Regional intelligent electricity utilization method and system based on source-grid-load-storage cooperation and dynamic electricity price

Abstract

The invention provides a regional intelligent power utilization method and system based on source-grid-load-storage cooperation and dynamic electricity price. The method comprises the following steps: collecting photovoltaic output, energy storage SOC, load demand and time-of-use electricity price data in real time; determining a corresponding dynamic control mode in combination with the economic target model, the environmental protection target model and the constraint conditions; when the running dynamic control mode is a light storage and charging cooperative arbitrage mode, dynamically determining an energy storage charging and discharging strategy; when the running dynamic control mode is a virtual power plant aggregation mode, forming scheduling capacity by aggregating distributed resources, and responding to a power grid peak regulation instruction; when the running dynamic control mode is a carbon-electricity joint optimization mode, a carbon trading platform is connected, green certificate binding sales is carried out, scheduling capacity is formed through dynamic integration of energy storage charging and discharging strategies and aggregation of distributed resources, the carbon trading platform is connected, green certificate binding sales is carried out, economic benefits are effectively improved, carbon emission is reduced, and the service life of the system is prolonged. And the double-target optimization of the electric cost and the carbon emission is realized.

Description

Technical Field

[0001] The present invention relates to the field of smart grid and energy management technology, and in particular to a regional smart electricity consumption method based on source-grid-load-storage coordination and dynamic electricity pricing. Background Art

[0002] Source, grid, load and storage is the core operating model of the new power system, which consists of four key parts. Source (power source): includes renewable energy (such as solar energy, wind energy, and hydropower) and traditional energy (such as coal power and natural gas). Grid (grid): transmission network and power system equipment, responsible for energy transmission and distribution. Load: terminal electricity demand, covering electricity users such as industry, residents, and businesses. Storage (energy storage): stores excess electricity through technologies such as batteries and pumped storage, and releases it when there is an imbalance between supply and demand. Its integrated model achieves a dynamic balance between supply and demand through intelligent scheduling, such as storing energy when there is excess power generation from renewable energy, and releasing stored power when power generation is insufficient.

[0003] While a certain current platform achieves the coordinated optimization of solar-powered storage and charging, it doesn't cover virtual power plants participating in peak-shaving ancillary services (for example, one pilot project generated 236,000 yuan in monthly revenue) or carbon trading. Current energy storage scheduling patents focus on the economic benefits of charging station operators but fail to consider global optimization under the varying electricity prices of various electricity users (e.g., industrial, commercial, and residential).

[0004] Among them, the main issues are as follows: Insufficient coordination between sources, grids, loads and storage: The traditional power grid has a low absorption rate of new energy (for example, the self-generation and self-consumption rate of photovoltaic power generation is less than 70%), and the energy storage scheduling strategy is single, without dynamic optimization based on the difference between the competitive grid-connected electricity price on the power generation side and the time-of-use electricity price on the power consumption side.

[0005] Poor adaptability of market-based mechanisms: The existing system cannot adapt to the 15-minute clearing model of the electricity spot market (for example, the price difference fluctuated by more than 1.5 yuan / kWh within a certain pilot day), and the requirements for energy storage response speed (required to be at the minute level) and prediction accuracy (error ≤ 5%) are insufficient.

[0006] Lack of carbon asset accounting: The existing scheme does not integrate carbon trading and green electricity revenue models, and cannot achieve the dual superposition of carbon quota and electricity price difference benefits.

[0007] However, the current traditional power grid has a low absorption rate of new energy and a single energy storage scheduling strategy, resulting in relatively poor economic and environmental performance. Summary of the Invention

[0008] The present invention provides a regional intelligent electricity consumption method and system based on source-grid-load-storage coordination and dynamic electricity price, which is used to solve the defects of the existing technology such as low new energy consumption rate and single energy storage scheduling strategy, resulting in relatively poor economy and environmental protection.

[0009] In a first aspect, the present invention provides a regional smart electricity consumption method based on source-grid-load-storage coordination and dynamic electricity pricing, comprising: Real-time collection of photovoltaic output, energy storage SOC, load demand and time-of-use electricity price data; Determine the dynamic control mode corresponding to the photovoltaic output, energy storage SOC, load demand and time-of-use electricity price data by combining the economic target model, environmental target model and constraint conditions; When the dynamic control mode is a photovoltaic storage and charging arbitrage mode, dynamically determining the energy storage charging and discharging strategy; When the dynamic control mode in operation is the virtual power plant aggregation mode, the dispatch capacity is formed by aggregating distributed resources to respond to the peak load regulation instructions of the power grid; When the dynamic control mode in operation is the carbon-electricity joint optimization mode, it is connected to the carbon trading platform to carry out green certificate bundled sales.

[0010] According to a regional smart electricity consumption method based on source-grid-load-storage coordination and dynamic electricity pricing provided by the present invention, when the dynamic control mode is a photovoltaic-storage-charging coordinated arbitrage mode, the energy storage charging and discharging strategy is dynamically determined, including: Determine the data input system corresponding to multi-source data fusion, and use a hybrid algorithm framework based on the data input system to perform dynamic electricity price forecasting; Determine the state space, action space, and reward function of the predicted dynamic electricity price using a photovoltaic storage and charging collaborative arbitrage model; Based on the state space, action space and reward function, combined with a rolling optimization mechanism, abnormal event response and online learning, the energy storage charging period and the discharging period are dynamically determined.

