CN-122022313-A - Mobile storage and charging system scheduling and path planning method based on multi-objective optimization

CN122022313ACN 122022313 ACN122022313 ACN 122022313ACN-122022313-A

Abstract

The invention relates to the technical field of mobile storage and charging collaborative scheduling, and particularly discloses a mobile storage and charging system scheduling and path planning method based on multi-objective optimization, which comprises the following steps of responding to a charging demand request and state data of a renewable energy access point in an area, executing multi-source data acquisition and collaborative preprocessing, and generating a system state set containing space-time characteristics; based on the system state set, a multi-objective optimization system comprising a user side, an operation side, a power grid side, an energy storage side and a renewable energy source absorption side is constructed, and conflict resolution factors are introduced to balance conflict relations among targets. According to the invention, a multidimensional target system comprising users, operation, power grid and renewable energy consumption is constructed, and a layered collaborative optimization and conflict resolution mechanism is introduced, so that the comprehensive scheduling efficiency and robustness of the system in a complex dynamic environment are remarkably improved.

Inventors

GUO YANPING
SHAO YONG
LI SHAOQI
CHU BOWEN
HUANG XINYU

Assignees

易合(庐江)新能源科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260128

Claims (10)

1. The mobile storage and charging system scheduling and path planning method based on multi-objective optimization is characterized by being applied to a central control system of a mobile storage and charging system and comprising the following steps: responding to a charging demand request in an area and state data of a renewable energy access point, and executing multi-source data acquisition and collaborative preprocessing to generate a system state set containing space-time characteristics; Based on the system state set, constructing a multi-objective optimization system comprising a user side, an operation side, a power grid side, an energy storage side and a renewable energy source absorption side, introducing conflict resolution factors, and dynamically balancing conflict relations among targets based on gradient directions or weight duty ratios of the target functions; Inputting the multi-objective optimization system into a layered collaborative optimization model for solving, and outputting a scheduling decision and a path planning scheme aiming at the mobile storage and charging vehicle; issuing the scheduling decision and path planning scheme to the mobile storage and charging vehicle so as to trigger the mobile storage and charging vehicle to execute charging tasks, dynamic obstacle avoidance and renewable energy collaborative charging and discharging operations; The solving process of the hierarchical collaborative optimization model comprises the following steps: Starting global layer optimization, determining regional distribution of a mobile storage and charging vehicle and access point charging priority according to space-time prediction data of renewable energy sources by taking a first preset time period as a reference, and generating a global path frame; Starting local layer optimization, and refining the service sequence and obstacle avoidance path of a single mobile storage and charging vehicle according to the real-time output data of renewable energy sources and road network dynamic obstacle information by taking a second preset time period as a reference; And responding to the burst state signal in the system operation, triggering an emergency adjustment flow, and executing power adjustment in a first preset response time or task redistribution in a second preset response time.
2. The method of claim 1, wherein performing multi-source data acquisition and collaborative preprocessing to generate a set of system states including spatio-temporal features comprises: Collecting real-time output data, space-time prediction data and weather-related data of a renewable energy access point; smoothing the real-time output data by adopting a moving average method to inhibit data fluctuation noise; Invoking a preset spatial feature extraction network model to extract the spatial features of the real-time output data, and invoking a preset time sequence feature extraction network model to extract the time sequence features of the real-time output data; and generating renewable energy output fluctuation risk levels comprising at least three preset levels by combining the weather-related data, and combining the risk levels into the system state set.
3. The method of claim 1, wherein the constructing a multi-objective optimization system comprising a user side, an operation side, a grid side, an energy storage side, and a renewable energy consumption side comprises: Establishing a total objective function, wherein the total objective function is formed by multiplying a user side objective function, an operation side objective function, a power grid side objective function, an energy storage side objective function, a dynamic obstacle avoidance objective function, an abnormal loss objective function and a renewable energy source absorption objective function by corresponding dynamic weights respectively and then weighting and summing the obtained products; setting a conflict resolution factor for the total objective function, wherein the value of the conflict resolution factor is positioned in a preset numerical correction interval and is used for dynamically correcting numerical conflicts among all sub objective functions in the optimization process; The calculating logic of the renewable energy source absorption objective function is used for calculating and obtaining regional renewable energy source absorption rate based on the ratio of the sum of the power charged by the mobile storage and charging vehicle to the sum of the real-time output power of the renewable energy source and the maximum allowable power rejection power of the power distribution network, and maximizing the absorption rate.
4. The method of claim 1, wherein inputting the multi-objective optimization system into a hierarchical collaborative optimization model for solution further comprises applying renewable energy collaborative constraints, the renewable energy collaborative constraints comprising: defining that the charging power of the mobile storage vehicle at a renewable energy access point is smaller than or equal to the difference between the real-time output of the access point and the local load of the power distribution network, and defining the charging duration to cover the output duration of the access point; Defining that the total power of the mobile storage and charging vehicles simultaneously charged by a single renewable energy access point is less than or equal to a preset percentage threshold value of the rated capacity of the access point; And limiting the power change rate of the mobile storage and charging vehicle when the charging and discharging operation is performed not to exceed a preset power climbing rate threshold.
