Vehicle-road-network carbon reduction scheduling method and system considering new energy access

Abstract

The invention relates to the technical field of power grid dispatching, in particular to a vehicle-road-network carbon reduction dispatching method and system considering new energy access, and the method comprises the steps: building a control model, determining a target function and a constraint condition of a regulation and control model, and solving a current optimal generator set power output distribution strategy for the regulation and control model through a dream optimization algorithm; determining an objective function and constraint conditions of the demand response model, solving the demand response model by using the load demand, the dynamic carbon emission factor and the time-of-use electricity price to obtain a current moving path and a current charging scheme of the electric vehicle, and outputting the current moving path and the current charging scheme to the regulation and control model; carrying out iterative solution on the demand response model and the regulation and control model until a preset maximum iteration number is reached, and carrying out vehicle-road-network cooperative scheduling based on the obtained power output distribution strategy, the moving path and the charging scheme of the generator set; the stability and utilization rate of the power grid can be improved.

Description

Technical Field

[0001] This invention relates to the field of power grid dispatching technology, specifically to a vehicle-road-grid carbon reduction dispatching method and system that considers the integration of new energy sources. Background Technology

[0002] With the advancement of the "dual carbon" goals, new energy sources, mainly wind and solar power, have been massively integrated into the power grid. However, their strong volatility and anti-peak-shaving characteristics have exacerbated the supply-demand imbalance in the power system (such as widening peak-valley differences and curtailment of wind and solar power). At the same time, the large-scale development of electric vehicles (EVs) has provided the power system with dispatchable EV resources. However, their charging and discharging behavior is highly dependent on user travel demand and road network traffic conditions, making it difficult for traditional grid dispatching models to effectively coordinate the consumption of new energy sources, grid stability, and the economical and low-carbon operation of EVs.

[0003] Against this backdrop, the coordinated scheduling of "vehicle-road-network" has become a research hotspot: by integrating grid load demand, renewable energy output, road network traffic flow, and the spatiotemporal characteristics of electric vehicles, multi-dimensional resource optimization can be achieved. However, existing technologies often sever the dynamic coupling relationship between the power grid and the transportation network, or focus only on a single objective (such as economic efficiency), lacking a systematic design for the coordinated optimization of multiple objectives of "carbon reduction-economy-resilience". Summary of the Invention

[0004] The purpose of this invention is to provide a vehicle-road-grid carbon reduction scheduling method and system that takes into account the access of new energy sources, aiming to reduce system carbon emissions while promoting the participation of electric vehicles in grid peak shaving and new energy consumption, thereby increasing grid stability and utilization.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] In a first aspect, embodiments of the present invention provide a vehicle-road-network carbon reduction scheduling method considering the access of new energy sources, the method comprising the following steps:

[0007] S100 establishes a control model for the interaction between the power distribution network and electric vehicles. This control model includes an upper-level power distribution network optimization control model and a lower-level electric vehicle user carbon reduction demand response model.

[0008] S200, determine the objective function and constraints of the control model, and use the dream optimization algorithm to solve the current optimal generator set power output allocation strategy for the control model;

[0009] S300: Determine the objective function and constraints of the demand response model, determine the dynamic carbon emission factor of the current grid node, solve the demand response model using load demand, dynamic carbon emission factor and time-of-use electricity price, obtain the current movement path and charging scheme of electric vehicles, and output the current movement path and charging scheme to the control model.

[0010] S400 iteratively solves the demand response model and control model until the preset maximum number of iterations is reached, and performs vehicle-road-network coordinated scheduling based on the obtained generator set power output allocation strategy, movement path and charging scheme.

[0011] Optionally, the control model is represented as follows:

[0012]

[0013] satisfy

[0014]

[0015] Where p represents the alternative solutions for the leader. The current optimal solution representing the leader; X leader U represents the leader's optimization space. leader The utility function representing the leader; q and These correspond to the alternative solutions and the current optimal solution for the followers, respectively; X follower U represents the optimization space for followers. follower The utility function representing the followers, f represents the current carbon emission factor vector of the followers. CEF This represents the function for calculating dynamic carbon emission factors.

[0016] Optionally, the objective function of the regulation model is:

[0017]

[0018] Where, p grid p represents the exchange power of the tie line between the distribution network and the external power grid. av ω represents the average value of the tie-line exchange, and T represents the number of control time periods; new The weighting coefficient representing the penalty for curtailing wind and solar power is a fixed constant. P represents the maximum output power of the wind turbine. wind (t) represents the current actual output of the wind turbine; P represents the maximum output power of the photovoltaic unit. PV (t) represents the current actual output of the photovoltaic unit.

[0019] Optionally, the objective function of the demand response model is:

[0020]

[0021] in, α represents the cost of vehicle k moving along road segment (i, j); α represents the cost per unit distance; dij β represents the path length of segment (i, j); β represents the unit time cost. This represents the time that vehicle k travels on road segment (i, j); λ(t) represents the path selection variable, and λ(t) represents the time-of-use electricity price. This represents the charging power of vehicle k at time t. p represents the dynamic carbon emission factor of vehicle k at time t. carbon K represents the carbon emission price; Z represents the set of electric vehicles; T represents the set of road network path nodes; and T represents the control time.

[0022] Optionally, the step of using the dream optimization algorithm to solve for the current optimal generator set power output allocation strategy in the control model includes:

[0023] S210, Set the parameters of the dream optimization algorithm. In the initialization phase, generate a random population in the search space as the initial population; wherein, the search space is the upper and lower limits of the generator set's power generation; the parameters include the maximum number of iterations in the exploration phase, the forgetting dimension of each group in the exploration phase, the problem dimension, and the ratio between the forgetting and supplementation strategies and the dream sharing strategy in the exploration phase.