[0011] According to a regional smart electricity consumption method based on source-grid-load-storage collaboration and dynamic electricity pricing provided by the present invention, the state space, action space, and reward function of the predicted dynamic electricity price are determined by using a photovoltaic-storage-charging collaborative arbitrage model, including: Based on the photovoltaic storage and charging collaborative arbitrage model, a Markov decision process is constructed using deep reinforcement learning; The Markov decision process is used to determine the real-time electricity price, the electricity price forecast curve for a preset period, the real-time photovoltaic power and forecast curve, the energy storage SOC / SOH, the current total charging demand and the load distribution of each charging pile, and the maximum charging and discharging power allowed by the power grid as the state space; Determine the energy storage charging power, energy storage discharging power, and charging pile power allocation using the Markov decision process as an action space; The Markov decision process is used to determine arbitrage benefits, photovoltaic absorption rewards, battery life penalties, user satisfaction and grid constraint penalties as reward functions.

[0012] According to the present invention, a regional smart electricity consumption method based on source-grid-load-storage coordination and dynamic electricity pricing is provided. The method dynamically determines the energy storage charging and discharging periods based on the state space, action space, and reward function, combined with a rolling optimization mechanism, abnormal event response, and online learning, including: Determining the operating status and external environment of the power grid system through the state space; In combination with the operating conditions and external environment, selecting an energy storage charging action, an energy storage discharging action, or an charging pile power adjustment action from the action space; After completing any of the energy storage charging actions, energy storage discharging actions, or charging pile power adjustment actions, a reward function is determined based on the revenue; By rolling optimization and updating the charging and discharging plan, and combining the abnormal event response with online learning of the charging and discharging strategy, the energy storage charging and discharging periods are determined.

[0013] According to a regional smart electricity consumption method based on source-grid-load-storage coordination and dynamic electricity pricing provided by the present invention, when the dynamic control mode is the virtual power plant aggregation mode, distributed resources are aggregated to form a dispatching capacity to respond to the power grid peak load regulation instruction, including: Build a distributed resource access platform to collect real-time data on power generation, remaining energy storage capacity, and load power consumption; Combined with the power generation of the power source, the remaining power of the energy storage, and the real-time power consumption of the load, the power generation trend of the distributed power source, the change pattern of the load, and the charging and discharging capacity of the energy storage are predicted to build a resource capacity map of the virtual power plant; Taking the peak-shaving demand of the power grid as the target and combining the resource capability map, a resource scheduling plan is formulated to respond to the peak-shaving instruction of the power grid.

[0014] According to a regional smart electricity consumption method based on source-grid-load-storage coordination and dynamic electricity pricing provided by the present invention, when the dynamic control mode is the carbon-electricity joint optimization mode, the method connects to the carbon trading platform to carry out green certificate bundled sales, including: By building a multi-data perception network through the perception layer, integrating it with the carbon price trend and green certificate supply and demand data of the carbon trading platform, a dynamic panoramic view of carbon-electricity data is formed; Based on the panoramic view of carbon-electricity data, the decision-making layer is used to build an intelligent decision-making center to dynamically generate green certificate bundling sales strategies; Based on the green certificate bundling sales strategy, the carbon-electricity trading alliance chain and smart contract trading model are used to realize the green certificate bundling sales.

[0015] According to the present invention, a regional smart electricity consumption method based on source-grid-load-storage coordination and dynamic electricity pricing is provided. The method uses a perception layer to build a multi-data perception network, which is integrated with the carbon price trend and green certificate supply and demand data of the carbon trading platform to form a dynamic carbon-electricity data panorama, including: On the power generation side, real-time data collection is used to measure the power generation, power generation period, and equipment operating status of photovoltaic and wind power. On the power consumption side, data collection is used to obtain the power consumption curves and energy consumption structures of enterprises and residents. Use environmental monitoring equipment combined with meteorological conditions to estimate power generation data; The photovoltaic and wind power generation, power generation period, equipment operating status, electricity consumption curves and energy consumption structure of the enterprises and residents are integrated with the carbon price trend and green certificate supply and demand data of the carbon trading platform to form a dynamic panoramic view of carbon-electricity data.

[0016] According to the present invention, a regional smart electricity consumption method based on source-grid-load-storage coordination and dynamic electricity pricing also includes: Establish a carbon-electricity joint optimization effect evaluation system; By using the carbon-electricity joint optimization effect evaluation system, the green certificate bundling sales strategy is quantitatively evaluated from different dimensions to achieve an adaptive optimization closed loop.

[0017] In a second aspect, the present invention further provides a regional smart electricity consumption system based on source-grid-load-storage coordination and dynamic electricity pricing, comprising: The acquisition module is used to collect real-time data on photovoltaic output, energy storage SOC, load demand, and time-of-use electricity prices; A determination module is used to determine the dynamic control mode corresponding to the photovoltaic output, energy storage SOC, load demand and time-of-use electricity price data by combining the economic target model, the environmental target model and the constraint conditions; A photovoltaic storage and charging collaborative arbitrage module is used to dynamically determine the energy storage charging and discharging strategy when the dynamic control mode is the photovoltaic storage and charging collaborative arbitrage mode; A virtual power plant aggregation module, configured to form a dispatch capacity by aggregating distributed resources and respond to power grid peak load regulation instructions when the dynamic control mode in operation is the virtual power plant aggregation mode; The carbon-electricity joint optimization module is used to connect to the carbon trading platform and sell green certificates in bundled form when the dynamic control mode in operation is the carbon-electricity joint optimization mode.

[0018] In a third aspect, the present invention also provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, a regional intelligent electricity consumption method based on source-grid-load-storage coordination and dynamic electricity prices as described above is implemented.

[0019] In a fourth aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a regional intelligent electricity consumption method based on source-grid-load-storage coordination and dynamic electricity prices as described above.

