Urban and rural charging and storage combined scheduling method and system based on mobile energy storage

Abstract

The invention discloses an urban and rural charging and storage combined scheduling method and system based on mobile energy storage, and belongs to the crossing field of energy internet and power markets, and the method comprises the steps: obtaining and preprocessing urban and rural power grid, traffic and market multi-source measurement data, and constructing a power-traffic double-flow weighted fusion graph model; establishing a day-ahead and intra-day double-layer prediction model of village and town new energy power generation and urban and rural node electricity price; a two-stage optimization scheduling model including day-ahead planning and intra-day scheduling is designed, a cooperative scheduling mechanism of fixed energy storage and mobile energy storage of villages and towns and an abnormal scene coping strategy are introduced, and a dynamic scheduling scheme is generated through alternate iteration solution by taking the maximum net benefit of operation of a mobile energy storage vehicle as a scheduling target. According to the method, a rural storage and urban use commercial operation mode is adopted, the economic benefits of mobile energy storage vehicle operators are guaranteed, the new energy consumption rate of villages and towns is improved, urban power grid congestion is relieved, and key technical support is provided for large-scale marketization application of mobile energy storage.

Description

Technical Field

[0001] This invention belongs to the intersection of the energy internet and the electricity market, specifically involving a method and system for joint urban and rural charging and storage dispatch based on mobile energy storage. Background Technology

[0002] In recent years, new energy sources such as rural distributed photovoltaic and decentralized wind power have developed rapidly. However, due to the intermittent nature of power generation and insufficient local absorption capacity, the phenomenon of "wind and solar curtailment" is prominent, and the value of low electricity prices during peak periods of new energy power generation has not been effectively explored. At the same time, the electricity load in urban centers is showing rigid growth, with peak-hour electricity prices soaring and the risk of transmission and distribution line congestion increasing, resulting in a significant supply-demand imbalance. Mobile energy storage vehicles, serving as "mobile carriers" connecting urban and rural energy resources, possess flexible charging, discharging, and spatial transfer capabilities. Theoretically, they can achieve arbitrage through "low-cost energy storage in rural areas and high-cost discharge in cities." However, their actual commercial operation faces three core bottlenecks: First, timing mismatch: peak periods for renewable energy generation in rural areas (such as midday solar power) are difficult to match with the travel plans of mobile energy storage vehicles (such as low transportation efficiency during morning and evening rush hours). Second, lack of coordination: fixed energy storage in rural areas serves only as "passive power supply facilities," failing to coordinate with the scheduling rhythm of mobile energy storage vehicles and thus unable to pre-store renewable energy to meet the charging needs of mobile energy storage vehicles during off-peak hours. Third, poor robustness: lacking mechanisms to cope with abnormal scenarios such as traffic disruptions, sudden drops in renewable energy output, and electricity price fluctuations, resulting in a single profit model. Traditional mobile energy storage scheduling solutions often focus on single technical objectives (such as maximizing renewable energy consumption) and fail to construct an optimization framework centered on sustainable commercial benefits, making it difficult to support large-scale application. Summary of the Invention

[0003] To address the problems of timing mismatch, lack of coordination, single profit model, and poor robustness between mobile energy storage and urban and rural energy supply and demand in existing technologies, this invention proposes a method and system for joint urban and rural charging and storage scheduling based on mobile energy storage.

[0004] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0005] The urban-rural joint dispatching method based on mobile energy storage includes the following steps:

[0006] Step 1: Acquire urban planning data, historical multi-source measurement data, and real-time multi-source measurement data for the target area. Urban planning data includes power system network topology, transportation network backbone topology, location and capacity of fixed energy storage nodes in townships, and location and load characteristics of urban load centers. Multi-source measurement data includes township renewable energy generation data, urban and rural node marginal price (LMP) data, urban load data, traffic flow data, and fixed energy storage status data. Preprocess the acquired historical and real-time multi-source measurement data.

[0007] Step 2: Create a weighted fusion graph model of power and transportation flows based on the acquired urban planning data;

[0008] Step 3: Based on the preprocessed historical multi-source measurement data, use the pre-trained day-ahead prediction model to output the prediction curves for township new energy power generation, urban and rural node electricity prices, and traffic flow for the next day; based on the preprocessed real-time multi-source measurement data, use the pre-trained intraday prediction model to output the prediction curves for township new energy power generation, urban and rural node electricity prices, and traffic flow for the future preset time period within the scheduling day.

[0009] Step 4: Construct a two-stage optimization scheduling model that includes a day-ahead planning model and an intraday scheduling model, with the goal of maximizing the net operating revenue of mobile energy storage vehicles. The day-ahead planning model takes the prediction results of the day-ahead prediction model as input and solves for the optimal driving route in the power-traffic dual-flow weighted fusion graph to obtain the day-ahead optimized driving route scheme for each mobile energy storage vehicle. The intraday scheduling model takes the day-ahead optimized driving route scheme output by the day-ahead planning model and the prediction results of the intraday prediction model as input and iteratively solves for the goal of maximizing the net operating revenue of mobile energy storage vehicles to obtain the optimized charging and discharging power of mobile energy storage vehicles.

