An event-driven scheduling-based mobile energy storage and mobile substation collaborative operation optimization method

Abstract

The application discloses a kind of based on event-driven scheduling mobile energy storage and mobile substation collaborative operation optimization method, belong to energy storage and distribution network operation field.The method of the present application first analyzes the structure and emergency capability of both, classifies and enters parameter to module library, quantitatively evaluates emergency capability;Again, the operation mode of both is dispatched, access, modeling and constraint setting are adjusted layer;Finally, based on switching system theory, map distribution network state as subsystem mode, switch mode by preset switching law when event occurs, call optimization model scheduling.The application realizes event-driven automatic identification and mode switching, improves calculation efficiency and resource utilization, reduces energy waste, enhances the reliability and economy of distribution network.

Description

Technical Field

[0001] This invention belongs to the field of energy storage and distribution network operation, specifically involving a collaborative operation optimization method for mobile energy storage and mobile substations based on event-driven scheduling. Background Technology

[0002] With the increasing penetration of renewable energy into distribution networks and the growing diversification of distributed energy access and load patterns, the operating environment of distribution networks is becoming increasingly complex and variable. Against this backdrop, mobile energy storage systems and mobile substations, with their flexible deployment and rapid response capabilities, have become crucial resources for emergency power supply, fault recovery, and dynamic peak and frequency regulation in distribution networks. Mobile energy storage can rapidly release or absorb electrical energy when the grid experiences short-term power shortages or sudden load surges; while mobile substations can connect between different voltage levels and achieve voltage regulation. Their synergistic application not only overcomes the limitations of traditional fixed facilities but also provides mobile, reusable, and efficient power backup for the distribution network during natural disasters, equipment maintenance, or emergencies.

[0003] However, existing technologies have not yet formed a systematic optimization framework for the coordinated operation of mobile energy storage and mobile substations. Most scheduling strategies only provide simple configurations for single resources or static scenarios, making it difficult to take into account the real-time coordination of the two types of equipment under event-driven conditions. The results often manifest as energy scheduling redundancy, insufficient consideration of converter and transformer delays, and low resource utilization. Especially in complex scenarios with multiple faults or concurrent events, rapid decision-making is urgently needed, but the lack of coordination mechanisms leads to slow response, which in turn affects the stability and economy of the distribution network. Summary of the Invention

[0004] Based on the above shortcomings, this invention aims to propose a collaborative operation optimization method for mobile energy storage and mobile substations based on event-driven scheduling. This method solves the problem that existing strategies for collaborative optimization frameworks of mobile energy storage and mobile substations are designed for single devices or static scenarios, resulting in low resource utilization. At the same time, it also solves the problems of slow response, difficulty in handling multiple concurrent events, impact on distribution network stability, and difficulty in balancing reliability and economy under event-driven scheduling.

[0005] The technical solution adopted in this invention is as follows: A collaborative operation optimization method for mobile energy storage and mobile substations based on event-driven scheduling, comprising the following steps:

[0006] S1. Conduct structural and emergency response capability analysis on mobile energy storage and mobile substations, classify the types of mobile energy storage and mobile substations respectively, collect key performance parameters and enter them into the module library in the specified format, and then carry out quantitative assessment of emergency response capability based on the modular resource model, calculate the emergency response capability index and integrate it into the resource model.

[0007] S2. Model the operation mode of mobile substations and mobile energy storage in three layers: dispatch layer, access layer, and regulation layer. Set corresponding constraints in each layer to build an operation model.

[0008] S3. Based on the switching system theory, the operating state variables of the distribution network are mapped to several subsystem modes. By pre-setting the switching law, the system automatically switches to the corresponding mode when different events occur, and calls the corresponding optimization model to achieve customized collaborative scheduling.

[0009] Furthermore, in step S1,

[0010] When performing structural analysis on mobile energy storage, it is divided into two types: integrated and separable. The key performance parameters collected include the rated capacity, maximum charge and discharge power, physical dimensions, mobile platform compatibility information, maximum number of start-stop cycles, start response time, and continuous discharge duration for each type of battery compartment. The input format is "compartment number-vehicle type-performance parameters".

[0011] When performing structural analysis on mobile substations, they are classified into four types: single-vehicle loading, two-vehicle loading, three-vehicle loading, and compartmentalized loading. The information collected includes: main transformer capacity, switch equipment parameters, medium-voltage distribution unit specifications, load and mobility performance of each vehicle, start-up delay and continuous operation capability. The input format is "loading method-vehicle number-equipment parameters".

[0012] When conducting quantitative assessments of emergency response capabilities, simulation schemes for two typical scenarios—active peak shaving and passive emergency response—are developed. The equipment delivery path, loading and unloading process, and grid-connected operation are simulated in the distribution network nodes and road transportation network. The transportation time, start-up delay, energy output curve, and utilization rate are recorded. The emergency response capability index for each type of module is calculated through statistical analysis.

