A method and system for integrated energy dynamic regulation in multi-energy coupling scenarios

By collecting and forecasting multi-energy load, meteorological data, and energy prices, and combining a multi-energy coupling regulation strategy library and reinforcement learning models, dynamic regulation of the integrated energy system is achieved, solving the problems of supply and demand balance and stability, and improving regulation efficiency and economy.

CN123121836APending Publication Date: 2026-11-17HUAKANG SHENGTAI ENVIRONMENTAL TECH (BEIJING) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611263109.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-19
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In integrated energy systems with multi-energy coupling scenarios, existing single-energy dispatching methods cannot perceive the coupling effect of supply and demand fluctuations between different energy sources in real time, resulting in insufficient supply and demand balance and stability, triggering chain reactions and low regulation efficiency.

Method used

By collecting multi-energy load, weather forecasts, and real-time energy price data, and using a comprehensive energy time-series forecasting model for supply and demand prediction, combined with a multi-energy coupling regulation strategy library and reinforcement learning model, the system achieves coordinated scheduling of distributed energy storage devices and renewable energy, dynamic regulation, and closed-loop feedback optimization.

Benefits of technology

It enables real-time sensing of supply and demand fluctuations in the electricity, heat, and cooling energy subsystems, ensuring the accuracy and stability of energy supply and demand balance, improving overall regulation efficiency, and adaptively adjusting to enhance the economy and reliability of system operation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123121836A_ABST
    Figure CN123121836A_ABST
Patent Text Reader

Abstract

This invention relates to the field of integrated energy system optimization and scheduling technology, and discloses an integrated energy dynamic adjustment method and system in a multi-energy coupling scenario. The system includes a data acquisition and sensing module, an energy prediction and analysis module, a strategy matching and decision-making module, a collaborative scheduling and control module, and a feedback optimization and learning module. By collecting multi-energy load data, meteorological forecast data, and real-time energy price data, and performing short-term energy supply and demand forecasts based on these data, it can perceive the supply and demand fluctuation coupling effects between multiple energy subsystems such as electricity, heat, and cooling in real time, avoiding the chain reaction caused by neglecting coupling relationships in a single energy scheduling method, and ensuring the accuracy and stability of energy supply and demand balance.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of integrated energy system optimization and scheduling technology, specifically to an integrated energy dynamic adjustment method and system in a multi-energy coupling scenario. Background Technology

[0002] An integrated energy system refers to a system that utilizes advanced physical and information technologies and innovative management models within a certain region to integrate various energy sources such as coal, oil, natural gas, electricity, and heat, and to achieve coordinated planning, optimized operation, collaborative management, interactive response, and mutual support among multiple heterogeneous energy subsystems.

[0003] Currently, due to the strong coupling relationship between various energy sources such as electricity, heat, and cooling in integrated energy systems, existing single-energy dispatching methods cannot perceive the supply and demand fluctuation coupling effect between different energy sources in real time when performing dynamic adjustment in multi-energy coupling scenarios. When a supply and demand imbalance occurs in a certain energy subsystem, it will trigger a chain reaction in other energy subsystems, causing a decrease in the overall system regulation efficiency and failing to guarantee the accuracy and stability of energy supply and demand balance.

[0004] Therefore, a comprehensive energy dynamic regulation method and system for multi-energy coupling scenarios is proposed to solve the above problems. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a comprehensive energy dynamic regulation method and system for multi-energy coupling scenarios, which solves the problems mentioned in the background technology, such as the decline in overall system regulation efficiency and the inability to guarantee the accuracy and stability of energy supply and demand balance.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method and system for dynamic regulation of integrated energy in a multi-energy coupling scenario, the method comprising the following steps: S1. Collect multi-energy load data, meteorological forecast data, and real-time energy price data of the integrated energy system; S2. Based on the multi-energy load data, the meteorological forecast data, and the real-time energy price data, perform energy supply and demand forecasting processing to generate short-term energy supply and demand forecast data. S3. Based on the short-term energy supply and demand forecast data, perform energy system operation status assessment processing to generate system operation status assessment data; S4. When the system operation status assessment data indicates that there is a risk of energy supply and demand imbalance, the system performs adjustment strategy matching processing based on the short-term energy supply and demand forecast data and the preset multi-energy coupling adjustment strategy library to generate target adjustment strategy data. S5. Based on the target adjustment strategy data, perform coordinated scheduling processing on distributed energy storage devices, renewable energy power generation devices, and controllable loads to generate device coordinated scheduling instruction data. S6. Send the device collaborative scheduling instruction data to the corresponding execution terminal to perform dynamic energy adjustment processing, and simultaneously obtain the system operation status data after dynamic adjustment; S7. Based on the dynamically adjusted system operating status data, monitor the adjusted system operating status in real time, and perform closed-loop feedback optimization processing based on the monitoring data to update the preset multi-energy coupling adjustment strategy library.