[0010] Step 5: Update the scheduling period and real-time multi-source measurement data with a preset time resolution, iteratively solve the two-stage optimized scheduling model until the end of the scheduling day, output a complete optimized scheduling scheme for mobile energy storage vehicles, and upload it to the scheduling terminal to guide the mobile energy storage vehicles to perform charging, discharging and driving operations.

[0011] Furthermore, in step 1, the power system network topology in the urban planning data includes line impedance and rated capacity, while the main road topology of the transportation network includes road length, speed limit, and traffic capacity. Historical multi-source measurement data includes rural renewable energy generation data, urban and rural node marginal electricity price data, urban load data, traffic flow data, and fixed energy storage status data for the past year. Real-time multi-source measurement data includes rural renewable energy generation data, urban and rural node marginal electricity price data, urban load data, traffic flow data, and fixed energy storage status data for the hours preceding the current time period on the dispatch day. Among these, rural renewable energy generation data includes photovoltaic and wind power generation data, traffic flow data includes average travel time for road segments, and fixed energy storage status data includes the state of charge (SOC) and charging / discharging power of fixed energy storage nodes. Preprocessing of historical multi-source measurement data and implemented multi-source measurement data includes outlier removal, missing value interpolation, and data standardization.

[0012] Furthermore, a weighted fusion graph model of electricity and transportation flows is created based on urban planning data. Its mathematical model is defined as follows:

[0013] G = {V, E, W}

[0014] Wherein, V is the set of nodes, including Ve: power network nodes, Vt: traffic road nodes, Vs: township fixed energy storage nodes, and V1: urban load nodes; E is the set of edges, including Ee: power flow edges representing power line connection relationships, and Et: traffic flow edges representing road connection relationships; W is the set of weights, including We: power flow edge weights representing line impedance, and Wt: traffic flow edge weights representing road segment travel time.

[0015] The creation process is as follows: First, the coordinates of each node are marked in the Geographic Information System (GIS) to complete the mapping of physical locations; then, edge sets are generated according to the connection relationships, that is, power network nodes are connected to township fixed energy storage nodes and urban load nodes respectively to generate power flow edges, and traffic road nodes are connected to township fixed energy storage nodes and urban load nodes respectively to generate traffic flow edges; finally, the traffic flow edge weight and power flow edge weight of each edge are calculated to generate a power-traffic dual-flow weighted fusion map.

[0016] Furthermore, in step 2, the pre-trained day-ahead prediction model uses a Long Short-Term Memory (LSTM) network. It takes time-series data consisting of historical township renewable energy power generation data, urban and rural node marginal electricity price data, traffic flow data, meteorological data, and date type data as input, and outputs the renewable energy power generation data prediction curve, urban and rural node electricity price prediction curve, and traffic flow prediction curve with a preset time resolution for the next 24 hours. The meteorological data includes light intensity and wind speed, and the date type includes weekdays and holidays. The pre-trained intraday prediction model uses a combination of Rolling Time Domain Optimization (RTO) and XGBoost algorithms. It takes real-time sequence data consisting of township renewable energy power generation data, urban and rural node marginal electricity price data, and traffic flow data consisting of several hours before the current time period of the scheduling day as input, and outputs the renewable energy power generation data prediction curve, urban and rural node electricity price prediction curve, and traffic flow prediction curve with a preset time resolution for the period after the current time period.

[0017] Furthermore, step 4 includes the following specific steps:

[0018] Step 4.1: Construct a scheduling objective function that includes transportation costs and line impedance loss costs;

[0019] Step 4.2: Construct the core constraints for solving the scheduling objective function;

[0020] Step 4.3: Based on the scheduling objective function, construct a two-stage optimized scheduling model that includes a day-ahead planning model and an intraday scheduling model.

[0021] Furthermore, the specific expression of the scheduling objective function constructed in step 4.1 is as follows:

[0022]

[0023] In the formula, f represents the total net daily revenue of mobile energy storage vehicles; T represents the total scheduling time period; K represents the set of mobile energy storage vehicles; and N represents the set of nodes. Let be the discharge electricity price of urban load node i during time period t. Let k be the discharge charge of mobile energy storage vehicle k at urban load node i during time period t. Let t be the charging electricity price for fixed energy storage node j in the township, and λ be the charging service fee for the fixed energy storage unit in the township. The charging power of mobile energy storage vehicle k at the fixed energy storage node j in the township during time period t; C transit D represents the unit distance transportation cost. k Let C be the travel distance of the mobile energy storage vehicle k. deg C represents the cost of battery degradation per unit charge. loss,k,i,t The cost is the cost of line impedance loss; the specific calculation formulas for the battery degradation cost per unit charge and the cost of line impedance loss are as follows:

[0024] C deg = Initial battery cost / (Cycle life × Rated capacity)

[0025] Where U is the voltage of the urban power distribution network, R i Let Δt be the line impedance corresponding to city node i, and Δt be the scheduling time interval.