[0013] Furthermore, in step S2,

[0014] When modeling the mobile substation dispatch layer, binary decision variables are introduced to characterize the instructions issued by the vehicle from the base to the fault node. A vehicle dispatch uniqueness constraint is set to ensure that a vehicle of a set of substations can only be dispatched to a node where an event occurs at any given time. At the same time, an immediate response constraint is set to ensure that at least one set of mobile substations is dispatched from the origin after an event occurs.

[0015] When modeling the access layer of a mobile substation, set cumulative dispatch constraints to ensure that vehicles arrive at the destination node after being dispatched, set arrival time constraints to specify the arrival time of vehicles, set all vehicle integration constraints to ensure that access is only allowed after all necessary vehicles have arrived and equipment setup time has been considered, and set access status retention constraints to ensure that the access status is maintained after access unless special operation is performed.

[0016] When modeling the regulating layer of a mobile substation, an adjustable access constraint is set to allow the on-load tap changer to start voltage regulation only after the mobile substation is connected. An on-load tap changer regulation constraint is set to clarify the selection and regulation rules of the on-load tap changer tap position and limit the maximum single adjustment amount of the tap position.

[0017] The mobile energy storage dispatch layer sets vehicle dispatch uniqueness constraints, immediate response constraints, and turnaround transportation constraints to ensure reasonable vehicle dispatch and fuel consumption does not exceed limits.

[0018] The mobile energy storage access layer adopts cumulative dispatch constraints and arrival time constraints similar to those of mobile substations;

[0019] The mobile energy storage regulation layer sets traffic constraints to ensure reasonable charging and discharging actions and the number of connected modules. It sets charging and discharging and battery energy storage health constraints to limit charging and discharging parameters and battery status. It sets a unique constraint on the charging and discharging status of integrated modules to ensure that the charging and discharging action commands are consistent when multiple modules are connected to the same node.

[0020] Furthermore, in step S3,

[0021] The defined state variables include the net load of the node at time t, the working status flag of the distribution network line, the node voltage at time t during day-ahead dispatch of the distribution network, the maintenance status flag of the substation at the node, and the status flag of the temporary increase in power supply to the load at the node. The net load of the node is calculated by the node load power and the renewable energy power generation power.

[0022] The preset switching law corresponds to different events, including load / distributed generation over-limit events, line fault events, voltage regulation events, substation maintenance events, and temporary service power supply events. Different subsystems and optimization models are invoked for different events. The optimization model is a mixed integer linear programming problem. By importing resource parameters, network topology, and fault event data, a commercial solver is selected to solve the problem, generating a scheduling sequence and performing robustness analysis.

[0023] Furthermore, in step S3,

[0024] After the commercial solver completes the solution, it extracts the optimal decision variable values, generates specific scheduling sequences for the cabin and vehicles, calculates the energy output, transportation routes, and arrival times for each time period, draws a scheduling Gantt chart, and summarizes the key performance indicators of operating costs, energy benefits, and response time to form a complete report. Finally, it conducts robustness analysis on sensitive parameters such as transportation delay, cabin availability, and fault response window to evaluate the stability of the scheme to parameter fluctuations. The commercial solver solution process is then integrated into the online scheduling platform, which quickly calls the commercial solver interface when new events occur or parameters are updated, and iteratively generates the adjusted optimal scheduling scheme.

[0025] The beneficial effects and advantages of this invention are as follows: On the one hand, based on switching system theory and a hierarchical scheduling framework of dispatch layer-access layer-regulation layer, this invention achieves event-driven automatic identification and mode switching of mobile energy storage and mobile substations, avoiding global large-scale optimization, thereby enabling rapid response within minutes to hours and significantly improving computational efficiency. On the other hand, through customized target optimization for various scenarios such as overload, fault, voltage regulation, maintenance, and temporary capacity expansion, it organically integrates load reduction, network loss minimization, voltage deviation correction, and node power supply guarantee, achieving resource synergy, reducing energy waste, and taking into account economic efficiency, significantly enhancing the reliability and operational efficiency of the distribution network. This invention significantly improves the utilization rate of mobile energy storage and mobile substations, significantly reduces energy loss and reserve capacity requirements of the distribution network, and effectively compresses operating costs; it can further reduce dispatch response latency under various event scenarios; it can further reduce the risk of power outages and the probability of voltage exceeding limits in the distribution network, and enhances the overall operational reliability and economy of the system. Attached Figure Description

[0026] Figure 1 This is an overall flowchart of the present invention;

[0027] Figure 2 This is a schematic diagram of the event switching law of the present invention.

[0028] Detailed implementation method

[0029] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings.