[0024] S220: Calculate individual fitness and select the best individual. Calculate the current fitness value of the population through the objective function of the control model, and determine the best individual as the one with the smallest fitness value.

[0025] S230, during the exploration phase, the population is divided into multiple groups based on differences in memory capacity, and the individuals in each group are updated sequentially during each iteration;

[0026] S240, During the development phase, multiple forgetting dimensions are randomly selected, and the position of each individual in the multiple forgetting dimensions is updated based on the position information of the best individual in the group in previous iterations;

[0027] S250, for individuals that exceed the search boundary, update the position of the individuals that exceed the search boundary in the search space based on the question dimension;

[0028] S260: Determine whether the dream optimization algorithm has reached the maximum number of iterations. If so, output the decision and fitness values ​​of the current best individual, i.e., the current best generator power output allocation strategy; otherwise, return to S230.

[0029] Optionally, updating the individuals in each group sequentially includes:

[0030] Determine the location information of the best individual in the group, and reset the location information of each individual to the location information of the best individual in the group.

[0031] Using the memory strategy, individuals forget and self-organize the positional information in the forgetting dimension, and update the individual's position in that dimension.

[0032] By employing a dream-sharing strategy, individuals randomly acquire the location information of other individuals in the forgetting dimension and update their position in that dimension for the next iteration.

[0033] Optionally, updating the position of each individual in the multiple forgetting dimensions based on the position information of the best individual in the group in previous iterations includes:

[0034] Using a memory strategy, in dimensions other than the selected multiple forgetting dimensions, individuals remember the position information of the best individual in the group in previous iterations and retain it while dreaming;

[0035] By employing a forgetting and replenishment strategy, individuals forget the position information of the best individual in the group in previous iterations while dreaming, and randomly obtain the position information of other individuals from multiple selected forgetting dimensions to update the individual's position in that dimension for the next iteration.

[0036] Secondly, embodiments of the present invention provide a vehicle-road-network carbon reduction scheduling system that considers the access of new energy sources, the system comprising:

[0037] At least one processor;

[0038] At least one memory for storing at least one program;

[0039] When the at least one program is executed by the at least one processor, the at least one processor performs the method as described in any of the preceding statements.

[0040] Thirdly, embodiments of the present invention provide a computer-readable storage medium storing a processor-executable program, which, when executed by a processor, is used to perform the method as described in any of the preceding claims.

[0041] The beneficial effects of this invention are as follows:

[0042] This invention provides a vehicle-road-grid carbon reduction scheduling method and system that considers the integration of new energy sources. It establishes a control model for interaction between the vehicle, road, and grid layers, encompassing an upper-layer distribution network optimization control model and a lower-layer electric vehicle user demand response model for carbon reduction. The upper-layer control model is solved using a dream algorithm to obtain an optimized generator output strategy, based on which a dynamic carbon emission factor is calculated. This dynamic carbon emission factor is combined with a given time-of-use electricity price to solve the lower-layer electric vehicle user demand response model for carbon reduction, yielding corresponding vehicle movement paths and charging schemes. The generator output strategy is then optimized again based on the vehicle movement paths and charging schemes, and this process is repeated until a pre-set maximum number of iterations is reached. Finally, the most optimized generator output strategy, vehicle movement paths, and charging schemes are output. This achieves low-carbon coordinated regulation of the vehicle-road-grid system that fully considers both new energy consumption and carbon emission flows. By coupling the transportation network and the power grid and implementing upper-lower-layer interactive control, this invention can significantly promote the grid's absorption of new energy sources, effectively reduce carbon emissions, and ensure the safe and reliable operation of the power grid. This invention can increase the stability and utilization rate of the power grid. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 This is a flowchart illustrating the vehicle-road-network carbon reduction scheduling method considering the access of new energy sources in an embodiment of the present invention.

[0045] Figure 2 This is a schematic diagram of the extended IEEE 33-node system power grid used in the embodiments of the present invention.

[0046] Figure 3 These are the fitness curves obtained by using the dream optimization algorithm and the particle swarm optimization algorithm to optimize the upper-level distribution network optimization system model in this embodiment of the invention.

[0047] Figure 4 This is a graph showing the total system load before and after optimization in an embodiment of the present invention;

[0048] Figure 5 This is a schematic diagram of the vehicle-road-network carbon reduction scheduling system considering the access of new energy sources in an embodiment of the present invention. Detailed Implementation

[0049] The following will clearly and completely describe the concept, specific structure, and technical effects of the present invention in conjunction with embodiments and accompanying drawings, so as to fully understand the purpose, solution, and effects of the present invention. It should be noted that, unless otherwise specified, the embodiments and features described in the embodiments of the present invention can be combined with each other.

[0050] The "vehicle-road-network" collaborative scheduling technology in related technologies includes:

[0051] (1) Peak-valley smoothing technology based on fixed energy storage: This type of method mainly adjusts the peak-valley difference of the grid load by configuring centralized energy storage power stations (such as lithium batteries and pumped storage). However, the energy storage capacity is fixed and the investment cost is high, which cannot adapt to the real-time fluctuations of new energy output and load.

