and discharging strategy.

[0011] Furthermore, the specific process of step S1 is as follows:

[0012] Step S1.1: Construct the economic objective function; the economic objective function aims to minimize the total lifecycle cost of the deterministic charging optimization model. The total cost includes two parts: first, the cost of purchasing electricity from the main grid at different times of the day; second, the battery depreciation cost incurred by the electric vehicle during its connection to the charging station due to charge-discharge cycles. Here, a negative value for the exchange power between the electric vehicle and the microgrid represents charging, a positive value represents discharging, and zero represents standby. Economic objective function Represented as:

[0013] ;

[0014] In the formula, express Power purchased from the main grid at any given time; express Real-time electricity pricing; This represents the battery discount cost coefficient; express The exchange power between electric vehicles and microgrids at any given time; , These represent the times when the electric vehicle arrives at and leaves the charging station, respectively.

[0015] Step S1.2: Supplement the constraints of the deterministic charging optimization model, including dynamic constraints on battery state of charge, operating mode constraints, user-specified state of charge boundary constraints, equipment safe operation constraints, and power balance constraints.

[0016] Furthermore, the dynamic constraint of the battery state of charge, taking into account energy conversion losses, is expressed as follows:

[0017] ;

[0018] In the formula, express The state of charge of the electric vehicle battery at all times; express The state of charge of the electric vehicle battery at all times; Indicates the time interval for scheduling; Indicates the rated capacity of the electric vehicle battery; Represents the time-varying efficiency function;

[0019] The operation mode constraint that limits the number of state transitions is represented as follows:

[0020] ;

[0021] In the formula, express The binary state variable at time t. Represents the charge / discharge activation state. Indicates standby status; express The binary state variable at any given time; Indicates the maximum allowed number of state transitions;

[0022] User-specified charge state boundary constraints are represented as follows:

[0023] ;

[0024] In the formula, , These represent the user-specified states of charge upon arrival and departure, respectively. express The time is less than or equal to the time the electric vehicle arrives at the charging station. The state of charge; express The time is greater than or equal to the time when the electric vehicle leaves the charging station The state of charge;

[0025] Equipment safety operation constraints are expressed as follows:

[0026] ;

[0027] In the formula, , These represent the minimum and maximum charging and discharging power limits for electric vehicles, respectively. , These represent the minimum and maximum permissible states of charge set to protect battery life, respectively.

[0028] The power balance constraint is expressed as:

[0029] ;

[0030] In the formula, express Total load power at any given time; express Photovoltaic power generation at any given time.

[0031] Furthermore, the specific process of step S2 is as follows:

[0032] Step S2.1: Define the set of uncertain variables as... The variables in the set of uncertain variables include load power. Photovoltaic power generation and electricity price Define robust boundary values ​​for each variable in the set of uncertain variables, including robust boundary values ​​for load power. Robust boundary values ​​of photovoltaic power generation and electricity price robust boundary value Robust boundary values ​​represent the maximum permissible fluctuation of each variable relative to its predicted value;

[0033] Step S2.2: Set the objective of the robust optimization model of the information gap decision theory to maximize the robust boundary, i.e. ;

[0034] Step S2.3: Construct the worst-case uncertainty scenario: For the load power that increases cost, the worst-case realization value is For electricity prices that increase costs, the worst-case scenario is... For cost-reducing photovoltaic power generation, the worst-case scenario is... ; This represents the predicted load power. This represents the electricity price forecast. This represents the predicted value of photovoltaic power generation.

[0035] Step S2.4: Substitute the worst-case uncertainty scenario into the constraints of the deterministic charging optimization model to update and form the constraints of the robust optimization model of information gap decision theory; at the same time, retain the dynamic constraints of battery state of charge, operating mode constraints, user-specified state of charge boundary constraints, and equipment safe operation constraints, and set robust performance indicators.

[0036] Furthermore, the robustness performance index is expressed as: , Let represent the objective function of the robust optimization model of the information gap decision theory, i.e., the electricity purchase cost considering the worst-case uncertainty. This represents the parameters used to balance robustness and economy;

[0037] The objective function of the robust optimization model of information gap decision theory Represented as:

[0038] .

[0039] Furthermore, by substituting the worst-case uncertainty scenario into the constraints of the deterministic charging optimization model, the power balance constraint is updated as follows:

[0040] .