[0030] Example 1:

[0031] like Figure 1 As shown, a collaborative operation optimization method for mobile energy storage and mobile substations based on event-driven scheduling is described, with the following specific steps:

[0032] Step 1: Conduct structural and emergency response capability analysis on mobile energy storage and mobile substations, classify them into different types, collect key performance parameters and enter them into the module library in a specified format, then conduct a quantitative assessment of emergency response capability based on the modular resource model, calculate the emergency response capability index and integrate it into the resource model; details are as follows:

[0033] First, a modular structure analysis of the mobile energy storage system is conducted, and all available devices are divided into two main categories: integrated and separable.

[0034] Integrated type refers to the battery compartment being fixedly installed in one piece with the chassis of the transport vehicle, which can be quickly deployed without on-site disassembly.

[0035] The detachable type designs the battery compartment as a standardized independent module, which can be loaded, unloaded and transported in parallel by various matching vehicles, so as to realize the stacking of compartments as needed and flexible expansion.

[0036] For each specific cabin model, detailed data must be collected and recorded, including its rated capacity, maximum charging and discharging power, physical dimensions (length × width × height), compatible vehicle platform type, maximum number of start-stop cycles, start-up response time (delay from receiving the command to being ready), and continuous discharge duration, among other key performance indicators. All parameters are uniformly entered into the resource module library in the format of "cabin number—vehicle type—performance parameters" to ensure that the corresponding cabin model can be accurately called and resources allocated according to the equipment characteristics during subsequent scheduling.

[0037] Modular structure analysis was also conducted on mobile substations, which were classified into four types: single-vehicle loading, two-vehicle loading, three-vehicle loading, and compartmentalized loading.

[0038] In the single-vehicle loading mode, all high-voltage switches, main transformers and medium-voltage power distribution units are loaded onto a single tractor.

[0039] In the dual-vehicle loading mode, the main transformer and the distribution cabinet are loaded on two separate vehicles, each with independent operating capabilities.

[0040] The three-vehicle loading mode places the high-voltage switchgear, main transformer and distribution cabinet in three separate modules on three vehicles to meet the requirements of high voltage and large capacity.

[0041] Compartmental loading further involves making the equipment into detachable compartmental units for flexible combination.

[0042] For each vehicle number under each loading method, data such as main transformer capacity, rated current and short-circuit breaking capacity of switchgear, rated voltage and rated current of medium-voltage distribution unit, vehicle load and mobility performance, start-up delay and continuous operation capability need to be collected and entered into the same module library in the format of "loading method - vehicle number - equipment parameters" to ensure that appropriate substation resource units can be quickly matched and deployed under different voltage levels and capacity requirements.

[0043] After completing the resource model construction, a quantitative assessment of emergency response capabilities based on modular libraries was conducted.

[0044] First, simulation schemes are designed for two typical operating scenarios: active peak shaving and passive emergency response. Based on the distribution of distribution network nodes and road transport network information, the transportation time, loading and unloading process and grid-connected operation process of each resource module on multiple selectable paths are simulated.

[0045] Secondly, the total transportation time, start-up delay, energy output curve in the power grid, and operation utilization rate of each resource unit are recorded in real time from order acceptance to delivery.

[0046] Finally, based on the collected simulation data, the emergency response capability index of each type of module is calculated using normalization and weighted scoring methods, and the index is integrated back into the modular resource library.

[0047] Step Two: Model the operation of mobile substations and mobile energy storage in three layers: dispatch layer, access layer, and regulation layer. Set corresponding constraints in each layer to construct the operation model; details are as follows:

[0048] The design employs a three-layer scheduling approach—dispatch layer, access layer, and regulation layer—to characterize the entire process of a mobile substation from receiving a service request to initiating regulation.

[0049] First, let's introduce the following variables: This indicates whether the c-th vehicle from the s-th set of mobile substations of type k is dispatched from node i to the destination node j at time t. This variable is a binary decision variable (0 or 1) and is instantaneous, meaning that the command is only issued at a certain moment. It is assumed that after dispatching, this variable is reset to 0 at the next moment. This indicates whether the c-th vehicle in the s-th mobile substation has arrived at the destination node j at time t. This variable is a binary decision variable (0 or 1). This variable is state-type. If the vehicle has arrived at the destination, this variable is always 0; if it has not arrived, it is 1. This indicates whether the s-th mobile substation is connected to the power grid at time t. This variable is a binary decision variable (0 or 1). If the mobile substation is connected, it is always 1, and if it is disconnected, it is 0.

[0050] The mobile substation dispatch layer mainly models the issuance of dispatch instructions for mobile substations.

[0051] Vehicle dispatch uniqueness constraint:

[0052] (1)

[0053] In the formula: The set of nodes where events have occurred; To configure a set of mobile substations, the above formula ensures that a single vehicle from a set of substations can only be dispatched to one node where an event has occurred at any given time.