[0020] In a fifth aspect, the present invention also provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above-mentioned regional smart electricity consumption methods based on source-grid-load-storage coordination and dynamic electricity prices.

[0021] The present invention provides a regional intelligent electricity consumption method and system based on source-grid-load-storage coordination and dynamic electricity price, including: real-time collection of photovoltaic output, energy storage SOC, load demand and time-of-use electricity price data; combining economic target model, environmental target model and constraint conditions to determine the dynamic control mode corresponding to photovoltaic output, energy storage SOC, load demand and time-of-use electricity price data; when the dynamic control mode of operation is the photovoltaic-storage-charging coordinated arbitrage mode, dynamically determine the energy storage charging and discharging strategy; when the dynamic control mode of operation is the virtual power plant aggregation mode, form a dispatching capacity by aggregating distributed resources to respond to the peak-shaving instructions of the power grid; when the dynamic control mode of operation is the carbon-electricity joint optimization mode, connect to the carbon trading platform to carry out green certificate bundled sales, through dynamic integration of energy storage charging and discharging strategies, aggregate distributed resources to form dispatching capacity, connect to the carbon trading platform, and carry out green certificate bundled sales, effectively improve economic benefits and reduce carbon emissions, and achieve the dual goals of electricity cost and carbon emissions. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0023] Figure 1 This is a flow chart of a regional smart electricity consumption method based on source-grid-load-storage coordination and dynamic electricity pricing provided by this embodiment; Figure 2 This is a schematic diagram of the structure of a regional smart power consumption system based on source-grid-load-storage coordination and dynamic electricity pricing provided by this embodiment; Figure 3 Schematic diagram of the structure of the electronic device provided in this embodiment. DETAILED DESCRIPTION

[0024] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0025] Figure 1 This is a flow chart of a regional smart electricity consumption method based on source-grid-load-storage coordination and dynamic electricity price provided by this embodiment.

[0026] like Figure 1 As shown, an embodiment of the present invention provides a regional smart electricity consumption method based on source-grid-load-storage coordination and dynamic electricity pricing, which mainly includes the following steps: 101. Real-time collection of photovoltaic output, energy storage SOC, load demand and time-of-use electricity price data.

[0027] PV output data collection involves intelligent sensors installed on every photovoltaic panel and inverter in a PV power plant. These sensors act like "small scouts." They monitor in real time the light intensity absorbed by the panels, the voltage and current generated, and the output power of the inverter after converting DC power to AC power. These sensors then quickly transmit this data to the system's data collection center via a wired network (such as an Ethernet cable) or a wireless network (like WiFi). This allows users to monitor the current power generation of the PV power plant at any time.

[0028] Energy storage SOC data collection specifically involves the following: Energy storage systems (such as large battery packs) have a dedicated battery management system (BMS), which acts as a "dedicated butler" for the batteries. The BMS measures parameters such as battery voltage, current, and temperature, and uses a specific algorithm to calculate the battery's current state of charge (SOC), or remaining charge level, such as "60% remaining." The BMS regularly transmits this data via communication lines to a data collection center, allowing the system to monitor the energy storage device's charge status in real time.

[0029] Load demand data collection involves upgrading every meter in electricity-consuming areas, such as residential communities and factories, to a smart meter. These smart meters not only record electricity usage but also monitor current power consumption in real time—that is, the amount of power consumed by a user's device at a given moment. Smart meters transmit real-time power consumption data to a data collection center via power line carrier communication or wireless communication. This allows users to clearly understand current power consumption and load demand.

[0030] To collect TOU electricity price data, the official website, mobile app, or dedicated power communication network regularly publishes electricity price information for different time periods, such as peak and off-peak hours. The data collection center proactively "subscribes" to this information, much like subscribing to a newspaper, obtaining the latest TOU price data at a fixed time. Programs can also be set up to automatically receive and save new price information released by the power company to the system, allowing the system to schedule operations based on electricity prices at any time.

[0031] Through the above methods, the system can quickly and accurately collect real-time data on photovoltaic output, energy storage SOC, load demand and time-of-use electricity prices, providing an important basis for the subsequent reasonable arrangement of power generation, energy storage and electricity consumption.

[0032] 102. Combine the economic target model, environmental target model and constraints to determine the dynamic control mode corresponding to photovoltaic output, energy storage SOC, load demand and time-of-use electricity price data.

[0033] Specifically, the economic objective model is used to minimize the total electricity cost, including utility power purchases, energy storage losses, and green certificate revenue. This cost includes the cost of purchasing electricity from the grid, the energy losses caused by using energy storage equipment, and the profit from selling green certificates. By calculating the costs of different electricity usage plans, for example, if using energy storage equipment to power a home when electricity prices are low is more cost-effective than buying electricity directly from the grid, if there is excess photovoltaic power generated, how much can be saved by storing the excess power in energy storage equipment and using it when electricity prices are high, and how much income can be increased by selling excess green certificates, the most cost-effective electricity usage plan can be found by continuously comparing various options.

[0034] The environmental target model is designed to minimize carbon emissions, including through a carbon footprint tracking model that supports ISO 14064 reporting. This model, like labeling each kilowatt-hour of electricity with an "environmental label," clearly identifies the carbon emissions generated. This model also supports ISO 14064 reporting standards, acting as a formal environmental report card. For example, more photovoltaic power can be used, as it is a clean energy source with virtually no carbon emissions. Reduce electricity purchases from the grid, especially during peak hours, when electricity is likely to come from higher-carbon generation methods. Ultimately, this approach helps identify the most environmentally friendly electricity usage plan.