[0026] Furthermore, step 4.2, which constructs the core constraints for solving the scheduling objective function, includes:

[0027] 1) Constraints on orderly charging in townships:

[0028]

[0029] In the formula, SOC j,t The state of charge of fixed energy storage node j in the township during time period t. For the rated capacity of fixed energy storage node j in the township, T gen This is during peak periods for new energy power generation in rural towns. The charging power of fixed energy storage node j in the township during time period t1. The charging power of mobile energy storage vehicle k at the fixed energy storage node j in the township during time period t2;

[0030] 2) Urban load matching constraints:

[0031] t∈T peak,i

[0032]

[0033] In the formula T peak,i For the peak load period of urban load node i, Let i be the average electricity price per day for urban load node i.

[0034] 3) State of Charge (SOC) constraints for mobile energy storage vehicles:

[0035]

[0036] In the formula, For the rated capacity of mobile energy storage vehicle k, SOC k,t For the state of charge (SOC) of mobile energy storage vehicle k during time period t, min The minimum permissible state of charge, SOC max The maximum permissible state of charge;

[0037] 4) Power loss constraints:

[0038]

[0039] In the formula, P loss,max This represents the maximum allowable power loss of the line.

[0040] Furthermore, step 5 iteratively solves the two-stage optimization scheduling model to output a complete optimized scheduling scheme for mobile energy storage vehicles. The specific process includes:

[0041] Step 5.1: Input the preprocessed historical multi-source measurement data sequence of the days prior to the scheduling day into the day-ahead prediction model, and output the new energy power generation data prediction curve, urban and rural node electricity price prediction curve, and traffic flow prediction curve for the next 24 hours.

[0042] Step 5.2: The day-ahead planning model takes the prediction results output by the day-ahead prediction model as input, and aims to minimize transportation costs. It uses the Dijkstra algorithm to solve the optimal driving route of the mobile energy storage vehicle in the power-traffic dual-flow weighted fusion graph, and obtains the day-ahead optimized driving route of each mobile energy storage vehicle.

[0043] Step 5.3: Input the preprocessed real-time multi-source measurement data sequence from several hours prior to the current time period into the intraday prediction model, and output the new energy power generation data prediction curve, urban and rural node electricity price prediction curve, and traffic flow prediction curve for the preset time period after the current time period; input the day-ahead optimized driving route data and the prediction results output by the intraday prediction model into the intraday scheduling model, and use the interior point method to iteratively solve the problem under the conditions of satisfying the orderly charging constraints of fixed energy storage in towns, the time period matching constraints of urban load, and the SOC safety constraints of mobile energy storage vehicles, until the iteration termination condition is met, and obtain the optimized charging and discharging power of mobile energy storage vehicles;

[0044] Step 5.4: Update the current time period and real-time multi-source measurement data according to the preset time resolution. Repeat step 5.3 until the end of the scheduling day. Integrate the optimized driving routes and the optimized charging and discharging power values ​​of all time periods of the scheduling day, output the complete mobile energy storage vehicle optimization scheduling plan for the scheduling day, and upload it to the scheduling terminal at the same time to guide the mobile energy storage vehicles to perform charging, discharging and driving operations.

[0045] This invention also protects a joint urban and rural charging and storage scheduling system based on mobile energy storage, including a data acquisition module for acquiring urban planning data and historical multi-source measurement data in step 1; a first model building module for constructing a power-traffic dual-flow weighted fusion map based on the imported urban planning data, and preprocessing the acquired historical multi-source measurement data and real-time multi-source measurement data; a second model building module for constructing and training a day-ahead prediction model and an intraday prediction model based on machine learning; a third model building module for constructing a scheduling objective function and a two-stage optimized scheduling model including a day-ahead planning model and an intraday scheduling model, and iteratively solving the model through an optimization solver to output the day-ahead optimized driving route and intraday charging and discharging scheme for mobile energy storage vehicles; and a scheduling terminal module for the mobile energy storage vehicle's on-board terminal and operation management platform to implement the scheduling of mobile energy storage vehicles to perform charging, discharging, and driving operations.

[0046] Furthermore, the data acquisition module consists of an edge computing gateway, a data acquisition terminal, and an energy integrated service station; the edge computing gateway is deployed at fixed energy storage nodes in townships and urban load centers to collect township new energy power generation data, fixed energy storage status data, and urban load data; the data acquisition terminal collects traffic flow data and meteorological data.