[0054] Immediate response constraints:

[0055] (2)

[0056] In the formula: The time when the event occurred; This represents the time elapsed from the occurrence of an event to the receipt of the event message by maintenance personnel. The above formula indicates that for all nodes where events occur, at least one mobile substation must be dispatched from the starting point i after the event occurs to ensure a fault response.

[0057] The mobile substation access layer mainly models the issuance of instructions for connecting to the power grid when a mobile substation arrives at a predetermined location.

[0058] Cumulative dispatch constraint:

[0059] (3)

[0060] The above formula ensures A value of 1 indicates that if vehicle c at any time If ≤t is dispatched, the cumulative sum will be positive, leading to... It must be 1 to indicate that the vehicle has arrived at the destination node j.

[0061] (4)

[0062] The above equation represents the arrival state constraint, which ensures that if If the value at time t is 1, then there must be at least one dispatch action at or before this time. If there are no dispatch actions, It cannot be 1 because no vehicles have been dispatched.

[0063] Arrival time constraints:

[0064] (5)

[0065] In the above formula, =1 indicates that the c-th vehicle of the s-th mobile substation has arrived at its destination, either before time t.

[0066] All vehicle integrated constraints:

[0067] (6)

[0068] The above formula means that access is only allowed after all necessary vehicles have arrived. It indicates that for the s-th mobile substation, access is only permitted when all c vehicles have arrived at time t - If all devices arrive at their destination before time t (considering device setup time), The talent value is 1.

[0069] Access state maintenance constraints:

[0070] (7)

[0071] The above formula ensures that if a mobile substation is already connected, it should remain connected unless there is a special operation. This refers to the access time.

[0072] The mobile substation regulation layer initiates regulation actions when the mobile substation reaches the node where the event occurs.

[0073] Access adjustable constraints:

[0074] (8)

[0075] The above formula indicates that the on-load tapchanger (OLTC) of the mobile substation s can only start voltage regulation when the mobile substation s is in the connected state. Let n be a 0-1 variable representing the on-load tap position of the voltage regulator during time period t. Let be the set of on-load tap changer positions for the s-th mobile substation.

[0076] On-load tap changer regulation constraints:

[0077] (9)

[0078] (10)

[0079] In the formula: The voltage regulator ratio for time period t; The on-load tap changer position is determined by its 0-1 variables; Equation (8) indicates that the on-load tap changer position is selected by its 0-1 variables, and the on-load tap changer is only in one position at any given time; and These are 0-1 variables representing the on-load tap changer position increasing and decreasing during time period t; This represents the maximum adjustment amount for a single gear shift.

[0080] The design employs a three-layer scheduling approach—dispatch layer, access layer, and regulation layer—to characterize the entire process of mobile energy storage from receiving demand to entering regulation.

[0081] First, let's introduce the following variables: This indicates whether, at time t, node i dispatches the u-th energy storage module and uses the c-th vehicle to transport it to the destination node j. This indicates whether the u-th energy storage module arrives at its destination j at time t during the transport of the c-th vehicle; Node j indicates whether the u-th energy storage module is connected to the power grid at time t.

[0082] Mobile energy storage dispatch layer:

[0083] Vehicle dispatch uniqueness constraint:

[0084] (11)

[0085] In the formula: A set of nodes that require mobile energy storage services; The set of nodes for configuring mobile energy storage modules; u represents the set of nodes in the energy storage set. The energy storage module index. The set of nodes that need to be served can be represented as follows:

[0086] (12)

[0087] In the formula: l is the line number; A collection of lines requiring mobile energy storage for emergency response; For the set of routes; It is a set of nodes containing mobile energy storage.

[0088] Immediate response constraints:

[0089] (13)

[0090] In the formula: The time of the failure; The time from the occurrence of a fault to the time when maintenance personnel receive the fault message.

[0091] The above formula indicates that for all sets of nodes that have failed, at least one set of mobile energy storage must be dispatched from the starting point i after the failure occurs to ensure a response to the failure.

[0092] Turnaround transport constraints:

[0093] (14)

[0094] The above formula indicates that if vehicle c transports energy storage module u from node i to node j and returns or returns to other configured energy storage power stations, its total fuel consumption will not exceed its fuel capacity. . This refers to fuel consumption for a one-way trip. Here... This indicates the operation where the vehicle returns to i or continues to the next destination after completing the task at destination j.

[0095] Mobile energy storage access layer: using cumulative dispatch constraints and arrival time constraints similar to those of mobile substations.

[0096] Mobile energy storage regulation layer:

[0097] Traffic constraints:

[0098] (15)

[0099] (16)

[0100] Equation (15) ensures that, under the regulation of the mobile energy storage module u, only one charging / discharging action can be performed within each time interval. Equation (16) ensures that there is an upper limit to the number of mobile energy storage modules that can be inserted. .