[0007] Preferably, the step S1, which involves collecting multi-energy load data, meteorological forecast data, and real-time energy price data from the integrated energy system, includes the following steps: S11. Collect the multi-energy load data, including electrical load, thermal load, and cooling load, through a smart sensor network deployed on the user side; S12. Obtain the meteorological forecast data, including light intensity, wind speed, and ambient temperature, by accessing the meteorological service platform; S13. Obtain real-time energy price data, including peak-valley electricity prices and real-time electricity prices, through data interfaces with the electricity trading market and energy price platform.

[0008] Preferably, the generation of short-term energy supply and demand forecast data in step S2 includes the following steps: S21. Input the multi-energy load data, the meteorological forecast data, and the real-time energy price data into the pre-trained integrated energy time series forecast model; S22. The integrated energy time-series prediction model is based on a long short-term memory network and combined with an attention mechanism to predict the energy supply and demand situation within a preset time window in the future, and generates the short-term energy supply and demand prediction data, which includes the difference between electricity supply and demand, heat supply and demand, and cold supply and demand at future times.

[0009] Preferably, the generation of system operating status evaluation data in step S3 includes the following steps: S31. Obtain the short-term energy supply and demand forecast data; S32. Compare the short-term energy supply and demand forecast data with preset safe operation thresholds, the preset safe operation thresholds including electricity supply and demand balance threshold, heat supply and demand balance threshold and cold energy supply and demand balance threshold. When any supply-demand difference in the short-term energy supply and demand forecast data exceeds the corresponding preset safe operation threshold, the system operation status assessment data is determined to indicate a risk of energy supply and demand imbalance. When all supply and demand gaps are within the corresponding preset safe operating thresholds, the system operating status assessment data is determined to be in normal operating status.

[0010] Preferably, generating the target adjustment strategy data in step S4 includes the following steps: S41. Establish the preset multi-energy coupling regulation strategy library. The preset multi-energy coupling regulation strategy library contains multiple predefined regulation strategies. Each regulation strategy is associated with a specific set of energy supply and demand imbalance scenarios and corresponding equipment collaborative scheduling schemes. S42. When the system operation status assessment data indicates a risk of energy supply and demand imbalance, the short-term energy supply and demand forecast data is matched with the energy supply and demand imbalance scenario corresponding to each regulation strategy in the preset multi-energy coupling regulation strategy library. S43. Search for the adjustment strategy that best matches the short-term energy supply and demand forecast data, and determine it as the target adjustment strategy data.

[0011] Preferably, the generation of device collaborative scheduling instruction data in step S5 includes the following steps: S51. Analyze the target adjustment strategy data to obtain the required equipment collaborative scheduling scheme. The equipment collaborative scheduling scheme includes the power setpoints and start / stop status of phase change thermal storage devices, phase change cold storage devices, electrochemical energy storage devices, air source heat pumps, photovoltaic power generation equipment and wind power generation equipment. S52. Based on the device collaborative scheduling scheme, generate the device collaborative scheduling instruction data, which includes power adjustment instructions, charging and discharging instructions and start / stop instructions for each device.

