Vehicle-mounted V2G system energy management and power grid interaction coordination optimization platform
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-01-05
- Publication Date
- 2026-03-31
- Estimated Expiration
- Not applicable · inactive patent
Smart Images
Figure CN121769976A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of energy management technology, and more specifically to a platform for coordinated optimization of energy management and grid interaction in a vehicle-to-grid (V2G) system. Background Technology
[0002] Currently, with the rapid growth of electric vehicle (EV) ownership, V2G technology, as a key link connecting the transportation and power systems, enables bidirectional energy exchange between EVs and the power grid, and has significant value in peak shaving, valley filling, and providing ancillary services. However, existing V2G systems have the following technical shortcomings: The data acquisition in the perception layer lacks an effective preprocessing mechanism, and the multi-source data has problems such as noise and outliers, resulting in low data reliability. The communication layer has a large transmission delay, which makes it difficult to meet the real-time interaction requirements of the V2G system, and the reliability of command interaction is insufficient. Decision-makers often use single-objective optimization models, which fail to take into account user benefits, grid balance and battery life, resulting in poor optimization performance. The execution layer has low control accuracy, large deviation between charging and discharging power and commands, and slow grid response speed.
[0003] Therefore, there is an urgent need to design a V2G system energy management and grid interaction coordination optimization platform that features multi-dimensional collaboration, comprehensive optimization objectives, and high control precision. Summary of the Invention
[0004] In view of this, the present invention provides a vehicle-mounted V2G system energy management and grid interaction coordination optimization platform to solve the problems existing in the background art.
[0005] To achieve the above objectives, the present invention adopts the following technical solution: A vehicle-to-grid (V2G) system energy management and grid interaction coordination optimization platform includes a sensing layer, a communication layer, a decision-making layer, and an execution layer, with each layer working in tandem. The perception layer is used to collect and preprocess multi-dimensional data, including vehicle battery status data, vehicle operating status data, real-time power grid operating data, and environmental perception data. The preprocessing algorithms include an outlier detection algorithm based on the 3σ criterion and a multi-source data fusion algorithm based on variance weighting. The communication layer is used to establish multi-link communication channels and achieve low-latency, high-reliability data transmission and command interaction through an adaptive transmission rate adjustment algorithm. The communication links include the links between the perception layer and the decision layer, the decision layer and the execution layer, the platform and the external power grid dispatch center and the vehicle terminal. The decision layer incorporates a multi-objective optimization model and an improved NSGA-Ⅲ solution algorithm. Based on the data preprocessed by the perception layer, it solves for the optimal charging and discharging control command and the grid-side coordinated response command. The multi-objective optimization model aims to maximize user benefits, balance grid load, and minimize battery life loss. The execution layer includes an on-board charging and discharging execution unit and a grid-side response unit. It tracks the instructions from the decision layer through PID control algorithm and droop control algorithm to complete the energy interaction between the V2G system and the grid.
[0006] Optionally, the outlier detection algorithm formula based on the 3σ criterion of the perception layer is as follows: If Then determine These are outliers, where For the first i One set of raw collected data, The mean of the data. The standard deviation of the data is given; outliers are replaced using linear interpolation, with the formula being: ,in The replaced outlier; These are two normal data points adjacent to an outlier.