[0041] Furthermore, the specific process of step S3 is as follows:

[0042] Step S3.1, Population Initialization: Based on the minimum and maximum charge / discharge power limits of the electric vehicle and the times when the electric vehicle arrives at and leaves the charging station, randomly generate an initial particle swarm representing the all-weather charge / discharge power curve of the electric vehicle, and ensure that the particles satisfy... ;

[0043] Step S3.2, Particle Evaluation: Decode the position vector of each particle into photovoltaic power generation. Substituting the time series data into the robust optimization model of information gap decision theory, we calculate the three objective function values, i.e., the robust objective function values, under the worst-case uncertainty scenario, including the load power robust boundary value. Robust boundary values ​​of photovoltaic power generation and electricity price robust boundary value ;

[0044] Step S3.3, Non-dominated solution screening and storage: Based on the three objective function values ​​obtained in step S3.2, non-dominated solutions in the particle population are screened according to the Pareto dominance relation and stored in an external storage library;

[0045] Step S3.4, Adaptive Leader Selection: Using a grid-based density estimation method, a "leader" is probabilistically selected for each particle from an external storage library;

[0046] Step S3.5: Particle position and velocity update;

[0047] Step S3.6, Constraint Processing: Correct the updated particle positions so that the corrected particles satisfy... Robustness boundaries for load power, photovoltaic power generation, and electricity price;

[0048] Step S3.7, Storage Maintenance and Iteration: After each iteration, the non-dominated solutions in the current population are merged into the external storage and reordered. If the external storage exceeds the preset capacity, the non-dominated solutions in the densely populated region are removed. Repeat steps S3.2-S3.6 until the maximum number of iterations is reached. At this point, the solution set in the external storage is the Pareto optimal solution set of the robust optimization model.

[0049] Furthermore, the specific process of step S4 is as follows:

[0050] Step S4.1: Clarify the relationship between the decision objective and the Pareto optimal solution set: In the Pareto optimal solution set, each non-dominated solution... This includes the charge / discharge power plan and the corresponding three robust objective function values; the three robust objective function values ​​are: non-dominated solutions. Corresponding load power robust boundary Non-dominated solutions Corresponding photovoltaic power robust boundary Non-dominated solutions Corresponding robust boundary of electricity price ; Indicates non-dominated solutions The corresponding load power robustness boundary value, Indicates non-dominated solutions The corresponding robust boundary value for photovoltaic power, Indicates non-dominated solutions The corresponding robust boundary value for electricity prices;

[0051] Step S4.2: Calculate the satisfaction membership degree of each robustness objective: for each non-dominated solution Calculate the linear satisfaction membership for each of the three robust objectives;

[0052] Step S4.3: Evaluate the overall satisfaction of the non-dominated solutions: For each non-dominated solution... Take the minimum value of the satisfaction of its three robust objectives;

[0053] Step S4.4: Select the final charging strategy: In all non-dominated solutions In the process, the non-dominated solution with the maximum minimum satisfaction is selected as the final charging strategy.

[0054] A robust optimization-based intelligent charging and discharging system for electric vehicles includes:

[0055] The deterministic charging optimization model building module is used to build a deterministic charging optimization model: with the goal of minimizing grid power purchase cost and battery depreciation cost, while incorporating dynamic constraints on battery state of charge, operating mode constraints, user-specified state of charge boundary constraints, equipment safety operation constraints, and power balance constraints; the equipment safety operation constraints include charging and discharging power limits;

[0056] The robust optimization model building module is used to construct a robust optimization model based on the information gap decision theory based on the deterministic charging optimization model: set the load, photovoltaic, and electricity price as uncertain variables and define the robust boundary, construct the worst scenario, and substitute it into the constraint update in step S1;

[0057] The optimal solution set selection module is used to solve robust optimization models using the multi-objective particle swarm optimization algorithm: initialize particles according to charging and discharging power limits and time windows, evaluate the robustness of particles in the worst scenario during iteration, and select non-dominated solutions to obtain the Pareto optimal solution set.

[0058] The charging and discharging strategy screening module is used to select a strategy based on the optimal solution set using fuzzy satisfaction: calculate the satisfaction of each solution under the three robust objectives, take the lowest satisfaction of each solution, and select the solution with the largest lowest satisfaction as the final charging and discharging strategy.

[0059] A non-volatile computer storage medium storing computer-executable instructions that execute a robustly optimized intelligent charging and discharging method for electric vehicles.