[0012] Preferably, the dynamic energy adjustment process in step S6, and the simultaneous acquisition of the dynamically adjusted system operating status data, includes the following steps: S61. The device collaborative scheduling instruction data is sent to the controller of each execution terminal through the Internet of Things communication protocol; S62. The controllers of each execution terminal control the corresponding distributed energy storage devices, renewable energy power generation devices and controllable loads to perform actions according to the received instructions, so as to realize the dynamic regulation of energy. The dynamic regulation includes peak shaving and valley filling, promoting the local consumption of renewable energy and multi-energy complementarity. S63. During the dynamic adjustment process, the system operation status data after dynamic adjustment is acquired and recorded in real time as the basis for closed-loop feedback optimization. The system operation status data includes the real-time power of each device, the state of charge of the energy storage device, and the actual balance between energy supply and demand.

[0013] Preferably, updating the preset multi-energy coupling adjustment strategy library in S7 includes the following steps: S71. Based on the dynamically adjusted system operating status data, monitor the actual supply and demand balance of the system and the equipment operating status in real time. S72. Compare and analyze the monitored system operation status data with the short-term energy supply and demand forecast data, and calculate the adjustment deviation; S73. When the adjustment deviation exceeds a preset deviation threshold, closed-loop feedback optimization is triggered. Using a reinforcement learning model, the preset multi-energy coupling adjustment strategy library is updated online according to the adjustment deviation to optimize the subsequent adjustment strategy matching process.

[0014] Preferably, the system includes a data acquisition and sensing module, an energy prediction and analysis module, a strategy matching and decision-making module, a collaborative scheduling and control module, and a feedback optimization and learning module.

[0015] Preferably, the system specifically includes: The data acquisition and sensing module collects multi-energy load data through intelligent sensor networks and multi-energy meters, receives meteorological forecast data through meteorological service interfaces, receives real-time energy price data through energy trading platform interfaces, and outputs the above three types of data to the energy forecasting and analysis module. The energy forecasting and analysis module receives the multi-energy load data, meteorological forecast data and real-time energy price data, performs energy supply and demand forecasting processing through the time-series forecasting unit, and performs operational status assessment processing through the status assessment unit, and outputs short-term energy supply and demand forecast data and system operational status assessment data. The strategy matching and decision-making module receives the short-term energy supply and demand forecast data and the system operation status assessment data, performs matching processing with the preset multi-energy coupling regulation strategy library through the strategy matching unit, and outputs the target regulation strategy data. The collaborative scheduling and control module receives the target adjustment strategy data, generates equipment collaborative scheduling instruction data through the instruction generation unit, sends it to the execution terminal through the communication interface unit, and obtains the dynamically adjusted system operation status data. The feedback optimization and learning module receives the dynamically adjusted system operation status data, compares and analyzes it with the short-term energy supply and demand forecast data through the deviation calculation unit, and updates the preset multi-energy coupling regulation strategy library online through the strategy update unit.

[0016] Compared with the prior art, the present invention provides a comprehensive energy dynamic regulation method and system for multi-energy coupling scenarios, which has the following beneficial effects: 1. In this invention, by collecting multi-energy load data, meteorological forecast data, and real-time energy price data, and making short-term energy supply and demand forecasts based on these data, it is possible to perceive the supply and demand fluctuation coupling effect between multiple energy subsystems such as electricity, heat, and cooling in real time. This avoids the chain reaction caused by ignoring coupling relationships in a single energy dispatching method, and ensures the accuracy and stability of energy supply and demand balance.

[0017] 2. In this invention, by comparing short-term energy supply and demand forecast data with preset safe operation thresholds, subsystems with potential energy supply and demand imbalance risks can be accurately identified. Based on a preset multi-energy coupling regulation strategy library, the optimal target regulation strategy is matched, thereby enabling coordinated scheduling of distributed energy storage devices, renewable energy power generation devices, and controllable loads. This allows different energy sources to achieve peak shaving and valley filling, multi-energy complementarity, and improves the overall system regulation efficiency.