[0007] Optionally, the formula for the variance-weighted multi-source data fusion algorithm of the perception layer is as follows:
[0008] in, , where n is the number of data sources The weights of the k-th class of data are... For the preprocessed data of the k-th class, Let be the variance of the k-th class of data. Optionally, the adaptive transmission rate adjustment algorithm formula of the communication layer is:
[0009] Where R is the adjusted transmission rate. As the reference transmission rate, Here, SNR is the channel adaptation coefficient, and SNR is the channel signal-to-noise ratio; the transmission delay constraint satisfies... ,in D represents the amount of data. Due to signal propagation delay, The maximum allowable delay. Optionally, the multi-objective optimization model of the decision layer includes an objective function and constraints: Objective function: Maximize user benefits:
[0010] Power grid load balancing:
[0011] in
[0012] Minimize battery life loss:
[0013] Where T is the optimization time period; Let t be the battery discharge power at time t; Let t be the electricity price sold by the power grid. The battery charging power at time t; Let t be the grid purchase price of electricity; Subsidies for auxiliary services at time t; Let t be the total load of the power grid at time t; Let t be the base load of the power grid; The amount of battery health loss at time t; This is the depth of discharge loss coefficient; Let t be the depth of battery discharge. ; P(t) is the power loss coefficient; P(t) is the absolute value of the charging and discharging power at time t. ; The remaining battery charge percentage at time t; Constraints: Battery charging and discharging power constraints:
[0014] Battery SOC constraints: ,and
[0015] Power grid security constraints:
[0016] Charge-discharge mutual exclusion constraint:
[0017] in, This is the maximum charging power; This represents the maximum discharge power. Minimum allowable SOC; Maximum permissible SOC; For charging efficiency; Discharge efficiency, valued between 0.9 and 0.95; Cap is the battery capacity; V(t) is the time step; V(t) is the grid node voltage at time t. The minimum permissible voltage is 0.95 times the rated voltage. This is the maximum permissible voltage, taken as 1.05 times the rated voltage; f ( t Let t be the grid frequency at time t; Minimum allowable frequency; This is the maximum permissible frequency.
[0018] Optionally, the improved NSGA-III solution algorithm for the decision layer includes the following steps: Population initialization: Generate N individuals with charging / discharging strategies that satisfy the constraints. ; Non-dominated sorting: based on objective function f 1. f 2. f 3 pairs of populations were subjected to non-dominated sorting to obtain the frontier layer. F 1. F 2、…; Crowding Calculation: Calculate the crowding level of each individual.
[0019] Adaptive crossover and mutation: The crossover operator uses simulated binary crossover, and the mutation operator uses polynomial mutation, where... Cross factor It is a variable factor; Elite preservation: Merge the parent and offspring populations, reorder them, and select the top N individuals to enter the next generation; Termination condition: The iteration reaches the maximum number of iterations, and the compromise solution in the optimal frontier layer is output as the control command. Optionally, the PID control algorithm formula for the execution layer is:
[0020] in , To control the output signal, This is the proportionality coefficient. The integral coefficient is... These are the differential coefficients. For the target power of the decision-making level, This represents the actual charging and discharging power. Optionally, the grid-side droop control algorithm formula for the execution layer is:
[0021] in This represents the grid-side response power. This is the grid reference response power. This is the frequency droop factor. The rated frequency of the power grid. f( t Let t be the power grid frequency at time t.
[0022] As can be seen from the above technical solution, compared with the prior art, this invention discloses a vehicle-to-grid (V2G) system energy management and grid interaction coordination optimization platform, including a perception layer, a communication layer, a decision-making layer, and an execution layer, each layer working in tandem. The perception layer improves data reliability through outlier detection and multi-source data fusion algorithms, providing high-quality support for decision optimization. The communication layer adopts an adaptive transmission rate adjustment algorithm to achieve low-latency, high-reliability command and data interaction, meeting the real-time interaction requirements of V2G. The multi-objective optimization model of the decision-making layer takes into account user benefits, grid balance, and battery life, and improves the NSGA-Ⅲ algorithm to enhance the diversity and convergence speed of optimization solutions. The execution layer improves the charging and discharging power tracking accuracy and grid response speed through PID control and droop control algorithms, ensuring stable interaction between the V2G system and the grid. This invention can be widely applied to electric vehicle charging stations, distributed energy systems, smart grid dispatching, and other scenarios, and has significant economic and social value. Attached Figure Description
[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.
[0024] Figure 1 This is a schematic diagram of the platform structure provided by the present invention. Detailed Implementation
[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0026] This invention discloses a vehicle-to-grid (V2G) system energy management and grid interaction coordination optimization platform, such as... Figure 1 As shown, it includes a perception layer, a communication layer, a decision-making layer, and an execution layer, with each layer working in tandem; among which: The perception layer is used to collect and preprocess multi-dimensional data, including vehicle battery status data, vehicle operating status data, real-time power grid operation data, and environmental perception data. The preprocessing algorithms include an outlier detection algorithm based on the 3σ criterion and a multi-source data fusion algorithm based on variance weighting. The communication layer is used to establish multi-link communication channels and achieve low-latency, high-reliability data transmission and command interaction through an adaptive transmission rate adjustment algorithm. The communication links include the links between the perception layer and the decision layer, the decision layer and the execution layer, the platform and the external power grid dispatch center, and the vehicle terminal. The decision-making layer incorporates a multi-objective optimization model and an improved NSGA-Ⅲ solution algorithm. Based on the data preprocessed by the sensing layer, it solves for the optimal charging and discharging control command and the grid-side coordinated response command. The multi-objective optimization model aims to maximize user benefits, balance grid load, and minimize battery life loss. The execution layer includes an on-board charging and discharging execution unit and a grid-side response unit. It tracks the instructions from the decision-making layer through PID control algorithm and droop control algorithm to complete the energy interaction between the V2G system and the grid.