[0047] Furthermore, the operation and management platform has the following functions:

[0048] Data visualization: Real-time display of data including urban and rural node electricity prices, new energy power generation, location and status of charge of mobile energy storage vehicles, and line impedance, generating daily / weekly / monthly revenue reports;

[0049] Anomaly Alerts: When scenarios such as traffic disruptions or a sudden drop in renewable energy power generation are detected, alarm information including the anomaly type, scope of impact, and response suggestions will be pushed through the APP.

[0050] Manual intervention: Supports operators to manually adjust charging and discharging power and driving routes, and export dispatch plans.

[0051] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0052] This invention proposes a method and system for joint urban and rural charging and storage scheduling based on mobile energy storage. Through integrated power-transportation resource modeling, it accurately matches the travel routes of mobile energy storage vehicles with the spatiotemporal distribution of urban and rural energy supply and demand, overcoming the bottleneck of timing mismatch. By employing a collaborative scheduling mechanism between fixed energy storage in townships and mobile energy storage vehicles, it achieves "pre-storage-peak-shifting" of new energy power, improving the efficiency of green energy consumption. A two-stage optimized scheduling model that balances revenue stability and robustness addresses multi-source data fluctuations and abnormal scenarios, ensuring continuous profitability for operators. Through "revenue-driven" market-based scheduling, it simultaneously solves the problems of new energy consumption in townships and urban power grid congestion while realizing the commercial value of mobile energy storage vehicles, achieving a win-win situation for both economic and social benefits. Attached Figure Description

[0053] Figure 1 A flowchart of the method provided by the present invention;

[0054] Figure 2 The output curves of the day-ahead prediction model and the intraday prediction model provided by this invention;

[0055] Figure 3 A comparison diagram showing the implementation effects of the method provided by this invention and traditional methods. Detailed Implementation

[0056] To make the technical solution of the present invention clearer and more complete, the technical solution of the present invention will be described in further detail below with reference to the accompanying drawings and specific embodiments.

[0057] Taking a county in eastern my country as the application scenario, and combining the county's energy and transportation resources, the basic parameter configuration logic is determined, including:

[0058] Urban and rural node layout: Based on the urban planning data of the target county, 5 fixed energy storage nodes in townships and 3 urban load centers are selected to form a spatial closed loop of township energy storage and urban discharge;

[0059] Mobile energy storage vehicle configuration: Based on the new energy consumption demand and transportation network capacity of the target county, 10 mobile energy storage vehicles will be configured. The vehicle parameters will meet the requirements of "matching the charging power of townships with the fixed energy storage output and the discharge power of cities with the load demand", while also taking into account transportation costs and battery degradation characteristics.

[0060] Correlation between power grid and traffic parameters: Based on the power system topology of the target county, determine the impedance characteristics of the power lines corresponding to the load centers of each city; combined with the topology of the main traffic arteries, clarify the travel time patterns between fixed energy storage nodes in townships and load nodes in cities.

[0061] like Figure 1 As shown, based on the basic parameter configuration logic of the target county, the urban-rural joint scheduling method for mobile energy storage provided by this invention includes the following steps:

[0062] Step 1: Obtain urban planning data, historical multi-source measurement data, and real-time multi-source measurement data for the target county through the target county's integrated energy service station. Urban planning data includes power system network topology, transportation network backbone topology, location and capacity of fixed energy storage nodes in townships, and location and load characteristics of urban load centers. Multi-source measurement data includes township renewable energy generation data, urban and rural node marginal electricity price data, urban load data, traffic flow data, and fixed energy storage status data. Preprocess the acquired historical and real-time multi-source measurement data, including outlier removal, missing value interpolation, and data standardization.

[0063] Step 2: Create a weighted fusion graph model of power and transportation flows based on the acquired urban planning data;

[0064] Step 3: Based on the preprocessed historical multi-source measurement data, a pre-trained day-ahead prediction model based on a long short-term memory network is used. The input consists of time-series data comprising historical township renewable energy generation data, urban and rural node marginal electricity price data, traffic flow data, meteorological data, and date-type data. The output consists of prediction curves for township renewable energy generation data, urban and rural node electricity price prediction curves, and traffic flow prediction curves for the next 24 hours with a time resolution of 1 hour. Meteorological data includes light intensity and wind speed, and date types include weekdays and holidays. The pre-trained intraday prediction model uses a combination of rolling time-domain optimization and the XGBoost algorithm. The input consists of real-time sequence data comprising township renewable energy generation data, urban and rural node marginal electricity price data, and traffic flow data from several hours prior to the current time period of the scheduling day. The output consists of prediction curves for renewable energy generation data, urban and rural node electricity price prediction curves, and traffic flow prediction curves for the next 4 hours with a time resolution of 15 minutes. Figure 2 A comparison chart of the forecast curves output by the day-ahead forecast model and the intraday forecast model is presented;

[0065] Step 4: Construct a two-stage optimization scheduling model that includes a day-ahead planning model and an intraday scheduling model, with the goal of maximizing the net operating revenue of mobile energy storage vehicles; specifically including:

[0066] Step 4.1: Construct a scheduling objective function that includes transportation costs and line impedance loss costs. The specific expression is as follows:

[0067]

[0068] In the formula, f represents the total net daily revenue of mobile energy storage vehicles; T represents the total scheduling time period; K represents the set of mobile energy storage vehicles; and N represents the set of nodes. Let be the discharge electricity price of urban load node i during time period t. Let k be the discharge charge of mobile energy storage vehicle k at urban load node i during time period t. Let t be the charging electricity price for fixed energy storage node j in the township, and λ be the charging service fee for the fixed energy storage unit in the township. The charging power of mobile energy storage vehicle k at the fixed energy storage node j in the township during time period t; C transit D represents the unit distance transportation cost. k Let C be the travel distance of the mobile energy storage vehicle k. deg C represents the cost of battery degradation per unit charge. loss,k,i,t The cost is the cost of line impedance loss; the specific calculation formulas for the battery degradation cost per unit charge and the cost of line impedance loss are as follows:

[0069] C deg = Initial battery cost / (Cycle life × Rated capacity)

[0070] Where U is the voltage of the urban power distribution network, R i Let Δt be the line impedance corresponding to city node i, and Δt be the scheduling time interval.

[0071] Step 4.2: Construct the core constraints for solving the scheduling objective function, including:

[0072] 1) Constraints on orderly charging in townships:

[0073]

[0074] In the formula, SOC j,t The state of charge of fixed energy storage node j in the township during time period t. For the rated capacity of fixed energy storage node j in the township, T gen This is during peak periods for new energy power generation in rural towns. The charging power of fixed energy storage node j in the township during time period t1. The charging power of mobile energy storage vehicle k at the fixed energy storage node j in the township during time period t2;

[0075] 2) Urban load matching constraints:

[0076] t∈T peak,i

[0077]

[0078] In the formula T peak,i For the peak load period of urban load node i, Let i be the average electricity price per day for urban load node i.

[0079] 3) State of Charge (SOC) constraints for mobile energy storage vehicles:

[0080]

[0081] In the formula, For the rated capacity of mobile energy storage vehicle k, SOC k,t For the state of charge (SOC) of mobile energy storage vehicle k during time period t, min The minimum permissible state of charge, SOC max The maximum permissible state of charge;

[0082] 4) Power loss constraints:

[0083]

[0084] In the formula, P loss,max This represents the maximum allowable power loss of the line.

[0085] Step 4.3: Based on the scheduling objective function, construct a two-stage model that includes a day-ahead planning model and an intraday scheduling model;

[0086] Step 5: Update the current time and real-time multi-source measurement data with a preset time resolution, iteratively solve the two-stage optimized scheduling model until the end of the scheduling day, output a complete optimized scheduling scheme for mobile energy storage vehicles, and simultaneously upload it to the scheduling terminal to guide the mobile energy storage vehicles in performing charging, discharging, and driving operations; specifically:

[0087] Step 5.1: Initialize the scheduling day as d and the current time period t=1. Input the preprocessed historical multi-source measurement data sequence of the previous few days into the day-ahead prediction model, and output the new energy power generation data prediction curve, urban and rural node electricity price prediction curve and traffic flow prediction curve with a time resolution of 1 hour for the next 24 hours.

[0088] Step 5.2: The day-ahead planning model takes the prediction results output by the day-ahead prediction model as input, and aims to minimize transportation costs. It uses the Dijkstra algorithm to solve the optimal driving route of the mobile energy storage vehicle in the power-traffic dual-flow weighted fusion graph, and obtains the day-ahead optimized driving route of each mobile energy storage vehicle.

[0089] Step 5.3: Input the preprocessed real-time multi-source measurement data sequence m hours before the current time period t into the intraday prediction model, and output the new energy power generation data prediction curve, urban and rural node electricity price prediction curve, and traffic flow prediction curve with a time resolution of 15 minutes for the next 4 hours; input the day-ahead optimized driving route data and the prediction results output by the intraday prediction model into the intraday scheduling model, and use the interior point method to iteratively solve the problem with the goal of maximizing the net operating revenue of mobile energy storage vehicles, while satisfying the orderly charging constraints of fixed energy storage in towns, the time-matching constraints of urban load, and the SOC safety constraints of mobile energy storage vehicles. Set the initial iterative revenue f. init The corrected profit f is obtained by iterative solution. new; Calculate the difference in returns Δf = |f new -f init | / f init If the iteration termination condition is not met, then update f. init =f new Continue iteratively solving until the iteration termination condition is met to obtain the optimized charging and discharging power of the mobile energy storage vehicle;

[0090] Step 5.4: Update the current time period t = t + 1, update the real-time multi-source measurement data, and repeat step 5.3 until the end of the scheduling day, i.e., t = 24. Integrate the day-ahead optimized driving routes and the day-ahead optimized charging and discharging power values ​​for all time periods of the scheduling day, output the complete mobile energy storage vehicle optimized scheduling plan for the scheduling day, and push it to the mobile energy storage vehicle's on-board terminal and operation management platform to guide the vehicle in performing charging, discharging, and driving operations. During operation, monitor the scheduling effect in real time, and simultaneously record key indicators such as new energy consumption, operating revenue, and grid congestion changes to verify the effectiveness of the strategy. Figure 3 A comparison chart showing the implementation effects of the method provided by this invention and the traditional method is presented.