[0060] Compared with existing technologies, this invention has the following beneficial effects: By integrating four core innovative designs—robust optimization construction based on information gap decision theory, synergistic consideration of full lifecycle economics and battery lifespan, multiple physical constraint guarantees, and multi-objective solution and fuzzy decision optimization—this invention forms a synergistic technical solution, ultimately achieving the following combined effects: In uncertain environments where photovoltaic output, load demand, and electricity prices fluctuate randomly, this invention ensures that charging and discharging scheduling is feasible and cost-controllable within the fluctuation range by defining robust boundaries for uncertain variables and simulating worst-case scenarios, thus solving the problems of easy failure and soaring costs associated with traditional deterministic optimization methods; and by depreciating the battery into… This approach incorporates an economic objective function and adds operational mode constraints to limit the number of charge / discharge switching cycles, achieving a balance between battery life and economic efficiency. This avoids the problems of accelerated aging and increased depreciation costs caused by neglecting battery life in existing strategies. Simultaneously, it combines multiple physical constraints such as dynamic constraints on battery state of charge, equipment safety constraints, and power balance constraints to ensure equipment safety. Furthermore, it uses a multi-objective particle swarm optimization algorithm to select the Pareto optimal solution set and a fuzzy satisfaction method to select the optimal compromise solution, avoiding the one-sidedness of single-objective optimization. Ultimately, it achieves economical, reliable, and battery-friendly electric vehicle charge / discharge scheduling, which is fully compatible with smart microgrids and V2G scenarios, significantly improving overall operational performance in uncertain environments. Attached Figure Description

[0061] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0062] like Figure 1 As shown, the present invention provides a technical solution: a robust optimization-based intelligent charging and discharging method for electric vehicles, comprising the following steps:

[0063] Step S1: Establish a deterministic charging optimization model: with the goal of minimizing grid power purchase costs and battery depreciation costs, while incorporating dynamic constraints on battery state of charge, operating mode constraints, user-specified state of charge boundary constraints, equipment safety operation constraints, and power balance constraints; the equipment safety operation constraints include charging and discharging power limits.

[0064] Establishing a deterministic charging optimization model is to construct an energy management model for V2G-oriented electric vehicle and photovoltaic collaboration, which jointly ensures economic efficiency, equipment lifespan and operational stability through customized economic objectives and physical constraints.

[0065] Step S1.1: Construct an economic objective function to clarify the core direction of model optimization. The economic objective function aims to minimize the total lifecycle cost of the deterministic charging optimization model. The total cost includes two parts: first, the electricity purchase cost generated by purchasing power from the main grid at each time period of the day (the product of the purchased power and the corresponding electricity price for each time period needs to be calculated and summed, with the time index covering 24 time periods of the day); second, the battery depreciation cost generated by the electric vehicle during the charging station access period (from the arrival time to the departure time) due to the charging and discharging cycle (the sum of the absolute values ​​of the power exchanged between the electric vehicle and the microgrid during this period needs to be calculated and multiplied by the battery depreciation cost coefficient). Among them, a negative value of the power exchanged between the electric vehicle and the microgrid represents charging, a positive value represents discharging, and zero value represents standby.

[0066] Economic objective function Represented as:

[0067] ;

[0068] In the formula, express Power purchased from the main grid at any given time; express Real-time electricity pricing; This represents the battery discount cost coefficient; express The exchange power between electric vehicles and microgrids at any given time; , These represent the times when the electric vehicle arrives at and leaves the charging station, respectively.

[0069] Step S1.2: Supplement the constraints of the deterministic charging optimization model to ensure stable operation and equipment safety, including:

[0070] Battery State of Charge (SOC) Dynamic Constraints: Energy conversion losses during charging and discharging must be taken into account. The SOC for a certain period of time needs to be calculated based on the SOC of the previous period, the power exchange between the electric vehicle and the microgrid in the current period, the dispatch time interval (usually set to 1 hour), the battery rated capacity, and the charging and discharging efficiency. In charging mode, the charging efficiency (value between 0 and 1) is used, in discharging mode, the reciprocal of the discharging efficiency (discharging efficiency value between 0 and 1) is used, and in standby mode, the efficiency is calculated as 1. The value of the battery SOC ranges from 0 to 1.