[0027] In one specific embodiment, the perception layer includes a battery status sensor, a vehicle operating status sensor, a power grid parameter sensor, and an environmental sensor (temperature and humidity sensor) to collect multi-dimensional raw data and improve data quality through preprocessing algorithms.
[0028] Outlier detection algorithm (3σ criterion) principle: Assuming that the collected data follows a normal distribution, values that exceed the mean ± 3 times the standard deviation are judged as outliers and replaced by linear interpolation.
[0029] Outlier detection:
[0030] Outlier replacement
[0031] In the formula, For the first i One set of raw collected data, The mean of the data. The standard deviation of the data; The replaced outlier; These are two normal data points adjacent to an outlier.
[0032] The principle of multi-source data fusion algorithm: weights are assigned according to the variance (reliability) of different sensor data. The smaller the variance, the greater the weight, and the more accurate the data is obtained after fusion.
[0033] Fusion results:
[0034] Weight calculation:
[0035] In the formula, n represents the number of data sources. The weights of the k-th class of data are... For the preprocessed data of the k-th class, Let be the variance of the k-th class of data. In a specific embodiment, the communication layer adopts the 5G+LTE-V2X dual-mode communication protocol to establish a multi-link communication channel, and optimizes the transmission delay through an adaptive transmission rate adjustment algorithm to ensure real-time interaction between data and instructions.
[0036] The principle of adaptive transmission rate adjustment algorithm is to dynamically adjust the transmission rate based on the channel signal-to-noise ratio (SNR) to maximize transmission reliability while meeting delay constraints.
[0037] Transmission rate:
[0038] Transmission delay:
[0039] In the formula, R is the adjusted transmission rate. As the reference transmission rate, Here, SNR is the channel adaptation coefficient, and D is the channel signal-to-noise ratio; D is the data volume. Due to signal propagation delay, The maximum allowable delay is defined. In one specific embodiment, the decision layer is the core of the platform, which incorporates a multi-objective optimization model and an improved NSGA-Ⅲ solution algorithm. Based on the preprocessed data from the sensing layer, it generates optimal charging and discharging control commands and grid-side coordinated response commands.
[0040] The multi-objective optimization model for the decision-making level includes the objective function and constraints: Objective function: Maximize user benefits:
[0041] Power grid load balancing:
[0042] in
[0043] Minimize battery life loss:
[0044] Where T is the optimization time period; Let t be the battery discharge power at time t; Let t be the electricity price sold by the power grid. The battery charging power at time t; Let t be the grid purchase price of electricity; Subsidies for auxiliary services at time t; Let t be the total load of the power grid at time t; Let t be the base load of the power grid; The amount of battery health loss at time t; This is the depth of discharge loss coefficient; Let t be the depth of battery discharge. ; P(t) is the power loss coefficient; P(t) is the absolute value of the charging and discharging power at time t. ; The remaining battery charge percentage at time t; Constraints: Battery charging and discharging power constraints:
[0045] Battery SOC constraints: ,and
[0046] Power grid security constraints:
[0047] Charge-discharge mutual exclusion constraint:
[0048] in, This is the maximum charging power; This represents the maximum discharge power. Minimum allowable SOC; Maximum permissible SOC; For charging efficiency; Discharge efficiency, valued between 0.9 and 0.95; Cap is the battery capacity; V(t) is the time step; V(t) is the grid node voltage at time t. The minimum permissible voltage is 0.95 times the rated voltage. This is the maximum permissible voltage, taken as 1.05 times the rated voltage; f ( t Let t be the grid frequency at time t; Minimum allowable frequency; This is the maximum permissible frequency.