[0035] Constraints include power balance between source, grid, load, and storage, energy storage capacity limits (SOC 20%-90%), and grid safety margins (voltage fluctuation ≤ ±5%), ensuring safe and reasonable power consumption throughout the entire process. This power balance ensures that the power generated by photovoltaic power plants, purchased from the grid, and energy storage devices precisely match the household's electricity needs. Energy storage capacity limits (SOC 20%-90%) ensure that the energy storage device's power level is constantly monitored and charging and discharging are properly scheduled. A grid safety margin (voltage fluctuation ≤ ±5%) ensures that grid voltage fluctuations remain within a safe range during power consumption.

[0036] Finally, the most cost-effective solution calculated by the economic objective model, the most environmentally friendly solution calculated by the environmental objective model, and the required constraints are comprehensively considered. Like putting together a puzzle, the economic, environmental, and safety pieces must be perfectly assembled to find the optimal electricity plan that minimizes electricity costs and carbon emissions while meeting various safety and capacity constraints. This solution becomes the final dynamic control model.

[0037] 103. When the dynamic control mode is the photovoltaic storage and charging arbitrage mode, the energy storage charging and discharging strategy is dynamically determined.

[0038] Specifically, in terms of system architecture, the solar-storage-charging collaborative arbitrage system adopts a four-layer architecture to achieve full-process intelligent decision-making, as follows: Data perception layer: IoT sensors are used to collect real-time data on photovoltaic power station power generation, energy storage system state of charge (SOC) and health (SOH), and charging pile real-time charging demand (including vehicle type and user charging time preference). It also accesses the grid's real-time electricity prices, peak and valley time divisions, load forecast data, and meteorological information (light intensity, temperature, humidity, etc.).

[0039] Edge computing layer: Preprocesses real-time data, extracts key features (such as photovoltaic power fluctuation trends, electricity price fluctuation slopes, and charging demand peak periods), filters noise, and compresses data to provide efficient input for cloud-based decision-making.

[0040] Cloud-based decision-making layer: Builds a strategy optimization model based on algorithms such as reinforcement learning and time series prediction to dynamically generate energy storage charging and discharging strategies and charging pile power allocation plans, while taking into account grid constraints (such as power caps and safe operating ranges).

[0041] Execution feedback layer: Sends control instructions to the energy storage converter and charging piles, monitors the execution effect in real time, and corrects strategy deviations through a closed-loop feedback mechanism (such as adjusting the allocation priority when the actual charging load does not match the forecast).

[0042] In the control process, multi-source data fusion and dynamic prediction are first performed, including: determining the data input system corresponding to multi-source data fusion. The data input system is divided into different dimensions, including the power grid dimension: obtaining real-time electricity price curves (accurate to 15 minutes), next-day peak and valley time division, and power grid dispatch instructions (such as peak-shaving demand and backup capacity requirements). The new energy dimension: combining short-term photovoltaic power forecasts (15 minutes to 1 hour) and medium- and long-term weather forecasts (24-48 hours) to predict the fluctuation range of photovoltaic output. The energy storage dimension: real-time monitoring of battery state of charge (SOC) (to avoid overcharge and over-discharge), state of health (SOH) (to assess the remaining battery life and dynamically adjust the depth of charge and discharge), and charge and discharge efficiency curves (energy loss at different power levels). The user dimension: analyzing historical charging data to predict user charging time preferences (such as concentrated charging on weekday evenings and high-frequency demand during weekend lunchtime) and identify the ratio of fast charging to slow charging demand.

[0043] Based on the data input system, a hybrid algorithm framework is then used to perform dynamic electricity price forecasting. This includes capturing time series features: using a long short-term memory (LSTM) network to learn the cyclical patterns of historical electricity prices (such as weekday peak-valley differences and holiday price fluctuations). External factor integration: using the XGBoost gradient boosting tree (XGBoost) to integrate meteorological data (high and low temperatures may affect grid load) and market events (such as electricity price policy adjustments and sudden changes in energy supply and demand) to correct time series forecast biases. Attention mechanism weighting: using an attention model to dynamically assign prediction weights to LSTM and XGBoost (for example, prioritizing XGBoost output during extreme weather events), ultimately generating a 24-hour electricity price forecast curve.

[0044] The PV-storage-charging synergy arbitrage model is then used to determine the state space, action space, and reward function for the predicted dynamic electricity price. This involves using deep reinforcement learning to construct a Markov decision process based on the PV-storage-charging synergy arbitrage model. This process uses the Markov decision process to determine the real-time electricity price, the electricity price forecast curve for a preset time period, the real-time PV power and forecast curve, the energy storage SOC / SOH, the current total charging demand and the load distribution of each charging pile, and the maximum charge and discharge power allowed by the grid as the state space. Furthermore, the Markov decision process is used to determine the energy storage charging power (0 to rated charging power, subject to the upper SOC limit constraint), the energy storage discharging power (0 to rated discharging power, subject to the lower SOC limit constraint), and the charging pile power distribution (dynamically adjusting the output of each charging pile based on user priority, such as prioritizing emergency charging vehicles) as the action space.

[0045] A Markov decision process is used to determine arbitrage returns, PV consumption incentives, battery life penalties, user satisfaction, and grid constraint penalties as the reward function. Arbitrage returns are calculated based on the real-time difference between charging and discharging electricity prices (e.g., direct profits from charging during off-peak hours and discharging during peak hours). PV consumption incentives provide positive incentives for energy storage systems to consume excess PV power (reducing curtailment). Battery life penalties calculate battery loss costs based on charge and discharge depth and rate (e.g., deep discharge accelerates SOH degradation). User satisfaction is quantified based on charging wait time and power allocation rationality (e.g., prioritizing fast-charging vehicles). Grid constraint penalties negatively penalize actions that exceed the grid's permitted power range (e.g., overload discharge, which poses a risk to grid security).