[0071] The dynamic constraint on battery state of charge (SOC) (taking energy conversion losses into account) is expressed as follows:

[0072] ;

[0073] In the formula, express The state of charge of the electric vehicle battery at all times; express The state of charge of the electric vehicle battery at all times; Indicates the time interval for scheduling; Indicates the rated capacity of the electric vehicle battery; This represents the time-varying efficiency function, used to accurately calculate energy conversion losses during the charging and discharging process.

[0074] Wherein, the time-varying efficiency function Represented as:

[0075] ;

[0076] In the formula, Indicates charging efficiency ( ), representing the energy conversion efficiency from the power grid to the battery; Indicates discharge efficiency ( ), representing the energy conversion efficiency from battery to grid.

[0077] Operating mode constraints: To reduce the impact of frequent switching between charging and discharging states on the equipment, it is necessary to limit the total number of state transitions within the scheduling cycle (from the time period after the electric vehicle arrives to the time of departure) so that it does not exceed the set maximum number of transitions. Among them, the operating mode is represented by a binary variable. When the variable is 1, it means that the electric vehicle is in the charging and discharging active state (i.e., the absolute value of the exchange power is greater than 0), and when it is 0, it means that it is in the standby state (i.e., the exchange power is 0).

[0078] The operation mode constraint (limiting the number of state transitions) is expressed as follows:

[0079] ;

[0080] In the formula, express The binary state variable at time t. Represents the charge / discharge activation state. Indicates standby status; express The binary state variable at any given time; This indicates the maximum number of allowed state transitions.

[0081] User-specified state of charge boundary constraints: The state of charge of the electric vehicle when it arrives at the charging station (and the period before arrival) must be equal to the state of charge at arrival specified by the user, and the state of charge when it leaves the charging station (and the period after departure) must be equal to the state of charge at departure specified by the user.

[0082] The user-specified charge state boundary constraints are represented as follows:

[0083] ;

[0084] In the formula, , These represent the user-specified states of charge upon arrival and departure, respectively. express The time is less than or equal to the time the electric vehicle arrives at the charging station. The state of charge; express The time is greater than or equal to the time when the electric vehicle leaves the charging station The state of charge.

[0085] Equipment safety operation constraints: First, charging and discharging power constraints, the exchange power between electric vehicles and microgrids must be between the minimum charging and discharging power (a negative value, representing the maximum charging power) and the maximum charging and discharging power (a positive value, representing the maximum discharging power); Second, state of charge safety constraints, the state of charge at each time period must be between the set minimum allowable state of charge and the maximum allowable state of charge to protect battery life.

[0086] The constraints on safe operation of equipment are expressed as follows:

[0087] ;

[0088] In the formula, , These represent the minimum and maximum charging and discharging power limits for electric vehicles, respectively. , These represent the minimum and maximum permissible states of charge, respectively, set to protect battery life.

[0089] Power balance constraint: The power purchased from the main grid during a certain period must be equal to the total load power during that period minus the photovoltaic power generation power, and then minus the power exchanged between electric vehicles and the microgrid.

[0090] The power balance constraint is expressed as:

[0091] ;

[0092] In the formula, express Total load power at any given time; express Photovoltaic power generation at any given time.

[0093] Step S2: Based on the deterministic charging optimization model, construct a robust optimization model of information gap decision theory: set load, photovoltaic, and electricity price as uncertain variables and define robust boundaries, construct the worst scenario (load and electricity price take high values, photovoltaic takes low values), and substitute them into the constraint update in step S1 (to ensure cost controllability).

[0094] Step S2.1: Define the set of uncertain variables as... The variables in the set of uncertain variables include load power. Photovoltaic power generation and electricity price Define robust boundary values ​​for each variable in the set of uncertain variables, including robust boundary values ​​for load power. Robust boundary values ​​of photovoltaic power generation and electricity price robust boundary value Robust boundary values ​​represent the maximum permissible fluctuation of each variable relative to its predicted value.

[0095] Step S2.2: Set the objective of the robust optimization model of the information gap decision theory to maximize the robust boundary, i.e. This ensures that the charging strategy is feasible and cost-controllable within the range of uncertainty fluctuations.

[0096] Step S2.3: Construct the worst-case uncertainty scenario: For the load power that increases cost, its worst-case realization value is For electricity prices that increase costs, the worst-case scenario is... For photovoltaic power generation with reduced costs, its worst-case realization value is ; This represents the predicted load power. This represents the electricity price forecast. This represents the predicted value of photovoltaic power generation.