[0018] 3. In this invention, the system operation status data after adjustment is monitored in real time, and the data is compared and analyzed with short-term energy supply and demand forecast data to calculate the adjustment deviation. When the deviation exceeds the preset threshold, closed-loop feedback optimization is triggered. The preset multi-energy coupling adjustment strategy library is updated online using a reinforcement learning model, so that the adjustment strategy can be adaptively adjusted according to the dynamic changes of user energy consumption patterns, meteorological conditions and energy prices, thereby improving the economy and reliability of the integrated energy system operation. Attached Figure Description

[0019] Figure 1 This is a flowchart of a comprehensive energy dynamic adjustment method in a multi-energy coupling scenario according to the present invention; Figure 2 This is a schematic diagram of the integrated energy dynamic regulation system in a multi-energy coupling scenario according to the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] For specific implementation examples, please refer to: Figures 1-2 A method and system for integrated energy dynamic regulation in multi-energy coupling scenarios, the method comprising the following steps: S1. Collect multi-energy load data, meteorological forecast data, and real-time energy price data of the integrated energy system; S2. Based on multi-energy load data, meteorological forecast data and real-time energy price data, perform energy supply and demand forecasting to generate short-term energy supply and demand forecast data. S3. Based on short-term energy supply and demand forecast data, perform energy system operation status assessment and generate system operation status assessment data. S4. When the system operation status assessment data indicates that there is a risk of energy supply and demand imbalance, the system performs adjustment strategy matching processing based on short-term energy supply and demand forecast data and the preset multi-energy coupling adjustment strategy library to generate target adjustment strategy data. S5. Based on the target adjustment strategy data, perform coordinated scheduling processing on distributed energy storage devices, renewable energy power generation devices, and controllable loads to generate device coordinated scheduling instruction data; S6. Send the equipment collaborative scheduling instruction data to the corresponding execution terminal to perform dynamic energy adjustment processing, and simultaneously obtain the system operation status data after dynamic adjustment; S7. Based on the dynamically adjusted system operating status data, monitor the adjusted system operating status in real time, and perform closed-loop feedback optimization processing based on the monitoring data to update the preset multi-energy coupling adjustment strategy library.

[0022] The steps involved in collecting multi-energy load data, meteorological forecast data, and real-time energy price data from the integrated energy system in S1 are as follows: S11. Collect multi-energy load data, including electrical load, thermal load, and cooling load, through a smart sensor network deployed on the user side; S12. Obtain meteorological forecast data, including light intensity, wind speed, and ambient temperature, by accessing the meteorological service platform; S13. By connecting with the electricity trading market and energy price platform, real-time energy price data, including peak-valley electricity prices and real-time electricity prices, can be obtained in real time.

[0023] The steps involved in generating short-term energy supply and demand forecast data in S2 are as follows: S21. Input multi-energy load data, meteorological forecast data, and real-time energy price data into the pre-trained integrated energy time series forecast model; S22. The integrated energy time-series forecasting model, based on a long short-term memory network and combined with an attention mechanism, predicts the energy supply and demand situation within a preset future time window and generates short-term energy supply and demand forecast data. This data includes the future supply and demand differences for electricity, heat, and cooling energy, and comprises the following steps: S221. Input the multi-energy load data, meteorological forecast data, and real-time energy price data into the input gate, forget gate, and output gate of the long short-term memory network in sequence according to time steps, and extract the time-series feature sequence. S222. Input the temporal feature sequence into the attention mechanism layer, and generate a weighted context feature vector by calculating the attention weights between the features at each time step and the current prediction target. The attention weights are calculated according to the following formula: ; ; in, For the first Attention weights at each time step The total number of time steps. For the first The hidden state vector of the Long Short-Term Memory network at each time step This is the weight matrix. For bias vectors, For attention vectors, For the first Attention score at each time step For the first Attention score at each time step This is the transpose symbol.