[0091] This invention also protects a joint urban and rural charging and storage scheduling system based on mobile energy storage, including a data acquisition module for acquiring urban planning data and historical multi-source measurement data in step 1; a first model building module for constructing a power-traffic dual-flow weighted fusion map based on the imported urban planning data, and preprocessing the acquired historical multi-source measurement data and real-time multi-source measurement data; a second model building module for constructing a day-ahead prediction model based on a long short-term memory network and an intraday prediction model based on rolling time-domain optimization combined with the XGBoost algorithm, and training the model; a third model building module for constructing a scheduling objective function and a two-stage optimized scheduling model including a day-ahead planning model and an intraday scheduling model, solving the day-ahead planning model using the Dijkstra algorithm, iteratively solving the intraday scheduling model using the interior point method, and outputting the day-ahead optimized driving route and intraday charging and discharging scheme for mobile energy storage vehicles; and a scheduling terminal module, which is the on-board terminal and operation management platform for mobile energy storage vehicles, for implementing the scheduling of mobile energy storage vehicles to perform charging, discharging, and driving operations.

[0092] Specifically, the data acquisition module consists of an edge computing gateway, a data acquisition terminal, and an energy integrated service station; the edge computing gateway is deployed at fixed energy storage nodes in townships and urban load centers to collect township new energy power generation data, fixed energy storage status data, and urban load data; the data acquisition terminal collects traffic flow data and meteorological data.

[0093] Specifically, the functions of the operation and management platform include:

[0094] Data visualization: Real-time display of data including urban and rural node electricity prices, new energy power generation, location and status of charge of mobile energy storage vehicles, and line impedance, generating daily / weekly / monthly revenue reports;

[0095] Anomaly Alerts: When scenarios such as traffic disruptions or a sudden drop in renewable energy power generation are detected, alarm information including the anomaly type, scope of impact, and response suggestions will be pushed through the APP.

[0096] Manual intervention: Supports operators to manually adjust charging and discharging power and driving routes, and export dispatch plans.

[0097] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Any modifications, equivalent substitutions, or improvements made by those skilled in the art within the scope of the technology disclosed in this invention, based on the technical solution and concept of the present invention, should be included within the protection scope of this invention. Therefore, the protection scope of this invention should be determined by the scope of the claims.

Claims

1. A method for joint urban and rural charging and storage scheduling based on mobile energy storage, characterized in that, Includes the following steps: Step 1: Acquire urban planning data, historical multi-source measurement data, and real-time multi-source measurement data for the target area. Urban planning data includes power system network topology, transportation network backbone topology, location and capacity of fixed energy storage nodes in townships, and location and load characteristics of urban load centers. Multi-source measurement data includes township renewable energy generation data, urban and rural node marginal electricity price data, urban load data, traffic flow data, and fixed energy storage status data. Preprocess the acquired historical and real-time multi-source measurement data. Step 2: Create a weighted fusion graph model of power and transportation flows based on the acquired urban planning data; Step 3: Based on the preprocessed historical multi-source measurement data, use the pre-trained day-ahead prediction model to output the prediction curves for township new energy power generation, urban and rural node electricity prices, and traffic flow for the next day; based on the preprocessed real-time multi-source measurement data, use the pre-trained intraday prediction model to output the prediction curves for township new energy power generation, urban and rural node electricity prices, and traffic flow for the future preset time period within the scheduling day. Step 4: Construct a two-stage optimization scheduling model that includes a day-ahead planning model and an intraday scheduling model, with the goal of maximizing the net operating revenue of mobile energy storage vehicles. The day-ahead planning model takes the prediction results of the day-ahead prediction model as input and solves for the optimal driving route in the power-traffic dual-flow weighted fusion graph to obtain the day-ahead optimized driving route scheme for each mobile energy storage vehicle. The intraday scheduling model takes the day-ahead optimized driving route scheme output by the day-ahead planning model and the prediction results of the intraday prediction model as input and iteratively solves for the goal of maximizing the net operating revenue of mobile energy storage vehicles to obtain the optimized charging and discharging power of mobile energy storage vehicles. Step 5: Update the scheduling period and real-time multi-source measurement data with a preset time resolution, iteratively solve the two-stage optimized scheduling model until the end of the scheduling day, output a complete optimized scheduling scheme for mobile energy storage vehicles, and upload it to the scheduling terminal to guide the mobile energy storage vehicles to perform charging, discharging and driving operations.