[0097] Step S2.4: Substitute the worst-case uncertainty scenario into the constraints of the deterministic charging optimization model to update and form the constraints of the robust optimization model of information gap decision theory; at the same time, retain the dynamic constraints of battery state of charge, operating mode constraints, user-specified state of charge boundary constraints, and equipment safe operation constraints, and set robust performance indicators.

[0098] The robust performance index is expressed as follows: , Let represent the objective function of the robust optimization model of the information gap decision theory, i.e., the electricity purchase cost considering the worst-case uncertainty. This represents the uncertainty budget parameter used to balance robustness and economy.

[0099] Substituting the worst-case uncertainty scenario into the constraints of the deterministic charging optimization model, the power balance constraint is updated as follows:

[0100] .

[0101] Among them, the objective function of the robust optimization model of information gap decision theory is... Represented as:

[0102] .

[0103] Step S3: Solve the robust optimization model using a multi-objective particle swarm optimization algorithm: Initialize particles according to the charging and discharging power limit and time window, evaluate the robustness of particles in the worst scenario during iteration, and select non-dominated solutions to obtain the Pareto optimal solution set.

[0104] Step S3.1, Population Initialization: Based on the minimum and maximum charge / discharge power limits of the electric vehicle and the times when the electric vehicle arrives at and leaves the charging station, randomly generate an initial particle swarm representing the all-weather charge / discharge power curve of the electric vehicle, and ensure that the particles satisfy... .

[0105] Step S3.2, Particle Evaluation: Decode the position vector of each particle into photovoltaic power generation. Substituting the time series data into the robust optimization model of information gap decision theory, we calculate the three objective function values ​​(robust objective function values) under the worst-case uncertainty scenario, including the load power robust boundary value. Robust boundary values ​​of photovoltaic power generation and electricity price robust boundary value .

[0106] Step S3.3, Non-dominated solution screening and storage: Based on the three objective function values ​​obtained in step S3.2, non-dominated solutions in the particle population are screened according to the Pareto dominance relationship and stored in an external storage library with limited size.

[0107] The non-dominated solution is defined as: if the particle It performs no worse than particles on all optimization objectives. (e.g., "lower or equal cost" and "more robust or equal"), and at least one of the targets performs strictly better than the particle. (For example, "significantly lower cost" or "significantly stronger robustness"), then it is called a particle. Dominant Particle Conversely, if no particle can dominate the particles... Then the particle It is a non-dominant solution.

[0108] Step S3.4, Adaptive Leader Selection: A grid-based density estimation method is used to probabilistically select a "leader" for each particle from an external storage. Solutions in grid cells with fewer members are more likely to be selected to maintain Pareto front diversity.

[0109] Step S3.5, Particle Position and Velocity Update: Update particle velocity and position according to the particle swarm optimization formula. The flight direction is guided by its own historical best position and the selected "leader".

[0110] Step S3.6, Constraint Processing: Correct the updated particle positions to ensure they meet the constraints. And load power robustness boundary, photovoltaic power generation robustness boundary and electricity price robustness boundary.

[0111] Step S3.7, Storage Maintenance and Iteration: After each iteration, new non-dominated solutions in the current population are merged into the external storage and reordered. If the external storage exceeds the preset capacity, non-dominated solutions in densely populated areas are removed. Repeat steps S3.2-S3.6 until the maximum number of iterations is reached. At this point, the solution set in the external storage is the Pareto optimal solution set of the robust optimization model.

[0112] Step S4: Select a strategy based on the optimal solution set using fuzzy satisfaction: Calculate the satisfaction of each solution under the three robust objectives, take the lowest satisfaction of each solution, and select the solution with the largest lowest satisfaction as the final charging and discharging strategy.

[0113] Step S4.1: Clarify the relationship between the decision objective and the Pareto optimal solution set: In the Pareto optimal solution set, each non-dominated solution... This includes the charge / discharge power plan and the corresponding three robust objective function values; the three robust objective function values ​​are: non-dominated solutions. Corresponding load power robust boundary Non-dominated solutions Corresponding photovoltaic power robust boundary Non-dominated solutions Corresponding robust boundary of electricity price ; Indicates non-dominated solutions The corresponding load power robustness boundary value, Indicates non-dominated solutions The corresponding robust boundary value for photovoltaic power, Indicates non-dominated solutions The corresponding robust boundary value for electricity prices.