[0052] (2) Static vehicle-network coordinated control model: This type of method is based on EV scheduling strategy with fixed charging pile locations and preset driving routes. It does not consider dynamic traffic information (such as real-time road conditions and charging pile queuing time), making it difficult to maximize the spatiotemporal flexibility of electric vehicles.

[0053] (3) Carbon emission single-objective optimization technology: This type of method only aims to minimize the carbon emissions of the power grid or vehicles, ignoring the game relationship between multiple objectives such as peak-valley difference, new energy absorption rate, and user travel cost, resulting in insufficient practicality of the optimization results.

[0054] The related technologies have the following problems:

[0055] In summary, existing optimization scheduling methods that consider vehicle-network cooperation have the following drawbacks:

[0056] Lack of dynamic coupling: Existing models rarely dynamically link road network traffic conditions (such as congestion and route selection) with real-time grid load demand, resulting in inaccurate predictions of the dispatchable capacity and spatiotemporal distribution of electric vehicles.

[0057] Insufficient multi-objective coordination: Most existing methods solve the optimization objectives of the grid side (peak-valley difference, renewable energy consumption) and the vehicle side (path cost, charging and discharging revenue, carbon emissions) in isolation, lacking a global coordination mechanism and easily getting trapped in local optima.

[0058] Carbon cost accounting is crude: Existing methods for calculating carbon emissions are mostly based on fixed grid emission factors, without considering the impact of dynamic changes in the penetration rate of new energy sources on marginal carbon emissions, making it difficult to accurately quantify the carbon reduction benefits of electric vehicles.

[0059] To address the aforementioned shortcomings, this invention proposes a two-layer interactive optimization scheduling method for "vehicle-road-network," which aims to solve the following core problems by introducing a dynamic carbon factor:

[0060] (1) Dynamic coupling modeling: How to deeply integrate the dynamic characteristics of road network traffic flow (such as vehicle speed and path) with the flexibility requirements of power grid (such as peak-valley difference and new energy absorption gap) to construct a spatiotemporally coupled electric vehicle control model.

[0061] (2) Multi-objective game collaboration: How to design an interactive optimization mechanism between upper and lower layers to achieve optimal collaboration between the grid carbon reduction target (minimizing peak-valley difference and maximizing new energy consumption) and the vehicle economic target (path cost and charging and discharging revenue).

[0062] (3) Accurate quantification of carbon costs: How to calculate the dynamic carbon emission factor based on the power flow of the power grid and construct the mapping relationship between EV charging and discharging behavior and carbon costs, so as to realize that the carbon reduction benefits are measurable and optimizable.

[0063] See Figure 1 This invention provides a vehicle-road-network carbon reduction scheduling method considering the access of new energy sources, the method comprising the following steps:

[0064] S100 establishes a control model for the interaction between the power distribution network and electric vehicles. This control model includes an upper-level power distribution network optimization control model and a lower-level electric vehicle user carbon reduction demand response model.

[0065] Specifically, a vehicle-road-network collaborative interactive control model is built, with the upper and lower layers interacting through master-slave game theory. This model includes an upper-layer power distribution network optimization and control model and a lower-layer electric vehicle user participation carbon reduction demand response model.

[0066] S200, determine the objective function and constraints of the control model, and use the dream optimization algorithm to solve the current optimal generator set power output allocation strategy for the control model;

[0067] Specifically, objective functions and constraints are set for the upper-level power distribution network optimization and control model and the lower-level electric vehicle user participation carbon reduction demand response model.

[0068] It should be noted that the objective function of the control model is used to minimize the peak-to-valley difference and the penalty for wind and solar curtailment, while the constraints are used to ensure the safe operation of the power grid. The generator power output allocation strategy includes the active and reactive power output of each generator node. Based on the current optimal travel path selection and charging scheme of electric vehicles, the Dream Optimization Algorithm (DOA) is used to solve for the current optimal generator power output allocation strategy of the upper-level model. The Dream Optimization Algorithm is inspired by the changes in brain waves during rapid sleep.

[0069] S300: Determine the objective function and constraints of the demand response model, determine the dynamic carbon emission factor of the current grid node, solve the demand response model using load demand, dynamic carbon emission factor and time-of-use electricity price, obtain the current movement path and charging scheme of electric vehicles, and output the current movement path and charging scheme to the control model.

[0070] It should be noted that the objective function of the demand response model is to minimize the vehicle's electricity costs, and the constraints include movement constraints, battery capacity constraints, charging power constraints, and charging time constraints.

[0071] Furthermore, based on time-of-use pricing and dynamic carbon emission factors, the costs incurred by users in their travel routes, charging costs, and carbon emission costs are calculated. The core objective of the demand response model for lower-level electricity users to participate in carbon reduction is to minimize user costs.

[0072] S400 iteratively solves the demand response model and control model until the preset maximum number of iterations is reached, and performs vehicle-road-network coordinated scheduling based on the obtained generator set power output allocation strategy, movement path and charging scheme.

[0073] The road-grid coupled electric vehicle demand response optimization scheduling problem in this embodiment can be constructed as a mixed integer linear programming (MILP) optimization model. Based on the set cost objective function and corresponding constraints, combined with the nodal load demand, dynamic carbon emission factor and time-of-use electricity price given by the grid, this problem can be solved using the Gurobi solver. Based on the solution process, the current optimal travel path and charging scheme will be obtained and updated. Finally, the lower-level electric vehicle demand response model will transmit the obtained current optimal travel path and charging scheme to the upper-level control model. Subsequently, the upper-level control model will execute step S200 again, and so on, the upper and lower levels will continuously interact and iterate until the preset maximum number of iterations is reached. At this time, the control model considering the interaction between vehicles, roads and grids with new energy access will obtain the minimum peak-valley difference in electricity consumption and the penalty for wind and solar curtailment; at the same time, the lower-level electric vehicle demand response model will also obtain the optimal path cost, charging cost and carbon emission cost, thereby realizing low-carbon coordinated control of vehicles, roads and grids that fully considers the consumption of new energy and carbon emission flow. This invention, by coupling the transportation network and the power grid and implementing interactive control between the upper and lower layers, can significantly promote the grid's absorption of new energy sources, effectively reduce carbon emissions, and ensure the safe and reliable operation of the power grid.