2. The urban and rural charging and storage joint dispatching method based on mobile energy storage according to claim 1, characterized in that, In step 1, the power system network topology in the urban planning data includes line impedance and rated capacity, while the main road topology of the transportation network includes road length, speed limit, and traffic capacity. Historical multi-source measurement data includes rural renewable energy generation data, urban and rural node marginal electricity price data, urban load data, traffic flow data, and fixed energy storage status data for the past year. Real-time multi-source measurement data includes rural renewable energy generation data, urban and rural node marginal electricity price data, urban load data, traffic flow data, and fixed energy storage status data for the hours preceding the current time period on the dispatch day. Among these, rural renewable energy generation data includes photovoltaic and wind power generation data, traffic flow data includes average travel time for road segments, and fixed energy storage status data includes the state of charge and charging / discharging power of fixed energy storage nodes. Preprocessing of historical and implemented multi-source measurement data includes outlier removal, missing value interpolation, and data standardization.

3. The urban and rural charging and storage joint dispatching method based on mobile energy storage according to claim 1, characterized in that, A weighted fusion graph model of electricity and transportation flow is created based on urban planning data. Its mathematical model is defined as follows: G = {V, E, W} Wherein, V is the set of nodes, including Ve: power network nodes, Vt: traffic road nodes, Vs: township fixed energy storage nodes, and V1: urban load nodes; E is the set of edges, including Ee: power flow edges representing power line connection relationships, and Et: traffic flow edges representing road connection relationships; W is the set of weights, including We: power flow edge weights representing line impedance, and Wt: traffic flow edge weights representing road segment travel time. The creation process is as follows: First, the coordinates of each node are marked in the geographic information system to complete the mapping of physical locations; then, edge sets are generated according to the connection relationships, that is, the power network nodes are connected to the township fixed energy storage nodes and the urban load nodes respectively to generate power flow edges, and the traffic road nodes are connected to the township fixed energy storage nodes and the urban load nodes respectively to generate traffic flow edges; finally, the traffic flow edge weight and power flow edge weight of each edge are calculated to generate a power-traffic dual-flow weighted fusion map.

4. The urban and rural charging and storage joint dispatching method based on mobile energy storage according to claim 1, characterized in that, In step 2, the pre-trained day-ahead prediction model uses a long short-term memory network. It takes time-series data consisting of historical township renewable energy power generation data, urban and rural node marginal electricity price data, traffic flow data, meteorological data, and date type data as input, and outputs the township renewable energy power generation data prediction curve, urban and rural node electricity price prediction curve, and traffic flow prediction curve with a preset time resolution for the next 24 hours. The meteorological data includes light intensity and wind speed, and the date type includes weekdays and holidays. The pre-trained intraday prediction model uses a combination of rolling time-domain optimization and the XGBoost algorithm. It takes real-time series data consisting of township renewable energy power generation data, urban and rural node marginal electricity price data, and traffic flow data from several hours before the current time period of the scheduling day as input, and outputs the renewable energy power generation data prediction curve, urban and rural node electricity price prediction curve, and traffic flow prediction curve with a preset time resolution for the period after the current time period.

5. The urban and rural charging and storage joint dispatching method based on mobile energy storage according to claim 1, characterized in that, Step 4 includes the following specific steps: Step 4.1: Construct a scheduling objective function that includes transportation costs and line impedance loss costs; Step 4.2: Construct the core constraints for solving the scheduling objective function; Step 4.3: Based on the scheduling objective function, construct a two-stage optimized scheduling model that includes a day-ahead planning model and an intraday scheduling model.

6. The urban and rural charging and storage joint dispatching method based on mobile energy storage according to claim 5, characterized in that, The specific expression of the scheduling objective function constructed in step 4.1 is as follows: In the formula, f represents the total net daily revenue of mobile energy storage vehicles; T represents the total scheduling time period; K represents the set of mobile energy storage vehicles; and N represents the set of nodes. Let be the discharge electricity price of urban load node i during time period t. Let k be the discharge charge of mobile energy storage vehicle k at urban load node i during time period t. Let t be the charging electricity price for fixed energy storage node j in the township, and λ be the charging service fee for the fixed energy storage unit in the township. The charging power of mobile energy storage vehicle k at the fixed energy storage node j in the township during time period t; C transit D represents the unit distance transportation cost. k Let C be the travel distance of the mobile energy storage vehicle k. deg C represents the cost of battery degradation per unit charge. loss,k,i,t The cost is the cost of line impedance loss; the specific calculation formulas for the battery degradation cost per unit charge and the cost of line impedance loss are as follows: C deg = Initial battery cost / (Cycle life × Rated capacity) Where U is the voltage of the urban power distribution network, R i Let Δt be the line impedance corresponding to city node i, and Δt be the scheduling time interval.