[0024] Generating system operational status assessment data in S3 includes the following steps: S31. Obtain short-term energy supply and demand forecast data; S32. Compare short-term energy supply and demand forecast data with preset safe operating thresholds. The preset safe operating thresholds include electricity supply and demand balance thresholds, heat supply and demand balance thresholds, and cooling supply and demand balance thresholds. The preset safe operating thresholds adopt a dynamic threshold setting method, including the following steps: S321. Obtain historical data on electricity supply and demand balance, heat supply and demand balance, and cold supply and demand balance for the same period in history, and calculate the historical mean and standard deviation of the supply and demand difference for each energy source. S322. The standard deviation of the historical mean plus or minus a preset multiple is used as the initial threshold range; wherein, the initial threshold range is calculated according to the following formula: ; ; in, This represents the average of the supply and demand differences for various energy sources during the same historical period. Standard deviation, For preset multiples, and These are the lower threshold and the upper threshold, respectively; S323. Based on the current season, weather type, and holiday attributes, the initial threshold range is corrected. The correction coefficient is given by a pre-established mapping table, and finally the power supply and demand balance threshold, heat supply and demand balance threshold, and cold supply and demand balance threshold are obtained. When any supply-demand difference in the short-term energy supply and demand forecast data exceeds the corresponding preset safe operation threshold, the system operation status assessment data is determined to indicate that there is a risk of energy supply and demand imbalance. When all supply and demand gaps are within the corresponding preset safe operating thresholds, the system operating status assessment data is determined to be in normal operating status.

[0025] The steps involved in generating target adjustment strategy data in S4 are as follows: S41. Establish a preset multi-energy coupling regulation strategy library. The preset multi-energy coupling regulation strategy library contains multiple predefined regulation strategies. Each regulation strategy is associated with a specific set of energy supply and demand imbalance scenarios and corresponding equipment collaborative scheduling schemes. S42. When the system operation status assessment data indicates a risk of energy supply and demand imbalance, the short-term energy supply and demand forecast data is matched with the energy supply and demand imbalance scenarios corresponding to each regulation strategy in the preset multi-energy coupling regulation strategy library, including the following steps: S421. Set a similarity threshold. The value ranges from 0 to 1, and the similarity threshold is... Determined based on historical matching data, specifically including: collecting cosine similarity data between successfully matched adjustment strategies and their corresponding energy supply and demand imbalance scenarios within historical periods, and calculating the average value of the cosine similarity data. and standard deviation , similarity threshold Set as ,in This is a preset coefficient, with a value range of 0.5 to 2; S422. Extract the magnitude and trend of the electricity supply-demand difference, heat supply-demand difference, and cold supply-demand difference from the short-term energy supply-demand forecast data. S423. Extract the expected range and trend characteristics of the energy supply and demand imbalance in each regulation strategy; S424. Using the cosine similarity algorithm, short-term energy supply and demand forecast data is matched with energy supply and demand imbalance scenarios. The similarity between the changing trends of the two is calculated, and those with similarity exceeding a similarity threshold are considered. The adjustment strategies are included in the candidate set to complete the matching of short-term energy supply and demand forecast data with energy supply and demand imbalance scenarios; The cosine similarity is calculated using the following formula: ; in, For vectors and cosine similarity, This is a vector representing the changing trends of short-term energy supply and demand forecast data, where each element represents the difference in supply and demand between adjacent time steps. This represents the changing trend vector for scenarios of energy supply and demand imbalance, where each element represents the expected difference in the corresponding scenario. and Let be the magnitudes of the two vectors, respectively. S43. Search for the adjustment strategy that best matches the short-term energy supply and demand forecast data, and determine it as the target adjustment strategy data, including the following steps: S431. Construct the first feature vector from the electricity supply-demand difference, heat supply-demand difference, and cold supply-demand difference in the short-term energy supply-demand forecast data. S432. Construct the expected supply and demand difference in the energy supply and demand imbalance scenario corresponding to each regulation strategy in the preset multi-energy coupling regulation strategy library as the second feature vector. S433. Calculate the Euclidean distance between the first eigenvector and each second eigenvector, and determine the adjustment strategy with the smallest Euclidean distance as the adjustment strategy with the highest matching degree with the short-term energy supply and demand forecast data; wherein, the Euclidean distance is calculated according to the following formula: ; in, For vectors and The Euclidean distance between them The first eigenvector, , , These represent the predicted supply and demand gaps for electricity, heat, and cooling energy, respectively. This is the second feature vector. , , These represent the expected power supply-demand gap, expected heat supply-demand gap, and expected cooling supply-demand gap for the corresponding scenarios of the adjustment strategy, respectively.