[0046] Finally, based on the state space, action space, and reward function, combined with a rolling optimization mechanism, abnormal event response, and online learning, the energy storage charging and discharging periods are dynamically determined. This includes: determining the operating status and external environment of the power grid system through the state space (real-time collection of information such as current electricity prices, photovoltaic power generation, energy storage power, and charging demand); selecting energy storage charging actions, energy storage discharging actions, or charging pile power adjustment actions from the action space based on the operating status and external environment; and determining the reward function based on the revenue after completing any energy storage charging action, energy storage discharging action, or charging pile power adjustment action (scoring based on aspects such as money earned, photovoltaic utilization, and battery loss).

[0047] To cope with ever-changing circumstances, the system also implements rolling optimization updates, replanning the next charging and discharging schedule every 15 minutes, much like a constantly updated travel route. In the event of an unexpected situation, such as a sudden drop in photovoltaic power or a surge in charging demand, it immediately initiates an abnormal event response, prioritizing system stability. Furthermore, like a student learning new knowledge, the system adjusts its decision-making strategy hourly based on the latest operational data (through online learning), gradually acquiring more economic and environmental skills, ultimately accurately determining energy storage charging and discharging periods, achieving both economic benefits and system stability.

[0048] 104. When the dynamic control mode is the virtual power plant aggregation mode, the dispatching capacity is formed by aggregating distributed resources to respond to the peak load regulation instructions of the power grid.

[0049] Specifically, a distributed resource access platform will be built. Leveraging IoT technology, this platform connects distributed power sources (such as small photovoltaic power plants and wind turbines), energy storage devices (home storage, commercial and industrial storage systems), and controllable loads (adjustable air conditioners and electric vehicle charging stations with delayed charging), enabling real-time data collection and transmission. The collected data includes key information such as power generation, remaining energy storage capacity, and real-time load power usage, providing data support for subsequent scheduling.

[0050] By combining the power generation capacity of power sources, the remaining capacity of energy storage, and the real-time power consumption of loads, advanced data analysis and forecasting technologies are used to conduct in-depth assessment and forecasting of aggregated resources. Using machine learning algorithms, combined with historical data, meteorological information, and user electricity usage habits, the system predicts the generation trends of distributed power sources, load fluctuations, and the charging and discharging capacity of energy storage, thereby constructing a resource capacity map for the virtual power plant. For example, the system can predict the next day's power generation curve for photovoltaic power stations based on weather forecasts, and predict peak power consumption periods for commercial building air conditioners by analyzing user schedules. This resource capacity map clearly displays the available resource capacity and potential at different times.

[0051] When the power grid issues a peak-shaving order, the intelligent dispatch system quickly activates. Based on the grid's peak-shaving needs, combined with the resource capacity map, it uses optimization algorithms (such as mixed integer programming) to develop a resource dispatch plan to respond to the grid's peak-shaving order. When formulating the plan, it fully considers factors such as resource type, location, and regulation characteristics to achieve precise resource matching and coordination. For example, energy storage devices closer to the grid's load center are prioritized for discharge to quickly respond to peak-shaving demands. For adjustable loads, smart meters send flexible power usage recommendations to users, using economic incentives (such as electricity price discounts) to encourage users to adjust their power consumption time to achieve peak-shaving and valley-filling.

[0052] To ensure efficient execution of the dispatch plan, a real-time communication and feedback mechanism has been established. Using high-speed communication networks like 5G, dispatch instructions are quickly and accurately transmitted to each distributed resource end. Upon receiving the instructions, the resource end devices immediately execute the corresponding operations. Simultaneously, the operating status and execution results of the resources are monitored in real time, and this data is fed back to the virtual power plant control center. If an emergency occurs during execution (such as a failure of an energy storage device), the control center quickly activates the emergency plan, dynamically adjusts the dispatch plan, and mobilizes other available resources to supplement the peak-shaving task, ensuring the smooth completion of the peak-shaving task.

[0053] Furthermore, the introduction of blockchain technology enhances trust and collaborative efficiency within virtual power plants. Leveraging the decentralized and tamper-proof nature of blockchain, it records information such as the contribution of distributed resources and the distribution of revenue, ensuring transparent and fair transactions. Resource entities can use the blockchain platform to view their participation in peak-shaving and their revenue in real time, enhancing their enthusiasm for participation. Furthermore, blockchain technology enables cross-regional and cross-entity resource coordination, breaking down information barriers and improving the overall operational efficiency of virtual power plants.

[0054] Finally, a continuous optimization and learning system is established. The virtual power plant continuously collects data from peak-shaving tasks and analyzes the rationality of its scheduling strategies and resource utilization efficiency. Leveraging reinforcement learning algorithms, the scheduling model and strategy are adjusted based on feedback from execution results, enabling it to more accurately and efficiently aggregate resources and respond to grid demand in subsequent peak-shaving tasks. For example, through multiple peak-shaving practices, the system learns resource scheduling patterns under different weather conditions and electricity usage scenarios, gradually improving the virtual power plant's intelligence and peak-shaving capabilities.

[0055] 105. When the dynamic control mode is the carbon-electricity joint optimization mode, it will be connected to the carbon trading platform to carry out green certificate bundled sales.