[0101] Mobile energy storage charging and discharging and battery energy storage state of health (SOH) constraints:

[0102] (17)

[0103] (18)

[0104] (19)

[0105] (20)

[0106] (twenty one)

[0107] (twenty two)

[0108] In the formula: , These represent the charging and discharging variables, which are 0 and 1, respectively. The energy of the mobile energy storage device u at time t; , These are the lower and upper limits of the State of Charge (SOC) for mobile energy storage, respectively. , These are the charging and discharging efficiencies, respectively. For time step; , These are the upper limits of the mobile energy storage rate and the discharge rate, respectively. The average SOC; , These are the upper and lower limits of the average SOC, respectively.

[0109] The module integrates a unique constraint on charge / discharge states:

[0110] (twenty three)

[0111] The above formula represents the module index; it means that if multiple modules are connected to the same node at the same time, the charging or discharging commands of these modules must be the same at each point in time.

[0112] Step 3: Based on switching system theory, the operating state variables of the distribution network are mapped to several subsystem modes. By using a preset switching law, the system automatically switches to the corresponding mode when different events occur, and calls the corresponding optimization model to achieve customized collaborative scheduling, as detailed below: Figure 2 As shown, a switching system can generally be designed as multiple subsystems, as expressed by the following formula:

[0113] (twenty four)

[0114] in, For system state variables; The system input variable has the following switching law: ={( ,…, The state variable is defined as a function of f, and follows the switching law. Perform subsystem switching. These are the switching time values ​​for different subsystems; The mode number of the subsystem indicates the subsystem during the overall system switchover; Indicates a specific subsystem number. At a certain moment... The system switches to a subsystem according to the switching law. At that time, The handover trajectory of the handover system within the neighborhood can be determined by... To describe.

[0115] For a distribution network system containing mobile energy storage and mobile substations, the system's state variables are monitored. This allows us to determine the system's operational status. Different operational events require optimized scheduling of different subsystems; the determination of event types can be defined as a switching law. Input variables This can be considered as the number and status of mobile energy storage and mobile substations in a distribution network system.

[0116] The optimization steps for coordinated scheduling of mobile energy storage and mobile substations are as follows:

[0117] Define state variables including { , , , , } The net load of node i at time t is calculated as follows:

[0118] (25)

[0119] In the formula: Let be the load power of node i at time t. Let represent the renewable energy generation capacity of node i at time t. This is the working status flag bit for the distribution network line ij. It is 1 when working normally and 0 when there is a fault. The voltage at node i at time t during day-ahead dispatching of the distribution network; This is the maintenance status flag bit for the substation located at node i at time t (0 indicates no maintenance, 1 indicates maintenance). This is a status flag indicating that the load at node i needs a temporary increase in power supply at time t. (0 indicates no demand, 1 indicates demand)

[0120] In a distribution network dispatching system, subsystems include mobile energy storage, mobile substations, and the distribution network itself. The distribution network operation system can be represented as follows:

[0121] (26)

[0122] In the formula: Representing different events The objective function for the operation of the distribution network is as follows.

[0123] The constraints on the operation of the distribution network are as follows:

[0124] (27)

[0125] In the formula; To contribute to DG; , These represent the upper and lower limits of the output power of the distributed power source.

[0126] (28)

[0127] In the formula, Let be the voltage at node i; , These are the upper and lower limits of the node voltage, respectively.

[0128] (29)

[0129] In the formula, , These are the line resistance and reactance, respectively. , These are the active and reactive power of the line, respectively. This represents the line current.

[0130] (30)

[0131] (31)

[0132] In the formula, , These represent the active and reactive power of node j, respectively. and This is the susceptance of the line.

[0133] (32)

[0134] (33)

[0135] (34)

[0136] In the formula: The installed capacity of distributed power sources; This represents the maximum useful power of the distributed power source. Let be the active power of the load at node i; Let be the reactive power of the load at node i; Reduce active power for load; To reduce the active power of distributed power sources; To reduce reactive power from the load; To reduce reactive power in distributed power sources; It is a collection of distributed power sources.

[0137] The events determined by the state variables in the power system have different scheduling requirements for mobile energy storage and mobile substations. This prior knowledge includes two parts: event state determination and demand matching.

[0138] Event status determination, passive events are as follows:

[0139] Load / Distributed Generation Over-limit Events When the net load power of node i It is activated when the adjustable range is exceeded.

[0140] (35)

[0141] In the formula: , These are the upper and lower limits that are adjustable for net load.

[0142] When load / generation exceedances are detected, the subsystem only includes the distribution network and mobile energy storage. The objective functions for optimizing the system operation include minimizing load shedding and minimizing distributed generation curtailment, as follows:

[0143] (36)

[0144] The constraints also include distribution network operation constraints and mobile energy storage operation constraints.