[0114] Step S4.2: Calculate the satisfaction membership degree of each robustness objective: for each non-dominated solution The linear satisfaction membership degrees are calculated for each of the three robust objectives, and are expressed as follows:

[0115] ;

[0116] ;

[0117] ;

[0118] In the formula, Indicates non-dominated solutions Linear satisfaction membership degree for load power robust boundary targets; , These represent the minimum and maximum values ​​of the load power robustness boundary, respectively; Indicates non-dominated solutions Linear satisfaction membership degree for robust boundary targets of photovoltaic power; , These represent the minimum and maximum values ​​of the robust boundary conditions for photovoltaic power, respectively. Indicates non-dominated solutions Linear satisfaction membership degree for the electricity price robust boundary objective; , These represent the minimum and maximum values ​​of the robust boundary of electricity prices, respectively.

[0119] Step S4.3: Evaluate the overall satisfaction of the non-dominated solutions: For each non-dominated solution... The minimum value among the three robustness target satisfactions is taken, which represents the performance level of the charging and discharging plan in the least desirable robustness dimension.

[0120] Step S4.4: Select the final charging strategy: In all non-dominated solutions In this study, the non-dominated solution with the maximum minimum satisfaction is selected as the final charging strategy; this solution represents the optimal trade-off between robustness to the triple uncertainties of load, photovoltaics, and electricity price.

[0121] A robust optimization-based intelligent charging and discharging system for electric vehicles includes:

[0122] The deterministic charging optimization model building module is used to build a deterministic charging optimization model: with the goal of minimizing grid power purchase cost and battery depreciation cost, while incorporating dynamic constraints on battery state of charge, operating mode constraints, user-specified state of charge boundary constraints, equipment safety operation constraints, and power balance constraints; the equipment safety operation constraints include charging and discharging power limits;

[0123] The robust optimization model building module is used to construct a robust optimization model based on the information gap decision theory based on the deterministic charging optimization model: set the load, photovoltaic, and electricity price as uncertain variables and define the robust boundary, construct the worst scenario, and substitute it into the constraint update in step S1;

[0124] The optimal solution set selection module is used to solve robust optimization models using the multi-objective particle swarm optimization algorithm: initialize particles according to charging and discharging power limits and time windows, evaluate the robustness of particles in the worst scenario during iteration, and select non-dominated solutions to obtain the Pareto optimal solution set.

[0125] The charging and discharging strategy screening module is used to select a strategy based on the optimal solution set using fuzzy satisfaction: calculate the satisfaction of each solution under the three robust objectives, take the lowest satisfaction of each solution, and select the solution with the largest lowest satisfaction as the final charging and discharging strategy.

[0126] A non-volatile computer storage medium storing computer-executable instructions that execute a robustly optimized intelligent charging and discharging method for electric vehicles.

[0127] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A robust optimization-based intelligent charging and discharging method for electric vehicles, characterized in that, Includes the following steps: Step S1: Establish a deterministic charging optimization model: with the goal of minimizing grid power purchase costs and battery depreciation costs, while incorporating dynamic constraints on battery state of charge, operating mode constraints, user-specified state of charge boundary constraints, equipment safety operation constraints, and power balance constraints; the equipment safety operation constraints include charging and discharging power limits. Step S2: Based on the deterministic charging optimization model, construct a robust optimization model of information gap decision theory: set the load, photovoltaic, and electricity price as uncertain variables and define robust boundaries, construct the worst scenario, and substitute it into the constraint update in step S1; Step S3: Solve the robust optimization model using the multi-objective particle swarm optimization algorithm: initialize particles according to the charging and discharging power limit and time window, evaluate the robustness of particles in the worst scenario during iteration, and obtain the Pareto optimal solution set by screening non-dominated solutions. Step S4: Select a strategy based on the optimal solution set using fuzzy satisfaction: Calculate the satisfaction of each solution under the three robust objectives, take the lowest satisfaction of each solution, and select the solution with the largest lowest satisfaction as the final charging and discharging strategy.