[0074] In some embodiments, the control model is represented as follows:

[0075]

[0076] satisfy

[0077]

[0078] Where p represents the alternative solutions for the leader. The current optimal solution representing the leader; X leader U represents the leader's optimization space. leader The utility function representing the leader; q and These correspond to the alternative solutions and the current optimal solution for the followers, respectively; X follower U represents the optimization space for followers. follower The utility function representing the followers, f represents the current carbon emission factor vector of the followers. CEF This represents the function for calculating dynamic carbon emission factors.

[0079] Specifically, the upper level assumes the role of leader, denoted as leader; the lower level plays the role of follower, denoted as follower; at the leader level, p represents the alternative solutions for the leader. The current optimal solution represents the leader; the range within which the leader can optimize, i.e., the optimization space, is denoted by X. leader U leader This is used to measure the leader's utility function; for followers, q and These correspond to the alternative solutions and the current optimal solution for the followers, respectively; X follower U represents the optimization space for followers. follower This represents the utility function of the followers. Specifically, f represents the current carbon emission factor vector of the follower, which encompasses the dynamic carbon emission factor over multiple time points; CEF This represents the dynamic carbon emission factor calculation function, used to calculate the dynamic carbon emission factor under different conditions. These elements together constitute a key component of the vehicle-road-network upper and lower layer interactive control model.

[0080] In some embodiments, the objective function of the regulation model is:

[0081]

[0082] Where, p grid p represents the exchange power of the tie line between the distribution network and the external power grid. av ω represents the average value of the tie-line exchange, and T represents the number of control time periods; new The weighting coefficient representing the penalty for curtailing wind and solar power is a fixed constant. P represents the maximum output power of the wind turbine. wind (t) represents the current actual output of the wind turbine; P represents the maximum output power of the photovoltaic unit.PV (t) represents the current actual output of the photovoltaic unit.

[0083] The purpose of the constraints in the upper-level control model is to ensure the safe operation of the power grid. These constraints include power flow balance, power balance, active power constraints, reactive power constraints, voltage amplitude constraints, and other constraints.

[0084] The constraints of the regulation model are:

[0085]

[0086] V i min ≤V i ≤V i max i = 1, 2, ..., N b (11);

[0087]

[0088] in, This represents the active power output by the i-th generator node. This represents the active power demand of the i-th node. and These represent the lower and upper limits of the output power of the i-th generator node, respectively, to ensure that the active power of each node is within the permissible range and to avoid overload or underload. It is the reactive power output of the i-th generator node. This is the reactive power demand of the i-th node. and G represents the lower and upper limits of the reactive power output of the i-th generator node, respectively, used to maintain the reactive power balance of the power grid and stabilize the voltage; ij B represents the conductance of line (i, j). ij θ represents the inductive reactance of line (i, j); ij V represents the voltage phase angle difference between the i-th node and the j-th node; i With V j V represents the voltage magnitude of the i-th and j-th nodes, respectively. i min and V i max These represent its lower and upper limits, respectively; S l This represents the apparent power flow of the l-th line. This is the maximum allowable power flow for line l, to prevent the line from transmitting power beyond its rated capacity and ensure safe line operation; in addition, N i N represents the set of nodes directly connected to node i; g N represents the set of generator nodes;b Represents the PQ node set; N L This represents the set of node branches.

[0089] While ensuring the safe operation of the power grid, the output strategy of coal-fired power units should be adjusted to achieve the goal of minimizing the peak-valley difference in the power grid.

[0090] In some embodiments, the objective function of the demand response model is:

[0091]

[0092] in, α represents the cost of vehicle k moving along road segment (i, j); α represents the cost per unit distance; d ij β represents the path length of segment (i, j); β represents the unit time cost. This represents the time that vehicle k travels on road segment (i, j); The variable representing route selection is used to determine whether vehicle k selects route segment (i, j), with a value of 1 indicating selection and 0 indicating no selection; λ(t) represents the time-of-use electricity price. This represents the charging power of vehicle k at time t. p represents the dynamic carbon emission factor of vehicle k at time t. carbon K represents the carbon emission price; Z represents the set of electric vehicles; T represents the set of road network path nodes; and T represents the control time.

[0093] The constraints of the demand response model are:

[0094]

[0095] Where N represents the set of path nodes connected to path node i; This represents the speed of vehicle k on road segment (i, j); The maximum allowed travel time is represented by the road segment (i, j). V represents the initial state of charge of vehicle k; k C represents the energy consumption per unit distance of vehicle k; k η represents the battery capacity of vehicle k; c Represents charging efficiency; and These represent the minimum and maximum charge states of vehicle k, respectively. This represents the vehicle's maximum charging power limit. This represents the time it takes for vehicle k to reach the charging node. This represents the time when vehicle k is scheduled to leave the charging node.