5. The method of claim 1, wherein the issuing the scheduling decision and path planning scheme to the mobile storage and charging vehicle to trigger the mobile storage and charging vehicle to perform charging tasks, dynamic obstacle avoidance, and renewable energy collaborative charging and discharging operations comprises: Judging the current electricity price state and the renewable energy output state; If electricity price is in valley section and renewable energy source output is excessive, generating a deep charging instruction, indicating the mobile storage and charging vehicle to prolong the charging time at an access point, and charging the battery state of charge to be above a first preset state of charge threshold; If the electricity price is at the peak section and the renewable energy source output is insufficient, a discharge service instruction is generated, the mobile storage and charging vehicle is instructed to preferentially use the self energy storage electric quantity to supply power for a user, and the charging time at the access point is shortened or the charging is stopped.
6. The method according to claim 1, wherein the hierarchical collaborative optimization model includes road network traffic constraints for path planning, and the construction process of the road network traffic constraints includes: Constructing a layered topological map, and establishing a first obstacle avoidance branch based on geometric space segmentation and semantic risk weight on the layered topological map; constructing a Riemann manifold graph, equivalently mapping the motion characteristics of dynamic obstacles in a road network into a Riemann curvature tensor, and constructing a second obstacle avoidance branch based on curvature calculation on the Riemann manifold graph; and when a path planning scheme is generated, monitoring the Riemann curvature tensor in real time, judging that a high dynamic risk exists if the Riemann curvature tensor is detected to exceed a preset curvature threshold value, and switching from the first obstacle avoidance branch to the second obstacle avoidance branch to carry out path re-planning.
7. The method of claim 1, wherein triggering the contingency adjustment procedure in response to the burst status signal during system operation comprises: Monitoring states of the mobile storage vehicle and a communication link in real time through a redundant communication module, and generating a primary abnormal signal when an abnormal index is monitored; responding to the primary abnormal signal, triggering a paging process of the associated mobile storage and charging vehicle, and executing secondary state confirmation; And if the abnormal index is not released after the paging process is executed, judging that the abnormal state is confirmed, and starting a task taking-over process, namely, reassigning the charging task associated with the abnormal vehicle to the mobile storage and charging vehicle which is in a normal running state and the position of which meets a preset dispatching threshold value in a preset assignment period, and synchronously releasing the power grid resources and the access point resources occupied by the abnormal vehicle.
8. The method of claim 1, wherein the solving of the hierarchical collaborative optimization model employs a transfer learning mechanism, comprising: constructing a scene migration library, wherein the scene migration library is pre-stored with characteristic parameters of various renewable energy typical scenes and corresponding pre-training model parameters; Before solving the optimization problem at the current moment, calculating the matching degree of the real-time scene characteristics and the typical scene characteristics in the scene migration library; If the matching degree is higher than a preset matching threshold, directly calling the corresponding pre-training model parameters to initialize the population of the optimization algorithm, and reducing the iteration times of the optimization algorithm.
9. The method of claim 1, wherein issuing the scheduling decision and path planning scheme includes generating a discharge service instruction, and wherein before executing the discharge service instruction, further executing a mobility security check procedure based on road network traffic potential energy, the mobility security check procedure including: Calculating the minimum energy consumption value required by the mobile storage and charging vehicle to travel from the current geographic position to the nearest bottom-protecting energy-supplementing node in real time based on the Riemann manifold graph, wherein the calculation of the minimum energy consumption value introduces a road network friction coefficient which is positively correlated with the Riemann curvature tensor and is used for representing the nonlinear amplification effect of traffic jam on the traveling energy consumption; The method comprises the steps of monitoring the current charge state of the mobile storage and charging vehicle in real time, and predicting the estimated residual charge state after executing the discharging service instruction, wherein if the estimated residual charge state is smaller than the product of the dynamic return energy consumption threshold and a preset road network fluctuation risk coefficient, a discharging overrule mechanism is triggered; The method comprises the steps of responding to a discharging overrule mechanism, intercepting a discharging service instruction forcedly, switching the working state of a mobile storage and charging vehicle into a mobility locking state, prohibiting the mobile storage and charging vehicle from outputting electric energy outwards in the mobility locking state, generating an emergency evacuation path planning scheme, and indicating the mobile storage and charging vehicle to drive to the bottom protection and energy supplement node immediately, wherein the bottom protection and energy supplement node is defined as a charging station or a power exchange station which is nearest to the current position in a road network and has special power supply resources.
10. The method according to claim 7, wherein the reassigning the charging task associated with the abnormal vehicle to the mobile storage vehicle in the normal running state and in the position satisfying the preset scheduling threshold within the preset assignment period, and synchronously releasing the power network resource and the access point resource occupied by the abnormal vehicle, comprises the following specific steps: when a task taking over process is started, aiming at the power grid resources and the access point resources occupied by the abnormal vehicle, intercepting a conventional resource release instruction, and marking the states of the power grid resources and the access point resources as a shadow locking state; Generating a resource inheritance token containing the power quota parameter and the unique verification signature, and issuing the resource inheritance token to the mobile storage and charging vehicle in a normal running state along with the redistributed charging task; Responding to the mobile storage and charging vehicle in a normal running state to reach the access point and submitting the resource inheritance token, wherein an edge computing unit where the access point is located executes atomic identity replacement operation, namely keeping the output power set value of the access point unchanged, and switching a charging and control object identifier from the abnormal vehicle to the mobile storage and charging vehicle in the normal running state in the same clock period; and only after a preset safety time window after the completion of the hot switching handshake or when the abnormal vehicle is detected to generate a physical disconnection signal, releasing the shadow locking state and allowing a new power adjustment request.