[0145] Line fault event flag Triggered when node i experiences a local or widespread failure.

[0146] (37)

[0147] When a line fault is detected, the subsystem includes the active power portion of the distribution network and mobile energy storage. The optimization objectives of this system include minimizing power loss and minimizing network losses for the remaining lines of the distribution network that are operating normally.

[0148] (38)

[0149] In the formula: Let be the active power loss of node i at time t.

[0150] The constraints also include the active power component of the distribution network and the operational constraints of mobile energy storage;

[0151] Voltage regulation event flag :

[0152] (39)

[0153] The voltage at node i at the day-ahead dispatch time t in the distribution network; , These are the upper and lower limits of the voltage, respectively.

[0154] When a voltage regulation event is detected, the subsystem includes the distribution network, mobile substations, and mobile energy storage. The objective function for optimization is as follows:

[0155] (40)

[0156] Event status determination, active events are as follows:

[0157] Substation maintenance event flag : Triggered when node i is in the planned maintenance state.

[0158] (41)

[0159] In the formula: To request a maintenance signal, the maintenance time starts from time t, and the maintenance signal starts from... Send before To maintain a lower limit for the time of signal advance transmission.

[0160] Temporary service power event flag : Triggered when node i is in a temporary power supply state.

[0161] (42)

[0162] In the formula: To request a temporary power supply signal, the power supply time starts from time t, and the power supply signal starts from... Send before; To maintain a lower limit for the time of signal advance transmission.

[0163] The optimization model described above is a mixed-integer linear programming problem involving binary decision variables and linear constraints. First, all resource parameters (such as cabin capacity, vehicle maneuverability, loading method parameters, etc.), network topology, and fault event data obtained in the embodiments are imported into the mathematical modeling environment (YALMIP). Then, all linear constraint coefficients are normalized and preprocessed, and the binary variables that can be fixed in advance are relaxed and simplified.

[0164] Subsequently, commercial solvers such as Gurobi or CPLEX were selected, and key parameters such as solution time upper limit, optimality tolerance, branching strategy and number of parallel threads were set. The branch and bound algorithm was used to perform a global search on the model. By solving the linear slack subproblem at each node and pruning according to the difference between the upper and lower bounds, the global optimum was gradually approached.

[0165] After the solution is completed, the optimal decision variable values ​​are extracted, the specific scheduling sequence of the cabin and vehicles is generated, and the energy output, transportation route and arrival time of each time period are calculated to draw the scheduling Gantt chart. The key performance indicators such as operating cost, energy revenue and response time are summarized to form a complete report. Finally, robustness analysis is performed on sensitive parameters such as transportation delay, cabin availability and fault response window to evaluate the stability of the scheme to parameter fluctuations. The offline solution process is integrated into the online scheduling platform, and the solver interface is quickly called when new events occur or parameters are updated to iteratively generate the adjusted optimal scheduling scheme.

[0166] Example 2

[0167] In this embodiment, the city's power distribution network consists of nodes 1 to 5. Lines 1-2 are 2 km long, 2-3 are 2 km long, 3-4 are 3 km long, and 4-5 are 4 km long. Node 3 connects to important loads such as hospitals and shopping malls, while node 5 has a distributed photovoltaic system that can generate 1.5 MW of power output. The power distribution dispatch center and resource repository are located at node 0 in the western part of the city, 10 km away from node 3. The speed limit on urban roads is 40 km / h.

[0168] First, all deployable resources are modularly constructed: mobile energy storage is divided into integrated module X and detachable module Y. Module X (numbered X1) is pre-installed in the Node 2 plant area, integrated with a 6×4 m truck chassis, with a rated capacity of 1MWh, a maximum charge / discharge power of 0.5 MW, a volume of 3.0 m×2.5 m×2.5 m, allows 20 start / stop cycles, a response delay of 0.5 min from receiving the order to being able to output power, and a continuous discharge duration of 2 h. Module Y (numbered Y1) is stored in the dedicated base station at Node 0. The detachable module has a standard size of 3.5 m×2.7 m×2.7 m, is matched with a 10 t flatbed truck, has a rated capacity of 2 MWh, a maximum power of 1 MW, allows 25 start / stop cycles, a response delay of 2 min, and a continuous discharge duration of 2 h.