[0056] Specifically, a multi-data sensing network will be built at the perception layer, integrating it with carbon price trends and green certificate supply and demand data from the carbon trading platform to create a dynamic, panoramic view of carbon-electricity data. This includes leveraging the Internet of Things and digital twin technologies to build a monitoring system covering the entire energy production and consumption process. On the power generation side, smart sensors will be deployed to collect real-time data on photovoltaic and wind power generation, generation time periods, and equipment operating status. On the electricity consumption side, smart meters will be used to capture electricity consumption curves and energy consumption structures for businesses and residents. Furthermore, environmental monitoring equipment will be used to estimate power generation data based on meteorological conditions (such as sunlight and wind speed). This data will be integrated with carbon price trends and green certificate supply and demand data from the carbon trading platform. Privacy-preserving computing technologies will be used to break down data silos, creating a dynamic, panoramic view of carbon-electricity data to support subsequent decision-making.

[0057] Based on a panoramic view of carbon-electricity data, an intelligent decision-making center is constructed using the decision-making layer. Guided by economic benefits and carbon emission reduction targets, a green certificate bundling sales strategy is dynamically generated. By inputting multi-dimensional information such as real-time electricity prices, carbon market price fluctuations, green certificate inventory, and corporate carbon quota gaps, the output includes the optimal green certificate bundling ratio, sales timing, and target customer groups. For example, when carbon prices are on an upward trend and corporate carbon quotas are tight, an electricity sales plan with a high green certificate bundling ratio is automatically generated. Differentiated green certificate bundling packages are customized to meet the carbon emission reduction needs of customers in different industries (such as high-energy-consuming enterprises such as steel and internet companies). Simulation technology is also used to simulate the benefits and emission reduction effects of different strategies, pre-diagnosing risks and optimizing decision-making plans.

[0058] Based on the green certificate bundling strategy, the carbon-electricity trading alliance chain and smart contract trading model are used to achieve green certificate bundling sales. During the green certificate bundling sales process, "NFT green certificate" technology is used to give each green certificate a unique digital identity, recording its corresponding clean energy power generation, power generation time, environmental benefits and other information, thereby enhancing the traceability and credibility of green certificates. Through smart contracts, the needs of buyers and sellers are automatically matched. When the buyer submits a green certificate bundled with electricity purchase order, the smart contract automatically locks the corresponding green certificate and electricity resources according to the strategy output by the decision-making layer, calculates the transaction price according to the real-time carbon price and electricity price, and completes the one-click bundling transaction. After the transaction is completed, the alliance chain updates the green certificate holdings, carbon quota, capital flow and other information in real time, and generates a visual transaction voucher to facilitate supervision and auditing.

[0059] Finally, a carbon-electricity joint optimization effect evaluation system can be established; it can quantitatively evaluate the green certificate bundled sales strategy from the perspectives of economic benefits (such as carbon trading revenue and electricity sales profits), environmental benefits (carbon emission reduction, clean energy absorption rate), and customer satisfaction (order fulfillment rate, user repurchase rate). Utilizing federated learning technology, it can collaborate with other industry participants to share anonymized transaction data and optimization experience, and continuously iterate the parameters of the reinforcement learning model. For example, if a certain type of green certificate bundled product is found to be unsold, the system will automatically analyze the reasons and adjust the bundling ratio and marketing strategy. Based on changes in carbon market policies (such as new carbon quota allocation rules), the carbon-electricity joint decision-making model can be promptly optimized, allowing the entire system to continuously evolve in a dynamic environment and achieve continuous growth in the value of carbon-electricity resources.

[0060] In this embodiment, a two-way dynamic optimization mechanism integrates the differences between the on-grid electricity price (on the generator side) and the time-of-use electricity price (on the consumer side) to achieve synergistic arbitrage between photovoltaics, storage, and charging. Through intelligent aggregation of virtual power plants, blockchain evidence storage technology achieves sub-second regulation accuracy, supporting participation in spot market transactions and demand response. By integrating a carbon metering module through carbon-electricity dual-objective decision-making, carbon quota allocation and green electricity trading ratios are dynamically optimized to achieve optimal economic and environmental performance. Through multi-objective optimization models and dynamic control strategies, efficient synergy between source, grid, load, and storage resources is achieved, resulting in significant market competitiveness.

[0061] The method of the present invention was used to conduct a test in a certain industrial park, as follows: Scenario parameters: 2MW photovoltaic power, 1MWh energy storage, peak-valley electricity price 1.2 / 0.3 yuan / kWh, carbon price 50 yuan / ton.

[0062] Optimization process: 1. Directly supply loads and store energy during peak photovoltaic power generation periods (10:00-14:00), and the remaining electricity is auctioned to the grid at a price of 0.25 yuan / kWh; 2. Release stored energy during peak periods and participate in peak-shaving auxiliary services to obtain compensation of 0.8 yuan / kWh; 3. Charge during off-peak periods and purchase low-priced green electricity (0.28 yuan / kWh) to reduce carbon intensity.

[0063] Effect: The average monthly electricity cost was reduced by 28%, and the carbon quota income increased by 120,000 yuan.

[0064] The test was conducted in a rural microgrid as follows: It is mandatory to configure 50% of installed power with energy storage to realize the "photovoltaic + agriculture + energy storage" model, and village-level energy cooperatives distribute surplus electricity income.

[0065] Effect: The photovoltaic self-use rate increased to 95%, saving 1.5 million yuan in electricity bills annually.

[0066] Based on the same general inventive concept, the present invention also protects a regional smart power consumption system based on source-grid-load-storage coordination and dynamic electricity prices. The regional smart power consumption system based on source-grid-load-storage coordination and dynamic electricity prices described below and the regional smart power consumption method based on source-grid-load-storage coordination and dynamic electricity prices described above can be referenced to each other.

[0067] Figure 2 This is a structural diagram of the regional smart power consumption system based on source-grid-load-storage coordination and dynamic electricity price provided by this embodiment.