[0096] It should be noted that, based on the generator power output obtained from the above steps, the power flow and carbon emission flow in the power grid can be further calculated. The carbon emission flow is specifically constructed based on the proportional sharing principle.

[0097] The dynamic carbon emission factor (CEF) of the i-th node can be obtained in the following way:

[0098]

[0099] Where, δ i The dynamic carbon emission factor represents the i-th node; This represents the active power output of the generator set connected to the i-th node; P represents the carbon emission factor of the generator set connected at the i-th node; ij Ω represents the active power of the branch flowing from the j-th node to the i-th node; i δ represents the set of nodes connected to the inflow branch of the i-th node; j The dynamic carbon emission factor represents the j-th node.

[0100] In some embodiments, the step of using the dream optimization algorithm to solve for the current optimal generator set power output allocation strategy in the control model includes:

[0101] S210, Set the parameters of the dream optimization algorithm. In the initialization phase, generate a random population in the search space as the initial population; wherein, the search space is the upper and lower limits of the generator set's power generation; the parameters include the maximum number of iterations in the exploration phase, the forgetting dimension of each group in the exploration phase, the problem dimension, and the ratio between the forgetting and supplementation strategies and the dream sharing strategy in the exploration phase.

[0102] During the initialization phase, the initial iteration number it = 1 and the maximum iteration number T are set first. max =100. Similar to other artificial intelligence algorithms, in the initialization phase, the dream optimization algorithm first generates a random population in the search space as the initial population, thus beginning the optimization process.

[0103] The formula for obtaining the initial population is as follows:

[0104] X i =X l +rand×(X u -X l ), i = 1, 2, ..., N(21);

[0105]

[0106] Among them, N represents the number of individuals and also refers to the population size, that is, the size of the candidate solutions included in the group. In this embodiment, the value of N is 50; X i represents the i-th individual in the population, that is, the i-th candidate solution. In this embodiment, it refers to the power generation power distribution instruction of each generator set. Each X i is a 3x24 numerical matrix. 3 represents the decision vector dimension, and 24 represents the time; X l and X u respectively represent the lower and upper boundaries of the search space, that is, the upper and lower limits of the generator set power generation power. In this embodiment, X l takes 1200, and X u takes 0; rand is a Dim-dimensional vector, and each dimension is a random number between 0 and 1; Equation (22) represents the initial population obtained by initialization. Among them, x i,j represents the position of the i-th individual in the j-th dimension. Dim represents the dimension of the optimization problem, that is, the number of decision variables. In this embodiment, Dim takes 3, that is, 3 generator sets of fuel generator sets, wind turbine generator sets, and photovoltaic generator sets.

[0107] Considering the stability and applicability of the dream optimization algorithm, the parameters of the dream optimization algorithm can be set as follows:

[0108]

[0109] Among them, T d represents the maximum number of iterations in the exploration stage, and Tmax represents the total maximum number of iterations.

[0110]

[0111] Among them, randi(a, b) represents a random integer selected from the range from a to b. k q represents the forgotten dimension of q groups in the exploration stage, and Dim represents the problem dimension.

[0112]

[0113] Among them, randi(a, b) represents a random integer selected from the range from a to b. k r represents the forgotten dimension in the development stage, and Dim represents the problem dimension.

[0114] In addition, the parameter u is set to adjust the ratio between the forgetting and replenishment strategy and the dream sharing strategy in the exploration stage. When rand < u, the forgetting and replenishment strategy is executed; otherwise, the dream sharing strategy is executed. It is set that u = 0.9.

[0115] S220: Calculate individual fitness and select the best individual. Calculate the current fitness value of the population through the objective function of the control model, and determine the best individual as the one with the smallest fitness value.

[0116] Specifically, calculate the individual fitness and select the best individual. Calculate the current fitness value (also known as the objective function value) f(X) of the population using formula (4), and determine the individual with the smallest fitness value as the best individual, i.e., the current optimal solution.

[0117] S230, during the exploration phase, the population is divided into multiple groups based on differences in memory capacity, and the individuals in each group are updated sequentially during each iteration;

[0118] Specifically, during the exploration phase, the number of iterations is less than the boundary iteration number, that is, the number of iterations ranges from 1 to the boundary iteration number T. d First, the population was divided into 5 groups based on differences in memory ability, and the individuals in each group were updated as follows:

[0119] Each iteration is considered a dreaming action, and the optimal solution and optimal value are sought by continuously performing this action. Before each dream, all individuals in each group see the previous best dream (i.e., the best individual in the previous iteration). Since individuals randomly forget some information (i.e., information in certain dimensions, called forgetting dimensions) while dreaming, only the positions of the forgetting dimensions are updated. The memory capacity of each group is different, and the number of forgetting dimensions is also different, represented by parameters k1, k2, k3, k4, and k5. Therefore, each individual's position is first reset to the position of the best individual in its group in the previous iteration, and then K is randomly selected from the forgetting dimensions. q There are three dimensions, denoted as K1, K2, ..., K1. The positions in these dimensions are then updated, where q represents the group number, q = 1, 2, 3, 4, 5. In each iteration, updates are performed sequentially from the 1st individual to the Nth individual.

[0120] In some embodiments, updating the individuals in each group sequentially includes:

[0121] Determine the location information of the best individual in the group, and reset the location information of each individual to the location information of the best individual in the group.

[0122] Using the memory strategy, individuals forget and self-organize the positional information in the forgetting dimension, and update the individual's position in that dimension.

[0123] By employing a dream-sharing strategy, individuals randomly acquire the location information of other individuals in the forgetting dimension and update their position in that dimension for the next iteration.