Description

Mobile storage and charging system scheduling and path planning method based on multi-objective optimization Technical Field The invention relates to the technical field of mobile storage and charging collaborative scheduling, in particular to a mobile storage and charging system scheduling and path planning method based on multi-objective optimization. Background Along with the rapid increase of the energy conservation of new energy automobiles and the large-scale grid connection of distributed renewable energy sources, the traditional fixed charging infrastructure is proved to be a new part when coping with sudden charging demands with uneven space-time distribution, and a mobile storage and charging system is generated. In the prior art, the dispatching of mobile storage vehicles generally only depends on a path optimization algorithm of a single target, such as a shortest path or a minimum time, or only considers the mobile storage vehicles as a simple mobile power supply, and the mobile storage vehicles cannot be deeply integrated into a source-network-load-storage interaction system. However, in actual operation, the mobile storage and charging system faces the complex challenge of multi-physical field coupling, namely, on one hand, the output of the distributed renewable energy source has extremely strong intermittence and volatility, and the charging demands of users also have high randomness in time and space, so that the two are difficult to match, and on the other hand, the traffic jam condition of the urban road network directly influences the energy consumption and the arrival time of the vehicle, so that the dispatching strategy purely based on the geographic distance is often ineffective. Therefore, the core technical conflict to be solved at present is how to balance pursuing the operation economy of the mobile storage and filling system and guaranteeing the mobile viability of the vehicle, and how to solve the problem of space-time dislocation between the release of logic resources and the response of a physical power grid in an abnormal take-over scene. In particular, in the prior art, when a high-income discharging task is executed, the rapid increase of energy consumption caused by abrupt change of traffic conditions of a road network is often ignored, the vehicle is easy to be blocked on a congestion road section due to electric quantity exhaustion, and meanwhile, when the task of processing equipment faults is handed over, due to lack of consideration on physical inertia of a power grid, a simple resource release instruction is easy to cause power preemption or superposition between new and old tasks, so that overload protection tripping of a power distribution network is triggered. Therefore, how to construct a scheduling method which can cooperate with a multidimensional target, has a traffic-energy deadlock prevention mechanism and can realize atomization seamless take-over is a major concern for the technicians in the field. Disclosure of Invention The present invention aims to solve at least one of the technical problems in the related art to some extent. Therefore, the invention aims to provide a mobile storage and filling system scheduling and path planning method based on multi-objective optimization so as to realize safe, efficient and continuous operation of the mobile storage and filling system under multi-domain coupling. In order to achieve the above objective, an embodiment of a first aspect of the present invention provides a mobile storage and filling system scheduling and path planning method based on multi-objective optimization, which is applied to a central control system of a mobile storage and filling system, including: responding to a charging demand request in an area and state data of a renewable energy access point, and executing multi-source data acquisition and collaborative preprocessing to generate a system state set containing space-time characteristics; based on the system state set, constructing a multi-objective optimization system comprising a user side, an operation side, a power grid side, an energy storage side and a renewable energy source absorption side, and introducing conflict resolution factors to balance conflict relations among targets; Inputting the multi-objective optimization system into a layered collaborative optimization model for solving, and outputting a scheduling decision and a path planning scheme aiming at the mobile storage and charging vehicle; issuing the scheduling decision and path planning scheme to the mobile storage and charging vehicle so as to trigger the mobile storage and charging vehicle to execute charging tasks, dynamic obstacle avoidance and renewable energy collaborative charging and discharging operations; The solving process of the hierarchical collaborative optimization model comprises the following steps: Starting global layer optimization, determining regional distribution of a mobile storage and charging v