[0049] In a specific embodiment, the improved NSGA-III solution algorithm for the decision layer includes the following steps: Population initialization: Generate N individuals with charging / discharging strategies that satisfy the constraints. ; Non-dominated sorting: based on objective function f 1. f 2. f 3. Perform non-dominated sorting on the population to obtain the frontier layer, F 2、…;F 1 is the optimal frontier layer.
[0050] Crowding Calculation: Calculate the crowding level of each individual.
[0051] The higher the crowding level, the more sparsely distributed the individuals are in the population; Adaptive crossover mutation: The crossover operator uses simulated binary crossover. ,in Cross factor; The mutation operator employs polynomial mutation. ,in It is a variable factor; Elite preservation: Merge the parent and offspring populations, reorder them, and select the top N individuals to enter the next generation; Termination condition: Iterations reach the maximum number of iterations. The optimal frontier layer outputs a compromise solution (based on fuzzy membership function selection) as a control command. In a specific embodiment, the execution layer includes an on-board charging and discharging execution unit (OBC on-board charger, bidirectional DC / DC converter) and a grid-side response unit (SVG static var generator, energy storage converter), which accurately executes the decision commands through PID control algorithm and droop control algorithm.
[0052] The principle of PID charge / discharge power tracking algorithm: It eliminates the deviation between the target power and the actual power through proportional-integral-derivative control, thereby achieving accurate tracking of charge / discharge power.
[0053] The formula is:
[0054] in , To control the output signal, This is the proportionality coefficient. The integral coefficient is... These are the differential coefficients. For the target power of the decision-making level, This represents the actual charging and discharging power. The principle of the grid-side droop control algorithm is to dynamically adjust the grid-side response power based on the grid frequency deviation to maintain grid frequency stability.
[0055] The formula is:
[0056] in This represents the grid-side response power. This is the grid reference response power. This is the frequency droop factor. The rated frequency of the power grid. f ( t Let t be the power grid frequency at time t.
[0057] The platform operation process is as follows: The sensing layer collects raw data such as battery SOC, vehicle speed, grid voltage / frequency, and ambient temperature. Outliers are removed using the 3σ criterion, and the data is then fused using the variance-weighted average method to obtain preprocessed data. The communication layer transmits preprocessed data to the decision layer via the 5G link, and simultaneously transmits the decision layer's instructions to the execution layer. The transmission rate is dynamically adjusted based on SNR to ensure that the latency is ≤30ms. The decision-making level inputs preprocessed data, solves the problem using a multi-objective optimization model and an improved NSGA-Ⅲ algorithm, and generates hourly charge / discharge power commands for 24 hours. and grid-side response power command ; The on-board charging and discharging unit in the execution layer tracks the charging and discharging power command through a PID algorithm, while the grid-side response unit adjusts the response power through a droop control algorithm to complete the energy interaction between the V2G system and the grid. The perception layer collects real-time operational data from the execution layer and feeds it back to the decision-making layer, forming a closed-loop optimization.
[0058] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.