[0124] The specific update method and formula are as follows:

[0125] (a) Memory strategy: First, as shown in the basic memory strategy, individuals in group q can memorize the location information of the best individual in their group before dreaming, and reset their own location information to that of the best individual in their group:

[0126]

[0127] in, This represents the i-th individual at iteration number t+1; This represents the best individual in group q at iteration number t. The formula shows that, at K1, K2, ..., ... In dimensions other than those mentioned above, individuals can remember the location information of the best individuals in their group when they dream, retaining the accurate location in these dimensions.

[0128] (b) Forgetting and Replenishment Strategy: The forgetting and replenishment strategy combines global and local search capabilities. This strategy follows the memory strategy, allowing individuals to forget and self-organize locational information within the forgetting dimension. The update formula is as follows:

[0129]

[0130] in, This represents the position of the i-th individual in the j-th dimension at iteration number t+1; This represents the position of the best individual in group q in the j-th dimension at iteration number t; x l,j and x u,j These are the lower and upper bounds of the solution search space in the j-th dimension, respectively; rand is a random number between 0 and 1; t is the current iteration number, T max It is the maximum number of iterations, T d It represents the maximum number of iterations during the exploration phase.

[0131] The formula shows that K1, K2, ..., In this dimension, individuals forget the position information of the best individual in their group when dreaming and self-organize new positions. The cosine function reflects that as the number of iterations increases, the supplementary position information of the self-organized individuals becomes closer and closer to the original position information. This demonstrates the inherent logic of the self-organization process and controls the transition of the algorithm from global search to local search.

[0132] (c) Dream Sharing Strategy: The dream sharing strategy in the dream optimization algorithm enhances the ability to escape local optima. This strategy runs in parallel with the forgetting and replenishment strategies, following the memory strategy, and allows individuals to randomly acquire the location information of other individuals in the forgetting dimension. The update formula is as follows:

[0133]

[0134] in, Let represent the position of the i-th individual in the j-th dimension at iteration number t+1; m is a natural number randomly selected from the range [1, N] when updating each dimension. This represents the position of the m-th individual in the j-th dimension at iteration number t.

[0135] The formula shows that, at K1, K2, ..., In this system, individuals are allowed to randomly obtain the location information of other individuals in the population, thus achieving a certain degree of dream information sharing.

[0136] S240, During the development phase, multiple forgetting dimensions are randomly selected, and the position of each individual in the multiple forgetting dimensions is updated based on the position information of the best individual in the group in previous iterations;

[0137] Specifically, during the development phase, the number of iterations is between the boundary iteration number and the maximum iteration number, that is, the number of iterations ranges from the boundary iteration number T. d up to the maximum number of iterations T max Instead of grouping, the best dream from the previous few iterations (i.e., the best individual from the previous few iterations) is shown to the entire group before each dream. Then, the position of each individual in the forgetting dimension is updated.

[0138] In some embodiments, updating the position of each individual in the multiple forgetting dimensions based on the position information of the best individual in the population in previous iterations includes:

[0139] Using a memory strategy, in dimensions other than the selected multiple forgetting dimensions, individuals remember the position information of the best individual in the group in previous iterations and retain it while dreaming;

[0140] By employing a forgetting and replenishment strategy, individuals forget the position information of the best individual in the group in previous iterations while dreaming, and randomly obtain the position information of other individuals from multiple selected forgetting dimensions to update the individual's position in that dimension for the next iteration.

[0141] It should be noted that the forgetting dimension is the same for all individuals in the population, denoted as k. r Randomly select k r There are three forgetting dimensions, denoted as K1, K2, ..., K3. And update the positions in these dimensions. The update method is similar to formulas (23) and (24), and the update formula is as follows:

[0142] (a) Memory strategies: During the development phase, in K1, K2... In dimensions other than those of the group, individuals can remember the position information of the best individual in the group in previous iterations and retain the exact position in these dimensions while dreaming.

[0143]

[0144] in, This represents the i-th individual at iteration number t+1. This represents the best individual in the entire population at iteration number t.

[0145] (b) Forgetting and Reinforcement Strategies: During the development phase, in K1, K2..., In this dimension, when an individual is dreaming, they forget the position information of the best individual in the group in previous iterations and self-organize a new position.

[0146]

[0147] in, This represents the position of the i-th individual in the j-th dimension at iteration number t+1; x represents the position of the optimal individual in the entire population in the j-th dimension at iteration number t; l,j and x u,j These are the lower and upper bounds of the j-th dimension solution space, respectively. rand is a random number between 0 and 1; t is the current iteration number, and T... max This is the maximum number of iterations for the dream optimization algorithm.

[0148] S250, for individuals that exceed the search boundary, update the position of the individuals that exceed the search boundary in the search space based on the question dimension;

[0149] Specifically, for optimization problems of different dimensions, two different boundary condition handling methods are used.

[0150] The first method is suitable for problems where Dim ≤ 15. These problems have relatively few local optima due to their small dimensionality; therefore, a traditional randomized method is used to update points beyond the search boundary, as shown below:

[0151]

[0152] in, x represents the position of the i-th individual in the j-th dimension at iteration t+1; l,j and x u,j These are the lower and upper bounds of the search space in the j-th dimension, respectively; rand is a random number between 0 and 1.