[0026] The steps involved in generating device collaborative scheduling instruction data in S5 are as follows: S51. Analyze the target adjustment strategy data to obtain the required equipment collaborative scheduling scheme. The equipment collaborative scheduling scheme includes the power setpoints and start / stop status of phase change thermal storage devices, phase change cold storage devices, electrochemical energy storage devices, air source heat pumps, photovoltaic power generation equipment and wind power generation equipment. S52. Based on the equipment collaborative scheduling scheme, generate equipment collaborative scheduling instruction data. The equipment collaborative scheduling instruction data includes power adjustment instructions, charging and discharging instructions, and start / stop instructions for each device. Specific steps include: S521. Analyze the equipment collaborative scheduling scheme, obtain the power setting value, start / stop status and charging / discharging instructions of each device, and establish a priority list according to the device type. Among them, the priority of phase change heat storage device and phase change cold storage device is higher than that of electrochemical energy storage device, and the priority of air source heat pump is higher than that of photovoltaic power generation equipment and wind power generation equipment. S522. According to the priority list, check the current operating constraints of each device in sequence. The constraints include maximum power limit, minimum power limit, upper and lower limits of state of charge, charging and discharging power limit, and device start-stop count limit. S523. For devices that meet the constraints, directly generate the corresponding power adjustment command, charge / discharge command, and start / stop command; for devices that do not meet the constraints, adjust their power setting value and charge / discharge power to the constraint range according to priority, and recalculate the power allocation of the remaining devices until all device commands meet the constraints; wherein, the recalculation of power allocation is performed according to the following formula: ; in, For the first The adjusted power setting value of the device. and These are the maximum and minimum allowable power of the device, respectively. The power setpoint in the original scheduling scheme, The total number of devices participating in the scheduling. The total power requirement for all devices is determined by the target regulation strategy.

[0027] The dynamic energy adjustment process in S6, and the simultaneous acquisition of the dynamically adjusted system operating status data, includes the following steps: S61. Send the device collaborative scheduling instruction data to the controller of each execution terminal through the Internet of Things communication protocol; S62. The controllers of each execution terminal control the corresponding distributed energy storage devices, renewable energy power generation devices and controllable loads to perform actions according to the received instructions, so as to realize the dynamic regulation of energy. The dynamic regulation includes peak shaving and valley filling, promoting the local consumption of renewable energy and multi-energy complementarity. S63. During the dynamic adjustment process, the system operation status data after dynamic adjustment is acquired and recorded in real time as the basis for closed-loop feedback optimization. The system operation status data includes the real-time power of each device, the state of charge of the energy storage device, and the actual balance between energy supply and demand.