[0059] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A vehicle-mounted V2G system energy management and grid interaction coordination optimization platform, characterized in that, Comprise a perception layer, a communication layer, a decision layer and an execution layer, which are cooperated in turn; wherein: The perception layer is used for collecting multi-dimensional data and pre-processing, the multi-dimensional data includes vehicle-mounted battery state data, vehicle running state data, real-time operation data of power grid and environment perception data, the pre-processing algorithm includes an outlier detection algorithm based on 3σ criterion and a multi-source data fusion algorithm based on variance weighting; The communication layer is used for establishing multi-link communication channels, realizing low-delay and high-reliable data transmission and instruction interaction through an adaptive transmission rate adjustment algorithm, and the communication link includes the link between the perception layer and the decision layer, the link between the decision layer and the execution layer, the link between the platform and the external power grid dispatching center and the link between the vehicle-mounted terminal; The decision layer is internally provided with a multi-objective optimization model and an improved NSGA-Ⅲ solving algorithm, based on the pre-processed data of the perception layer, the optimal charging and discharging control instruction and the power grid side coordinated response instruction are obtained, the multi-objective optimization model takes the maximization of user benefits, the equalization of power grid load and the minimization of battery life loss as the target; The execution layer includes a vehicle-mounted charging and discharging execution unit and a power grid side response unit, which track the decision layer instruction through PID control algorithm and droop control algorithm, and complete the energy interaction between V2G system and power grid. 2.The energy management and grid interaction coordination optimization platform of a vehicle-mounted V2G system according to claim 1, characterized in that, The abnormal value detection algorithm formula of the perception layer based on 3σ criterion is: if , it is determined that is an abnormal value, wherein is the i original collected data, is the data mean, is the data standard deviation; the abnormal value is replaced by a linear interpolation method, and the formula is , wherein is the replaced abnormal value; is the two normal data adjacent to the abnormal value. 3.The energy management and grid interaction coordination optimization platform of a vehicle-mounted V2G system according to claim 1, characterized in that, The multi-source data fusion algorithm formula of the perception layer based on variance weighting is: wherein, n is the number of data sources, is the weight of the kth data, is the kth pre-processed data, is the variance of the kth data. 4.The energy management and grid interaction coordination optimization platform of a vehicle-mounted V2G system according to claim 1, wherein, The adaptive transmission rate adjustment algorithm formula of the communication layer is: where R is the adjusted transmission rate, is the reference transmission rate, is the channel adaptation coefficient, SNR is the channel signal-to-noise ratio; the transmission delay constraint satisfies where D is the data amount, is the signal propagation delay, is the maximum allowed delay. 5.The energy management and grid interaction coordination optimization platform of a vehicle-mounted V2G system according to claim 1, wherein, The multi-objective optimization model of the decision layer includes objective function and constraint condition: Objective function: Maximization of user benefits: Equalization of power grid load: wherein ; Minimization of battery life loss: Wherein, T is the optimization time period; is the battery discharging power at time t; is the grid electricity selling price at time t; is the battery charging power at time t; is the grid electricity buying price at time t; is the auxiliary service subsidy at time t; is the total grid load at time t; is the grid base load at time t; is the battery health state loss amount at time t; is the discharging depth loss coefficient; is the battery discharging depth at time t, ; is the power loss coefficient; P(t) is the absolute value of the charging and discharging power at time t, ; is the battery remaining capacity percentage at time t; Constraint condition: Battery charging and discharging power constraint: Battery SOC constraint: , and Power grid safety constraint: Charging and discharging mutual exclusion constraint: in, This is the maximum charging power; This represents the maximum discharge power. Minimum allowable SOC; Maximum permissible SOC; For charging efficiency; Discharge efficiency, valued between 0.9 and 0.95; Cap is the battery capacity; V(t) is the time step; V(t) is the grid node voltage at time t. Minimum allowable voltage; Maximum allowable voltage; f ( t Let t be the grid frequency at time t; Minimum allowable frequency; This is the maximum permissible frequency. 6.The energy management and grid interaction coordination optimization platform for a vehicle-mounted V2G system according to claim 1, wherein, The improved NSGA-Ⅲ solving algorithm of the decision layer includes the following steps: Initialization of population: generate N charging and discharging strategy individuals that meet the constraint condition; Non-dominated sorting: based on objective function f 1, f 2, f 3 Non-dominated sorting of the population, resulting in a front layer F 1, F 2,...; Crowding degree calculation: calculate the crowding degree of each individual; Adaptive crossover and mutation: the crossover operator adopts simulated binary crossover, and the mutation operator adopts polynomial mutation; Elite reservation: combine the parent and child populations, reorder and select the first N individuals into the next generation; Termination condition: iterate to the maximum iteration number, output the compromise solution in the optimal front layer as the control instruction. 7.The energy management and grid interaction coordination optimization platform of a vehicle-mounted V2G system according to claim 1, wherein, The PID control algorithm formula of the execution layer is: wherein , is a control output signal, is a proportional coefficient, is an integral coefficient, is a differential coefficient, is a decision layer target power, is an actual charge and discharge power. 8.The energy management and grid interaction coordination optimization platform of a vehicle-mounted V2G system according to claim 1, wherein, The power grid side droop control algorithm formula of the execution layer is: wherein is the grid-side response power, is the grid reference response power, is the frequency droop coefficient, is the grid rated frequency, f t is the grid frequency at time t.