[0153] The second method is suitable for problems where Dim > 15. These problems are more dimensional, complex, and have more local optima, requiring enhanced global optimization capabilities and the ability to escape local optima. Therefore, for surrogate points that exceed boundary conditions in these high-dimensional problems, we use a method similar to the dream-sharing strategy in the development phase for re-updating, as shown below:

[0154]

[0155] in, This represents the position of the i-th individual in the j-th dimension at iteration t+1; m is a random natural number in the range [1, N], which is different from the i-th natural number when updating each dimension.

[0156] Since the dream optimization algorithm updates each individual one by one, while formulas (31) and (32) only update the decision variables of individuals that exceed the boundary conditions, both methods can effectively ensure that all variables of all individuals in the population remain within the search space.

[0157] S260: Determine whether the dream optimization algorithm has reached the maximum number of iterations. If so, output the decision and fitness values ​​of the current best individual, i.e., the current best generator power output allocation strategy; otherwise, return to S230.

[0158] Specifically, determine whether to continue iterating. If it ≥ Tmax (in this embodiment, the maximum number of iterations Tmax = 100), output the decision and fitness value of the current best individual, that is, the optimal solution of the generator set output power allocation command; otherwise, return to step S230 to continue solving for optimization.

[0159] Please see Figure 2 This invention takes a control model combining an extended IEEE 33-node power grid system with a 29-node distance matrix in a certain region as the research object. In this model, the power grid includes different power output units such as gas turbines, wind turbines and photovoltaic units. In addition, the load includes multiple dispatchable mobile electric vehicle aggregation groups.

[0160] In this embodiment, a vehicle-road-network interaction control model is built using a master-slave game theory model. This model encompasses both the upper-level distribution network optimization control model and the lower-level electric vehicle user demand response model for carbon reduction. The overall implementation process is as follows: The dream algorithm is used to solve the upper-level control model, thereby obtaining the optimized generator output strategy. Based on this strategy, the dynamic carbon emission factor is calculated. The dynamic carbon emission factor is combined with a given time-of-use electricity price to solve the lower-level electric vehicle user demand response model for carbon reduction, obtaining the corresponding vehicle movement path and charging scheme. Then, based on the vehicle movement path and charging scheme, the generator output strategy is optimized again. This process is repeated until a pre-set maximum number of iterations is reached. Finally, the most optimized generator output strategy, vehicle movement path, and charging scheme are output.

[0161] Appendix Figure 3 The fitness curves of the vehicle-road-network carbon reduction control system considering new energy access in this embodiment, obtained by optimizing the upper-level distribution network optimization system model using the Dream Optimization Algorithm (DOA) and Particle Swarm Optimization (PSO) algorithms, are presented respectively. It can be seen that, compared to PSO, the generator power allocation command strategy obtained by DOA is a better solution, and its solution speed is also faster.

[0162] Appendix Figure 4 The total load curves of the vehicle-road-grid carbon reduction control system before and after optimization in this embodiment, considering the access of new energy sources, are presented. As can be seen from the curves, the load increases during valley and normal periods and decreases during peak periods, which plays a role in peak shaving and valley filling. The system load curve is also smoother. This shows that the upper and lower layer interactive control model can promote the participation of electric vehicles in grid peak shaving and new energy consumption while reducing system carbon emissions, thereby increasing grid stability and utilization.

[0163] Compared with related technologies, the present invention has the following technical improvements and effects:

[0164] (1) Refined Carbon Flow Tracking and Dynamic Incentives: This invention innovatively introduces the concept of carbon flow tracking. Carbon emission flows are tracked within the power grid and electric vehicle systems, setting carbon emissions as a clear and explicit scheduling optimization target. Within this framework, the design of electric vehicle charging strategies no longer focuses solely on economic benefits but also prioritizes carbon emission reduction, striving to achieve the dual goals of efficiency and environmental protection. Specifically, electric vehicles are encouraged to charge during periods of lower carbon emissions, while during periods of higher carbon emissions, energy is fed back to the power system through reasonable energy release strategies. This approach effectively helps the power system achieve low-carbon scheduling, comprehensively balancing utilization and environmental friendliness, and providing strong support for sustainable development.

[0165] (2) Multi-objective collaborative optimization mechanism: This invention uses a master-slave game framework: the upper layer (power grid) takes peak-valley difference and curtailment penalty as objectives, while the lower layer (EV aggregator) takes path cost, charging cost and carbon cost as objectives. Through electricity price and carbon emission factor, two-way interaction is achieved to minimize power transmission loss and improve the absorption rate of renewable energy.

[0166] (3) High-efficiency intelligent optimization algorithm: This invention utilizes a dream optimization algorithm. This algorithm coordinates the optimization process through a unique dream initialization process and the implementation of memory, forgetting and supplementation, and dream sharing mechanisms during the exploration and development phases. Compared to conventional heuristic algorithms, the dream optimization algorithm can cleverly balance global and local searches. It can adaptively adjust the search space resolution and search speed, thereby finding the optimal solution quickly and accurately, exhibiting significant characteristics of fast convergence speed and high solution accuracy.

[0167] and Figure 1 The corresponding method is referenced. Figure 5 This invention provides a vehicle-road-network carbon reduction scheduling system that considers the integration of new energy sources, comprising:

[0168] At least one processor;

[0169] At least one memory for storing at least one program;

[0170] When the at least one program is executed by the at least one processor, the at least one processor performs the method described above.