[0169] The mobile substation is divided into two schemes: a single-vehicle loading scheme A (A1) and a two-vehicle loading scheme B (B1 / B2). Scheme A integrates a 2 MVA main transformer, a medium-voltage distribution unit, and high-voltage switchgear on a single tractor vehicle; Scheme B mounts a 3 MVA main transformer on vehicle B1 and a 600 kVA distribution cabinet on vehicle B2. For each type of vehicle, parameters such as rated breaking current, rated voltage / current, maneuverability (minimum turning radius 12 m), vehicle load, start-up delay of 1 min, and continuous operation capability of 48 h are collected. All information is written to a database table in the format of "Resource ID—Vehicle Type—Key Parameters," with fields including "Resource ID," "Type," "Location Node," "Capacity," "Power," "Dimensions," "Compatible Vehicles," "Start-Stop Count," "Response Delay," and "Continuous Operation," ensuring that the system can quickly retrieve information through indexes and SQL queries.

[0170] After completing the database construction, the dispatch center used the MATLAB platform and a digital model of the municipal road network to conduct emergency response capability simulation assessments for two typical scenarios. The active peak-shaving scenario was set up daily from 08:00 to 10:00. When the photovoltaic system at node 5 experienced a significant overload (output > 1.5 MW for 10 minutes), the simulation system started X1 and Y1 from nodes 2 and 0 respectively, performing path planning and shortest time calculations along five pre-loaded optional paths. X1 traveled 2 km in 3 minutes, with loading / unloading and grid connection taking 1 minute; Y1 traveled 10 km in 15 minutes, with loading / unloading and grid connection taking 2 minutes. The passive emergency scenario simulated a sudden drop in photovoltaic output to 0.4 MW and a sudden increase in net load to 1.1 MW at 09:30, and a single-phase fault occurring on lines 3-4 at 10:15. The simulation program records the GPS location, remaining SOC, power output curve, vehicle idle rate, and communication delay with the dispatch center every 0.1 seconds. It ultimately outputs the total response delay from receiving the order to grid connection (X1: 4.5 min, Y1: 17 min), maximum single-time output energy (X1: 0.5 MWh, Y1: 1.0 MWh), and average idle rate (X1: 12%, Y1: 18%). After normalization, all data are weighted and scored according to weights of 0.4 / 0.3 / 0.2 / 0.1 for "response delay," "maximum output energy," "idle rate," and "communication reliability," respectively, yielding emergency response capability indices: X1 0.83, Y1 0.67, A1 0.75, B1 0.72. The results are then written to the "Emergency Index" field of the resource table.

[0171] During actual dispatching, the system monitors node voltage, load power, and transmission line status 24 / 7. When a sudden drop in output at node 3 is detected at 09:30, the strategy engine triggers the "mobile energy storage" subsystem according to a preset threshold (net load exceeding limit by 10%): the decision unit retrieves X1 with the highest emergency index via SQL, immediately executes a dispatch command at 09:30:00, and sends GPS navigation via V2I communication. X1 arrives within the service radius at 09:33:00 and connects to the grid at 09:34:00, entering the regulation layer to continuously discharge 0.5 MW until 10:34, outputting a total of 0.5 MWh. When the line fault flag is triggered at 10:15, the system logic automatically switches to the "distribution network + energy storage" mode, continuing to dispatch X1 and adding Y1. At 10:15:00, Y1 is immediately ordered to depart, arriving at 10:30:00 and connecting at 10:32:00, discharging 1.0 MW until 11:30, and then outputting 1.0 MWh. MWh; At 12:00, Node 2 enters the planned maintenance phase, and the system enters the "mobile substation + energy storage" joint mode. At 12:00:00, instruction scheme A mobile substation A1 is started, arrives at 12:01:00, and is connected to the grid at 12:02:00. After A1 is connected to the grid, it immediately performs step-down and medium-voltage power distribution tasks. X1 then starts charging at a rate of 0.5 MW to compensate for the initial reactive power loss, completing 1.0 MWh of energy charging within two hours. Throughout the process, the scheduling algorithm strictly follows the constraints of unique vehicle dispatch, response delay, access conditions, and upper and lower limits of charge / discharge SOC, ensuring that the status of each resource node is updated synchronously in the database for future query and statistical analysis.

[0172] Through the above implementation process, this embodiment achieves an average response delay of 6.5 min, a cumulative output energy of 1.5 MWh, an average no-load rate of 15% in multiple types of operating events, a 52% reduction in scheduling computation compared to traditional global optimization, and an annual operating cost saving of approximately 15%. At the same time, it reduces the number of user interruptions caused by long-term power outages and voltage over-limits by 40%, significantly improving the reliability and economy of the distribution network.