[0028] Updating the preset multi-energy coupling regulation strategy library in S7 includes the following steps: S71. Based on the dynamically adjusted system operating status data, monitor the actual supply and demand balance of the system and the operating status of the equipment in real time. S72. Compare and analyze the monitored system operating status data with short-term energy supply and demand forecast data, and calculate the adjustment deviation, including the following steps: S721. Obtain the actual power supply and demand difference, actual heat supply and demand difference, and actual cold supply and demand difference from the dynamically adjusted system operating status data. S722. Obtain the predicted electricity supply and demand difference, predicted heat supply and demand difference, and predicted cold supply and demand difference from the short-term energy supply and demand forecast data. S723. Calculate the sum of the absolute values ​​of the differences between the actual and predicted values ​​for each energy type, as the adjustment deviation: ; in, To adjust the deviation, , , These are the supply and demand differences for actual electrical energy, heat energy, and cooling energy, respectively. , , These are the corresponding projected supply and demand gaps; S73. When the adjustment deviation exceeds a preset deviation threshold, closed-loop feedback optimization is triggered. Using a reinforcement learning model, the preset multi-energy coupling adjustment strategy library is updated online based on the adjustment deviation to optimize the subsequent adjustment strategy matching process, including the following steps: S731. Encode the adjustment deviation, the current system operating state data, and the currently executed adjustment strategy into a state vector, which serves as the state input to the reinforcement learning model. S732. Define the adjustment direction of each adjustment strategy in the preset multi-energy coupling adjustment strategy library as the action space. The action space includes increasing the power setting value of each power schedulable device, decreasing the power setting value of each power schedulable device, and switching adjustment strategies. S733. The weighted sum of the reduction in system operating costs and the improvement in supply-demand balance after adjustment is used to obtain a comprehensive reward value as the immediate reward signal. A deep Q-network algorithm is employed to update the network parameters of the reinforcement learning model, aiming to maximize the cumulative reward. Based on the updated network parameters, the optimal action is selected to adjust the corresponding adjustment strategy in the pre-set multi-energy coupling adjustment strategy library. The specific steps for updating the network parameters of the reinforcement learning model include: S7311. Initialize the experience replay pool, setting its capacity to the preset value; S7312. After each adjustment is performed, the current state, the selected action, the immediate reward obtained, and the next state are combined into a quadruple and stored in the experience replay pool. S7313. Randomly sample a batch of quadruplets from the experience replay pool, calculate the target Q value, and use the gradient descent method to update the parameters of the deep Q network so that the current Q value approaches the target Q value.

[0029] The target Q-value of the deep Q-network algorithm is calculated according to the following formula: ; in, For the target Q value, For immediate reward signals, This is the discount factor, with a value between 0 and 1. The next state vector, For the next moment's action, For the parameters of the target network, For the target network state Take action below Q-value estimation; The parameter updates for gradient descent are calculated using the following formula: ; in, These are the parameters of the current network. For learning rate, This represents the number of quadruplets sampled in the batch. and The first The state and action in a quadruple. For the first The target Q value corresponding to each quadruple. For the current network state Take action below Q-value estimation.

[0030] The system includes a data acquisition and sensing module, an energy prediction and analysis module, a strategy matching and decision-making module, a collaborative scheduling and control module, and a feedback optimization and learning module.

[0031] The system specifically includes: The data acquisition and sensing module collects multi-energy load data through intelligent sensor networks and multi-energy meters, receives meteorological forecast data through meteorological service interfaces, receives real-time energy price data through energy trading platform interfaces, and outputs the above three types of data to the energy forecasting and analysis module. The energy forecasting and analysis module receives multi-energy load data, meteorological forecast data and real-time energy price data, performs energy supply and demand forecasting through the time series forecasting unit, and performs operational status assessment through the status assessment unit, outputting short-term energy supply and demand forecast data and system operational status assessment data. The strategy matching and decision-making module receives short-term energy supply and demand forecast data and system operation status assessment data, and performs matching processing with the preset multi-energy coupling regulation strategy library through the strategy matching unit, and outputs target regulation strategy data. The collaborative scheduling and control module receives target adjustment strategy data, generates equipment collaborative scheduling instruction data through the instruction generation unit, sends it to the execution terminal through the communication interface unit, and obtains the dynamically adjusted system operation status data. The feedback optimization and learning module receives the dynamically adjusted system operation status data, compares and analyzes it with short-term energy supply and demand forecast data through the deviation calculation unit, and updates the preset multi-energy coupling regulation strategy library online through the strategy update unit.

[0032] The operation steps of the integrated energy dynamic regulation method and system in a multi-energy coupling scenario are as follows: Step 1: Data Acquisition and Sensing First, the system collects multi-energy load data from the integrated energy system through a smart sensor network deployed on the user side. This data includes electrical load, heat load, and cooling load. Simultaneously, the system obtains weather forecast data, including solar irradiance, wind speed, and ambient temperature, by connecting to a meteorological service platform. Furthermore, the system acquires real-time energy price data, including peak-valley electricity prices and real-time electricity prices, through data interfaces with the electricity trading market and energy pricing platforms. These three types of data collectively form the foundational input for subsequent analysis and decision-making.