[0068] like Figure 2 As shown, this embodiment provides a regional smart power consumption system based on source-grid-load-storage coordination and dynamic electricity pricing, including: The acquisition module 201 is used to collect real-time data on photovoltaic output, energy storage SOC, load demand and time-of-use electricity price; Determination module 202, for determining the dynamic control mode corresponding to photovoltaic output, energy storage SOC, load demand and time-of-use electricity price data by combining the economic target model, environmental target model and constraint conditions; The photovoltaic storage and charging synergy arbitrage module 203 is used to dynamically determine the energy storage charging and discharging strategy when the dynamic control mode is the photovoltaic storage and charging synergy arbitrage mode; The virtual power plant aggregation module 204 is used to form a dispatch capacity by aggregating distributed resources and respond to power grid peak load regulation instructions when the dynamic control mode of operation is the virtual power plant aggregation mode; The carbon-electricity joint optimization module 205 is used to connect to the carbon trading platform and sell green certificates in bundles when the dynamic control mode of operation is the carbon-electricity joint optimization mode.

[0069] Figure 3 Schematic diagram of the structure of the electronic device provided in this embodiment.

[0070] like Figure 3As shown, the electronic device may include: a processor 310, a communication interface 320, a memory 330, and a communication bus 340, wherein the processor 310, the communication interface 320, and the memory 330 communicate with each other via the communication bus 340. The processor 310 may call logic instructions in the memory 330 to execute a regional smart electricity consumption method based on source-grid-load-storage coordination and dynamic electricity pricing. The method includes: real-time collection of photovoltaic output, energy storage SOC, load demand, and time-of-use electricity price data; combining an economic target model, an environmental target model, and constraints to determine a dynamic control mode corresponding to the photovoltaic output, energy storage SOC, load demand, and time-of-use electricity price data; when the dynamic control mode is a photovoltaic-storage-charging coordinated arbitrage mode, dynamically determining the energy storage charging and discharging strategy; when the dynamic control mode is a virtual power plant aggregation mode, forming a dispatch capacity by aggregating distributed resources to respond to grid peak load regulation instructions; when the dynamic control mode is a carbon-electricity joint optimization mode, connecting to a carbon trading platform to bundle and sell green certificates.

[0071] Furthermore, the logic instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product, stored in a storage medium, includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0072] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the regional intelligent electricity consumption method based on source-grid-load-storage coordination and dynamic electricity price provided by the above methods. The method includes: real-time collection of photovoltaic output, energy storage SOC, load demand and time-of-use electricity price data; combining the economic target model, environmental protection target model and constraint conditions to determine the dynamic control mode corresponding to the photovoltaic output, energy storage SOC, load demand and time-of-use electricity price data; when the dynamic control mode is the photovoltaic-storage-charging coordinated arbitrage mode, the energy storage charging and discharging strategy is dynamically determined; when the dynamic control mode is the virtual power plant aggregation mode, the dispatching capacity is formed by aggregating distributed resources to respond to the peak-shaving instructions of the power grid; when the dynamic control mode is the carbon-electricity joint optimization mode, the carbon trading platform is connected to carry out green certificate bundled sales.

[0073] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a regional intelligent electricity consumption method based on source-grid-load-storage coordination and dynamic electricity price provided by the above-mentioned methods, the method comprising: real-time collection of photovoltaic output, energy storage SOC, load demand and time-of-use electricity price data; combining the economic target model, the environmental protection target model and the constraint conditions, determining the dynamic control mode corresponding to the photovoltaic output, energy storage SOC, load demand and time-of-use electricity price data; when the dynamic control mode is the photovoltaic-storage-charging coordinated arbitrage mode, dynamically determining the energy storage charging and discharging strategy; when the dynamic control mode is the virtual power plant aggregation mode, forming a dispatching capacity by aggregating distributed resources to respond to the peak-shaving instructions of the power grid; when the dynamic control mode is the carbon-electricity joint optimization mode, connecting to the carbon trading platform to carry out green certificate bundled sales.

[0074] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0075] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0076] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A regional intelligent electricity consumption method based on source-grid-load-storage coordination and dynamic electricity pricing, characterized in that: include: Real-time collection of photovoltaic output, energy storage SOC, load demand and time-of-use electricity price data; Determine the dynamic control mode corresponding to the photovoltaic output, energy storage SOC, load demand and time-of-use electricity price data by combining the economic target model, environmental target model and constraint conditions; When the dynamic control mode is a photovoltaic storage and charging arbitrage mode, dynamically determining the energy storage charging and discharging strategy; When the dynamic control mode in operation is the virtual power plant aggregation mode, the dispatch capacity is formed by aggregating distributed resources to respond to the peak load regulation instructions of the power grid; When the dynamic control mode in operation is the carbon-electricity joint optimization mode, it is connected to the carbon trading platform to carry out green certificate bundled sales.

2. The regional intelligent electricity consumption method based on source-grid-load-storage coordination and dynamic electricity price according to claim 1 is characterized in that: When the dynamic control mode is a photovoltaic storage and charging synergy arbitrage mode, dynamically determining the energy storage charging and discharging strategy includes: Determine the data input system corresponding to multi-source data fusion, and use a hybrid algorithm framework based on the data input system to perform dynamic electricity price forecasting; Determine the state space, action space, and reward function of the predicted dynamic electricity price using a photovoltaic storage and charging collaborative arbitrage model; Based on the state space, action space and reward function, combined with a rolling optimization mechanism, abnormal event response and online learning, the energy storage charging period and the discharging period are dynamically determined.