2. The intelligent charging and discharging method for electric vehicles based on robust optimization according to claim 1, characterized in that: The specific process of step S1 is as follows: Step S1.1: Construct the economic objective function; the economic objective function aims to minimize the total lifecycle cost of the deterministic charging optimization model. The total cost includes two parts: first, the cost of purchasing electricity from the main grid at different times of the day; second, the battery depreciation cost incurred by the electric vehicle during its connection to the charging station due to charge-discharge cycles. Here, a negative value for the exchange power between the electric vehicle and the microgrid represents charging, a positive value represents discharging, and zero represents standby. Economic objective function Represented as: ； In the formula, express Power purchased from the main grid at any given time; express Real-time electricity pricing; This represents the battery discount cost coefficient; express The exchange power between electric vehicles and microgrids at any given time; , These represent the times when the electric vehicle arrives at and leaves the charging station, respectively. Step S1.2: Supplement the constraints of the deterministic charging optimization model, including dynamic constraints on battery state of charge, operating mode constraints, user-specified state of charge boundary constraints, equipment safe operation constraints, and power balance constraints.

3. The intelligent charging and discharging method for electric vehicles based on robust optimization according to claim 2, characterized in that: The dynamic constraint of battery state of charge taking into account energy conversion losses is expressed as follows: ； In the formula, express The state of charge of the electric vehicle battery at all times; express The state of charge of the electric vehicle battery at all times; Indicates the time interval for scheduling; Indicates the rated capacity of the electric vehicle battery; Represents the time-varying efficiency function; The operation mode constraint that limits the number of state transitions is represented as follows: ； In the formula, express The binary state variable at time t. Represents the charge / discharge activation state. Indicates standby status; express The binary state variable at any given time; Indicates the maximum allowed number of state transitions; User-specified charge state boundary constraints are represented as follows: ； In the formula, , These represent the user-specified states of charge upon arrival and departure, respectively. express The time is less than or equal to the time the electric vehicle arrives at the charging station. The state of charge; express The time is greater than or equal to the time when the electric vehicle leaves the charging station. The state of charge; Equipment safety operation constraints are expressed as follows: ； In the formula, , These represent the minimum and maximum charging and discharging power limits for electric vehicles, respectively. , These represent the minimum and maximum permissible states of charge set to protect battery life, respectively. The power balance constraint is expressed as: ； In the formula, express Total load power at any given time; express Photovoltaic power generation at any given time.

4. The intelligent charging and discharging method for electric vehicles based on robust optimization according to claim 3, characterized in that: The specific process of step S2 is as follows: Step S2.1: Define the set of uncertain variables as... The variables in the set of uncertain variables include Total load power at any given time , Photovoltaic power generation at any time and Real-time electricity purchase price ; Define robust boundary values ​​for each variable in the set of uncertain variables, including robust boundary values ​​for load power. Robust boundary values ​​of photovoltaic power generation and electricity price robust boundary value Robust boundary values ​​represent the maximum permissible fluctuation of each variable relative to its predicted value; Step S2.2: Set the objective of the robust optimization model of the information gap decision theory to maximize the robust boundary, i.e. ; Step S2.3: Construct the worst-case uncertainty scenario: For the load power that increases cost, the worst-case realization value is For electricity prices that increase costs, the worst-case scenario is... For cost-reducing photovoltaic power generation, the worst-case scenario is... ; This represents the predicted load power. This represents the electricity price forecast. This represents the predicted value of photovoltaic power generation. Step S2.4: Substitute the worst-case uncertainty scenario into the constraints of the deterministic charging optimization model to update and form the constraints of the robust optimization model of information gap decision theory; at the same time, retain the dynamic constraints of battery state of charge, operating mode constraints, user-specified state of charge boundary constraints, and equipment safe operation constraints, and set robust performance indicators.