[0171] It is evident that the content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0172] Furthermore, embodiments of the present invention also disclose a computer program product or computer program stored in a computer-readable storage medium. A processor of a computer device can read the computer program from the computer-readable storage medium, and the processor executes the computer program, causing the computer device to perform the described method. Similarly, the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0173] Those skilled in the art will understand that all or some of the methods and systems disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0174] The above is a detailed description of the preferred embodiments of this disclosure. However, this disclosure is not limited to the above embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of this disclosure. All such equivalent modifications or substitutions are included within the scope defined by the claims of this disclosure.

Claims

1. A vehicle-road-network carbon reduction scheduling method considering the integration of new energy sources, characterized in that, The method includes the following steps: S100 establishes a control model for the interaction between the power distribution network and electric vehicles. This control model includes an upper-level power distribution network optimization control model and a lower-level electric vehicle user carbon reduction demand response model. S200, determine the objective function and constraints of the control model, and use the dream optimization algorithm to solve the current optimal generator set power output allocation strategy for the control model; S300: Determine the objective function and constraints of the demand response model, determine the dynamic carbon emission factor of the current grid node, solve the demand response model using load demand, dynamic carbon emission factor and time-of-use electricity price, obtain the current movement path and charging scheme of electric vehicles, and output the current movement path and charging scheme to the control model. S400 iteratively solves the demand response model and control model until the preset maximum number of iterations is reached, and performs vehicle-road-network coordinated scheduling based on the obtained generator set power output allocation strategy, movement path and charging scheme.

2. The method according to claim 1, characterized in that, The control model is represented as follows: satisfy Where p represents the alternative solutions for the leader. The current optimal solution representing the leader; X leader U represents the leader's optimization space. leader The utility function representing the leader; q and These correspond to the alternative solutions and the current optimal solution for the followers, respectively; X follower U represents the optimization space for followers. follower The utility function representing the followers, f represents the current carbon emission factor vector of the followers. CEF This represents the function for calculating dynamic carbon emission factors.

3. The method according to claim 2, characterized in that, The objective function of the regulation model is: Where, p grid p represents the exchange power of the tie line between the distribution network and the external power grid. av ω represents the average value of the tie-line exchange, and T represents the number of control time periods; new The weighting coefficient representing the penalty for curtailing wind and solar power is a fixed constant. P represents the maximum output power of the wind turbine. wind (t) represents the current actual output of the wind turbine; P represents the maximum output power of the photovoltaic unit. PV (t) represents the current actual output of the photovoltaic unit.

4. The method according to claim 3, characterized in that, The objective function of the demand response model is: in, α represents the cost of vehicle k moving along road segment (i, j); α represents the cost per unit distance; d ij β represents the path length of segment (i, j); β represents the unit time cost. This represents the travel time of vehicle k on road segment (i, j); λ(t) represents the path selection variable, and λ(t) represents the time-of-use electricity price. This represents the charging power of vehicle k at time t. p represents the dynamic carbon emission factor of vehicle k at time t. carbon K represents the carbon emission price; Z represents the set of electric vehicles; T represents the set of road network path nodes; and T represents the control time.

5. The method according to claim 1, characterized in that, The method of using the dream optimization algorithm to solve for the current optimal generator power output allocation strategy in the control model includes: S210, Set the parameters of the dream optimization algorithm. In the initialization phase, generate a random population in the search space as the initial population; wherein, the search space is the upper and lower limits of the generator set's power generation; the parameters include the maximum number of iterations in the exploration phase, the forgetting dimension of each group in the exploration phase, the problem dimension, and the ratio between the forgetting and supplementation strategies and the dream sharing strategy in the exploration phase. S220: Calculate individual fitness and select the best individual. Calculate the current fitness value of the population through the objective function of the control model, and determine the best individual as the one with the smallest fitness value. S230, during the exploration phase, the population is divided into multiple groups based on differences in memory capacity, and the individuals in each group are updated sequentially during each iteration; S240, During the development phase, multiple forgetting dimensions are randomly selected, and the position of each individual in the multiple forgetting dimensions is updated based on the position information of the best individual in the group in previous iterations; S250, for individuals that exceed the search boundary, update the position of the individuals that exceed the search boundary in the search space based on the question dimension; S260: Determine whether the dream optimization algorithm has reached the maximum number of iterations. If so, output the decision and fitness values ​​of the current best individual, i.e., the current best generator power output allocation strategy; otherwise, return to S230.

6. The method according to claim 5, characterized in that, The step of updating each individual in each group sequentially includes: Determine the location information of the best individual in the group, and reset the location information of each individual to the location information of the best individual in the group. Using the memory strategy, individuals forget and self-organize the positional information in the forgetting dimension, and update the individual's position in that dimension. By employing a dream-sharing strategy, individuals randomly acquire the location information of other individuals in the forgetting dimension and update their position in that dimension for the next iteration.

7. The method according to claim 5, characterized in that, The update of each individual's position in the multiple forgetting dimensions based on the position information of the best individual in the group in previous iterations includes: Using a memory strategy, in dimensions other than the selected multiple forgetting dimensions, individuals remember the position information of the best individual in the group in previous iterations and retain it while dreaming; By employing a forgetting and replenishment strategy, individuals forget the position information of the best individual in the group in previous iterations while dreaming, and randomly obtain the position information of other individuals from multiple selected forgetting dimensions to update the individual's position in that dimension for the next iteration.

8. A vehicle-road-network carbon reduction scheduling system considering the integration of new energy sources, characterized in that, The system includes: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor performs the method as described in any one of claims 1 to 7.

9. A computer-readable storage medium storing a processor-executable program, characterized in that, The processor-executable program, when executed by the processor, is used to perform the method as described in any one of claims 1 to 7.

Images