3. The regional intelligent electricity consumption method based on source-grid-load-storage coordination and dynamic electricity price according to claim 2 is characterized in that: The state space, action space, and reward function of determining the predicted dynamic electricity price using the photovoltaic storage and charging collaborative arbitrage model include: Based on the photovoltaic storage and charging collaborative arbitrage model, a Markov decision process is constructed using deep reinforcement learning; The Markov decision process is used to determine the real-time electricity price, the electricity price forecast curve for a preset period, the real-time photovoltaic power and forecast curve, the energy storage SOC / SOH, the current total charging demand and the load distribution of each charging pile, and the maximum charging and discharging power allowed by the power grid as the state space; Determine the energy storage charging power, energy storage discharging power, and charging pile power allocation using the Markov decision process as an action space; The Markov decision process is used to determine arbitrage benefits, photovoltaic absorption rewards, battery life penalties, user satisfaction and grid constraint penalties as reward functions.

4. The regional intelligent electricity consumption method based on source-grid-load-storage coordination and dynamic electricity pricing according to claim 2 is characterized in that: The method dynamically determines the energy storage charging period and discharging period based on the state space, action space and reward function, combined with a rolling optimization mechanism, abnormal event response and online learning, including: Determining the operating status and external environment of the power grid system through the state space; In combination with the operating conditions and external environment, selecting an energy storage charging action, an energy storage discharging action, or an charging pile power adjustment action from the action space; After completing any of the energy storage charging actions, energy storage discharging actions, or charging pile power adjustment actions, a reward function is determined based on the revenue; By rolling optimization and updating the charging and discharging plan, and combining the abnormal event response with online learning of the charging and discharging strategy, the energy storage charging and discharging periods are determined.

5. The regional intelligent electricity consumption method based on source-grid-load-storage coordination and dynamic electricity price according to claim 1 is characterized in that: When the dynamic control mode is a virtual power plant aggregation mode, the distributed resources are aggregated to form a dispatch capacity to respond to a power grid peak load regulation instruction, including: Build a distributed resource access platform to collect real-time data on power generation, remaining energy storage capacity, and load power consumption; Combined with the power generation of the power source, the remaining power of the energy storage, and the real-time power consumption of the load, the power generation trend of the distributed power source, the change pattern of the load, and the charging and discharging capacity of the energy storage are predicted to build a resource capacity map of the virtual power plant; Taking the peak-shaving demand of the power grid as the target and combining the resource capability map, a resource scheduling plan is formulated to respond to the peak-shaving instruction of the power grid.

6. The regional intelligent electricity consumption method based on source-grid-load-storage coordination and dynamic electricity price according to claim 1 is characterized in that: When the dynamic control mode is the carbon-electricity joint optimization mode, the carbon trading platform is connected to conduct green certificate bundled sales, including: By building a multi-data perception network through the perception layer, integrating it with the carbon price trend and green certificate supply and demand data of the carbon trading platform, a dynamic panoramic view of carbon-electricity data is formed; Based on the panoramic view of carbon-electricity data, the decision-making layer is used to build an intelligent decision-making center to dynamically generate green certificate bundling sales strategies; Based on the green certificate bundling sales strategy, the carbon-electricity trading alliance chain and smart contract trading model are used to realize the green certificate bundling sales.

7. The regional intelligent electricity consumption method based on source-grid-load-storage coordination and dynamic electricity price according to claim 6 is characterized in that: The perception layer is used to build a data perception network, which is integrated with the carbon price trend and green certificate supply and demand data of the carbon trading platform to form a dynamic carbon-electricity data panorama, including: On the power generation side, real-time data collection is used to measure the power generation, power generation period, and equipment operating status of photovoltaic and wind power. On the power consumption side, data collection is used to obtain the power consumption curves and energy consumption structures of enterprises and residents. Use environmental monitoring equipment combined with meteorological conditions to estimate power generation data; The photovoltaic and wind power generation, power generation period, equipment operating status, electricity consumption curves and energy consumption structure of the enterprises and residents are integrated with the carbon price trend and green certificate supply and demand data of the carbon trading platform to form a dynamic panoramic view of carbon-electricity data.

8. The regional intelligent electricity consumption method based on source-grid-load-storage coordination and dynamic electricity pricing according to claim 6 is characterized in that: Also includes: Establish a carbon-electricity joint optimization effect evaluation system; By using the carbon-electricity joint optimization effect evaluation system, the green certificate bundling sales strategy is quantitatively evaluated from different dimensions to achieve an adaptive optimization closed loop.

9. A regional intelligent power consumption system based on source-grid-load-storage coordination and dynamic electricity pricing, characterized by: include: The acquisition module is used to collect real-time data on photovoltaic output, energy storage SOC, load demand, and time-of-use electricity prices; A determination module is used to determine the dynamic control mode corresponding to the photovoltaic output, energy storage SOC, load demand and time-of-use electricity price data by combining the economic target model, the environmental target model and the constraint conditions; A photovoltaic storage and charging collaborative arbitrage module is used to dynamically determine the energy storage charging and discharging strategy when the dynamic control mode is the photovoltaic storage and charging collaborative arbitrage mode; A virtual power plant aggregation module, configured to form a dispatch capacity by aggregating distributed resources and respond to power grid peak load regulation instructions when the dynamic control mode in operation is the virtual power plant aggregation mode; The carbon-electricity joint optimization module is used to connect to the carbon trading platform and sell green certificates in bundled form when the dynamic control mode in operation is the carbon-electricity joint optimization mode.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, it implements the regional intelligent electricity consumption method based on source-grid-load-storage coordination and dynamic electricity price as described in any one of claims 1 to 8.

Images