5. The intelligent charging and discharging method for electric vehicles based on robust optimization according to claim 4, characterized in that: Robust performance metrics are expressed as: , Let represent the objective function of the robust optimization model of the information gap decision theory, i.e., the electricity purchase cost considering the worst-case uncertainty. This represents the parameters used to balance robustness and economy; The objective function of the robust optimization model of information gap decision theory Represented as: 。 6. The intelligent charging and discharging method for electric vehicles based on robust optimization according to claim 5, characterized in that: Substituting the worst-case uncertainty scenario into the constraints of the deterministic charging optimization model, the power balance constraint is updated as follows: 。 7. The intelligent charging and discharging method for electric vehicles based on robust optimization according to claim 6, characterized in that: The specific process of step S3 is as follows: Step S3.1, Population Initialization: Based on the minimum and maximum charge / discharge power limits of the electric vehicle and the times when the electric vehicle arrives at and leaves the charging station, randomly generate an initial particle swarm representing the all-weather charge / discharge power curve of the electric vehicle, and ensure that the particles satisfy... ; Step S3.2, Particle Evaluation: Decode the position vector of each particle into a time series of photovoltaic power generation, substitute it into the robust optimization model of information gap decision theory, and calculate the three objective function values, i.e., the robustness objective function values, under the worst uncertainty scenario, including the load power robust boundary value. Robust boundary values ​​of photovoltaic power generation and electricity price robust boundary value ; Step S3.3, Non-dominated solution screening and storage: Based on the three objective function values ​​obtained in step S3.2, non-dominated solutions in the particle population are screened according to the Pareto dominance relation and stored in an external storage library; Step S3.4, Adaptive Leader Selection: A grid-based density estimation method is used to probabilistically select a leader for each particle from an external storage. Step S3.5: Particle position and velocity update; Step S3.6, Constraint Processing: Correct the updated particle positions so that the corrected particles satisfy... Robustness boundaries for load power, photovoltaic power generation, and electricity price; Step S3.7, Storage Maintenance and Iteration: After each iteration, the non-dominated solutions in the current population are merged into the external storage and reordered. If the external storage exceeds the preset capacity, the non-dominated solutions in the densely populated region are removed. Repeat steps S3.2-S3.6 until the maximum number of iterations is reached. At this point, the solution set in the external storage is the Pareto optimal solution set of the robust optimization model.

8. The intelligent charging and discharging method for electric vehicles based on robust optimization according to claim 7, characterized in that: The specific process of step S4 is as follows: Step S4.1: Clarify the relationship between the decision objective and the Pareto optimal solution set: In the Pareto optimal solution set, each non-dominated solution... This includes the charge / discharge power plan and the corresponding three robust objective function values; the three robust objective function values ​​are: non-dominated solutions. Corresponding load power robust boundary Non-dominated solutions Corresponding photovoltaic power robust boundary Non-dominated solutions Corresponding robust boundary of electricity price ; Indicates non-dominated solutions The corresponding load power robustness boundary value, Indicates non-dominated solutions The corresponding robust boundary value for photovoltaic power, Indicates non-dominated solutions The corresponding robust boundary value for electricity prices; Step S4.2: Calculate the satisfaction membership degree of each robustness objective: for each non-dominated solution Calculate the linear satisfaction membership for each of the three robust objectives; Step S4.3: Evaluate the overall satisfaction of the non-dominated solutions: For each non-dominated solution... Take the minimum value of the satisfaction of its three robust objectives; Step S4.4: Select the final charging strategy: In all non-dominated solutions In the process, the non-dominated solution with the maximum minimum satisfaction is selected as the final charging strategy.

9. A robust optimization-based intelligent charging and discharging system for electric vehicles, characterized in that, include: The deterministic charging optimization model building module is used to build a deterministic charging optimization model: with the goal of minimizing grid power purchase cost and battery depreciation cost, while incorporating dynamic constraints on battery state of charge, operating mode constraints, user-specified state of charge boundary constraints, equipment safety operation constraints, and power balance constraints; the equipment safety operation constraints include charging and discharging power limits; The robust optimization model building module is used to construct a robust optimization model based on the information gap decision theory based on the deterministic charging optimization model: set the load, photovoltaic, and electricity price as uncertain variables and define the robust boundary, construct the worst scenario, and substitute it into the constraint update in step S1; The optimal solution set selection module is used to solve robust optimization models using the multi-objective particle swarm optimization algorithm: initialize particles according to charging and discharging power limits and time windows, evaluate the robustness of particles in the worst scenario during iteration, and select non-dominated solutions to obtain the Pareto optimal solution set. The charging and discharging strategy screening module is used to select a strategy based on the optimal solution set using fuzzy satisfaction: calculate the satisfaction of each solution under the three robust objectives, take the lowest satisfaction of each solution, and select the solution with the largest lowest satisfaction as the final charging and discharging strategy.

10. A non-volatile computer storage medium storing computer-executable instructions, characterized in that, The computer can execute instructions to perform a robust optimization-based intelligent charging and discharging method for electric vehicles as described in any one of claims 1-8.

Images