Claims

1. A collaborative operation optimization method for mobile energy storage and mobile substations based on event-driven scheduling, characterized in that, Includes the following steps: S1. Conduct structural and emergency response capability analysis on mobile energy storage and mobile substations, classify them into different types, collect key performance parameters and enter them into the module library in a specified format, then conduct a quantitative assessment of emergency response capability based on the modular resource model, calculate the emergency response capability index and integrate it into the resource model; Specifically, when analyzing the structure of mobile energy storage, it is divided into integrated and separable types. The collected key performance parameters include the rated capacity, maximum charge / discharge power, physical dimensions, mobile platform compatibility information, maximum number of start-stop cycles, start-up response time, and continuous discharge duration for each battery compartment. The entry format is "compartment number-vehicle type-performance parameters"; for mobile substations... When conducting structural analysis, the equipment is classified into four types: single-vehicle loading, two-vehicle loading, three-vehicle loading, and compartmentalized loading. The information collected includes: main transformer capacity, switch equipment parameters, medium-voltage distribution unit specifications, load and mobility performance of each vehicle, start-up delay, and continuous operation capability. The data entry format is "loading method-vehicle number-equipment parameters". When conducting quantitative assessment of emergency response capabilities, simulation schemes for two typical scenarios, active peak shaving and passive emergency response, are developed. The equipment arrival path, loading and unloading process, and grid connection operation process are simulated in the distribution network node distribution and road transportation network. The transportation time, start-up delay, energy output curve, and utilization rate are recorded. The emergency response capability index of each type of module is calculated through statistical analysis. S2. The operation mode of mobile substations and mobile energy storage is modeled in three layers: dispatch layer, access layer, and regulation layer. Corresponding constraints are set at each layer to construct the operation model. Specifically, in the modeling of the mobile substation dispatch layer, binary decision variables are introduced to characterize the instructions issued by the vehicle from the base to the fault node. A vehicle dispatch uniqueness constraint is set to ensure that one vehicle from one substation can only be dispatched to one node where an event has occurred at any given time. Simultaneously, an immediate response constraint is set to ensure that at least one mobile substation is dispatched from the origin after an event occurs. In the modeling of the mobile substation access layer, a cumulative dispatch constraint is set to ensure that the vehicle arrives at the destination node after being dispatched. An arrival time constraint is set to specify the vehicle arrival time. An integration constraint for all vehicles is set to ensure that access is only allowed after all necessary vehicles have arrived and equipment setup time has been considered. An access status maintenance constraint is also set. Ensure that the connected state is maintained unless special operations are performed after connection; when modeling the mobile substation regulation layer, set access adjustable constraints, only allowing the on-load tap changer to start voltage regulation after the mobile substation is connected, set on-load tap changer regulation constraints, clarify the selection and regulation rules of the on-load tap changer, and limit the maximum single adjustment amount of the tap; set vehicle dispatch uniqueness constraints, immediate response constraints, and turnaround transportation constraints in the mobile energy storage dispatch layer to ensure reasonable vehicle dispatch and fuel consumption does not exceed limits; the mobile energy storage access layer adopts cumulative dispatch constraints and arrival time constraints similar to those of the mobile substation; the mobile energy storage regulation layer sets traffic constraints to ensure reasonable charging and discharging actions and the number of connected units, sets charging and discharging and battery energy storage health constraints to limit charging and discharging parameters and battery status, and sets module integration charging and discharging status uniqueness constraints to ensure consistent charging and discharging action commands when multiple modules are connected to the same node; S3. Based on switching system theory, the operating state variables of the distribution network are mapped to several subsystem modes. A preset switching law automatically switches to the corresponding mode when different events occur, and the corresponding optimization model is invoked to achieve customized collaborative scheduling. The defined state variables include the net load of a node at time t, the working status flag of the distribution network line, the node voltage at time t during day-ahead scheduling of the distribution network, the maintenance status flag of the substation at the node, and the status flag of temporary power supply increase for the load at the node. The node net load is calculated from the node load power and the renewable energy generation power. The preset switching law corresponds to different events, including load / distributed generation over-limit events, line fault events, voltage regulation events, substation maintenance events, and temporary service power supply events. Different subsystems and optimization models are invoked for different events. The optimization model is a mixed-integer linear programming problem. By importing resource parameters, network topology, and fault event data, a commercial solver is used to solve the problem, generating a scheduling sequence and performing robustness analysis.

2. The method for optimizing the coordinated operation of mobile energy storage and mobile substations based on event-driven scheduling according to claim 1, characterized in that, In step S3, after the commercial solver completes the solution, the optimal decision variable values ​​are extracted, a specific scheduling sequence for the cabin and vehicles is generated, and the energy output, transportation path, and arrival time indicators for each time period are calculated to draw a scheduling Gantt chart. The key performance indicators of operating costs, energy benefits, and response time are summarized to form a complete report. Finally, robustness analysis is performed on sensitive parameters such as transportation delay, cabin availability, and fault response window to evaluate the stability of the scheme to parameter fluctuations. The commercial solver solution process is integrated into the online scheduling platform, and the commercial solver interface is quickly invoked when new events occur or parameters are updated to iteratively generate the adjusted optimal scheduling scheme.

Images