7. The urban and rural charging and storage joint dispatching method based on mobile energy storage according to claim 6, characterized in that, Step 4.2 Construct the core constraints for solving the scheduling objective function, including: 1) Constraints on orderly charging in townships: In the formula, SOC j,t The state of charge of fixed energy storage node j in the township during time period t. For the rated capacity of fixed energy storage node j in the township, T gen This is during peak periods for new energy power generation in rural towns. The charging power of fixed energy storage node j in the township during time period t1. The charging power of mobile energy storage vehicle k at the fixed energy storage node j in the township during time period t2; 2) Urban load matching constraints: t∈T peak,i In the formula T peak,i For the peak load period of urban load node i, Let i be the average electricity price per day for urban load node i. 3) State of Charge (SOC) constraints for mobile energy storage vehicles: In the formula, For the rated capacity of mobile energy storage vehicle k, SOC k,t For the state of charge (SOC) of mobile energy storage vehicle k during time period t, min The minimum permissible state of charge, SOC max The maximum permissible state of charge; 4) Power loss constraints: In the formula, P loss,max This represents the maximum allowable power loss of the line.

8. The urban and rural charging and storage joint dispatching method based on mobile energy storage according to claim 7, characterized in that, Step 5 iteratively solves the two-stage optimization scheduling model and outputs a complete mobile energy storage vehicle optimization scheduling scheme. The specific process includes: Step 5.1: Input the preprocessed historical multi-source measurement data sequence of the days prior to the scheduling day into the day-ahead prediction model, and output the new energy power generation data prediction curve, urban and rural node electricity price prediction curve, and traffic flow prediction curve for the next 24 hours. Step 5.2: The day-ahead planning model takes the prediction results output by the day-ahead prediction model as input, and aims to minimize transportation costs. It uses the Dijkstra algorithm to solve the optimal driving route of the mobile energy storage vehicle in the power-traffic dual-flow weighted fusion graph, and obtains the day-ahead optimized driving route of each mobile energy storage vehicle. Step 5.3: Input the preprocessed real-time multi-source measurement data sequence from several hours prior to the current time period into the intraday prediction model, and output the new energy power generation data prediction curve, urban and rural node electricity price prediction curve, and traffic flow prediction curve for the preset time period after the current time period; input the day-ahead optimized driving route data and the prediction results output by the intraday prediction model into the intraday scheduling model, and use the interior point method to iteratively solve the problem under the conditions of satisfying the orderly charging constraints of fixed energy storage in towns, the time period matching constraints of urban load, and the SOC safety constraints of mobile energy storage vehicles, until the iteration termination condition is met, and obtain the optimized charging and discharging power of mobile energy storage vehicles; Step 5.4: Update the current time period and real-time multi-source measurement data according to the preset time resolution. Repeat step 5.3 until the end of the scheduling day. Integrate the optimized driving routes and the optimized charging and discharging power values ​​of all time periods of the scheduling day, output the complete mobile energy storage vehicle optimization scheduling plan for the scheduling day, and upload it to the scheduling terminal at the same time to guide the mobile energy storage vehicles to perform charging, discharging and driving operations.

9. A joint urban and rural charging and storage dispatching system based on mobile energy storage, characterized in that, The method for joint urban and rural charging and storage scheduling based on mobile energy storage as described in any one of claims 1-8 includes: a data acquisition module for acquiring urban planning data and historical multi-source measurement data from step 1; a first model building module for constructing a power-traffic dual-flow weighted fusion map based on the imported urban planning data and preprocessing the acquired historical and real-time multi-source measurement data; a second model building module for constructing and training a day-ahead prediction model and an intraday prediction model based on machine learning; a third model building module for constructing a scheduling objective function and a two-stage optimized scheduling model including a day-ahead planning model and an intraday scheduling model, iteratively solving the model through an optimization solver, and outputting the day-ahead optimized driving route and intraday charging and discharging scheme for mobile energy storage vehicles; and a scheduling terminal module, which is the on-board terminal and operation management platform for mobile energy storage vehicles, for scheduling mobile energy storage vehicles to perform charging, discharging, and driving operations.

10. The urban and rural charging and storage joint dispatching system based on mobile energy storage according to claim 9, characterized in that, The data acquisition module consists of an edge computing gateway, a data acquisition terminal, and an integrated energy service station. The edge computing gateway is deployed at fixed energy storage nodes in townships and urban load centers to collect township new energy power generation data, fixed energy storage status data, and urban load data. The data acquisition terminal collects traffic flow data and meteorological data; The functions of the operation and management platform include: Data visualization: Real-time display of data including urban and rural node electricity prices, new energy power generation, location and status of charge of mobile energy storage vehicles, and line impedance, generating daily / weekly / monthly revenue reports; Anomaly Alerts: When scenarios such as traffic disruptions or a sudden drop in renewable energy power generation are detected, alarm information including the anomaly type, scope of impact, and response suggestions will be pushed through the APP. Manual intervention: Supports operators to manually adjust charging and discharging power and driving routes, and export dispatch plans.

Images