[0033] Step 2: Energy Supply and Demand Forecasting The system inputs collected multi-energy load data, meteorological forecast data, and real-time energy price data into a pre-trained integrated energy time-series forecasting model. This model, based on a long short-term memory network and incorporating an attention mechanism, predicts energy supply and demand within a preset future time window, generating short-term energy supply and demand forecast data. This forecast data specifically includes the future supply and demand differences for electricity, heat, and cooling, thereby achieving synchronous perception of supply and demand fluctuations in the three energy subsystems: electricity, heat, and cooling.

[0034] Step 3: Operational Status Assessment After acquiring short-term energy supply and demand forecast data, the system compares it with preset safe operating thresholds. These preset thresholds include electricity supply and demand balance thresholds, heat supply and demand balance thresholds, and cooling supply and demand balance thresholds. When any supply and demand difference in the short-term energy supply and demand forecast data exceeds the corresponding safe operating threshold, the system determines that the operating status assessment data indicates a risk of energy supply and demand imbalance; when all supply and demand differences are within their respective thresholds, the operating status assessment data indicates normal operation. This assessment result provides a clear criterion for whether to initiate subsequent regulation.

[0035] Step 4: Adjust strategy matching: When system operation status assessment data indicates a risk of energy supply and demand imbalance, the system matches short-term energy supply and demand forecast data with a pre-defined multi-energy coupling regulation strategy library. This library contains multiple predefined regulation strategies, each associated with a specific set of energy supply and demand imbalance scenarios and corresponding equipment collaborative scheduling schemes. The system matches the current short-term energy supply and demand forecast data with the imbalance scenarios corresponding to each regulation strategy in the strategy library, searches for the regulation strategy with the highest matching degree, and identifies it as the target regulation strategy data.

[0036] Step 5: Coordinated Scheduling and Control The system analyzes the target adjustment strategy data to obtain the required equipment coordinated scheduling scheme. This scheme includes the power setpoints and start / stop status of phase change thermal storage devices, phase change cold storage devices, electrochemical energy storage devices, air source heat pumps, photovoltaic power generation equipment, and wind power generation equipment. Based on this scheme, the system generates equipment coordinated scheduling command data, which includes power adjustment commands, charging and discharging commands, and start / stop commands for each device, thereby realizing the coordinated scheduling of distributed energy storage devices, renewable energy power generation equipment, and controllable loads.

[0037] Step Six: Dynamically Adjust Execution and Data Synchronization Acquisition: The system transmits device collaborative scheduling instructions to the controllers of each execution terminal via an IoT communication protocol. Each execution terminal's controller, based on the received instructions, controls the corresponding distributed energy storage devices, renewable energy generation devices, and controllable loads to perform actions, achieving dynamic energy regulation. Specific aspects of dynamic regulation include peak shaving and valley filling, promoting local consumption of renewable energy, and multi-energy complementarity. During dynamic regulation, the system synchronously acquires and records real-time system operating status data after dynamic regulation. This data includes the real-time power of each device, the state of charge of the energy storage devices, and the actual balance between energy supply and demand, providing a basis for subsequent closed-loop feedback optimization.

[0038] Step 7: Closed-loop feedback optimization: The system monitors the actual supply and demand balance and equipment operating status in real time based on dynamically adjusted system operating status data. It compares and analyzes the monitored system operating status data with short-term energy supply and demand forecasts to calculate the adjustment deviation. When the adjustment deviation exceeds a preset deviation threshold, the system triggers closed-loop feedback optimization. A reinforcement learning model is used to update the preset multi-energy coupling adjustment strategy library online based on the adjustment deviation, thereby optimizing the subsequent adjustment strategy matching process. Through this mechanism, the adjustment strategy can adaptively adjust to dynamic changes in user energy consumption patterns, meteorological conditions, and energy prices, continuously improving the economy and reliability of the integrated energy system.

[0039] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0040] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for dynamic adjustment of integrated energy in a multi-energy coupling scenario, characterized in that: The method includes the following steps: S1. Collect multi-energy load data, meteorological forecast data, and real-time energy price data of the integrated energy system; S2. Based on the multi-energy load data, the meteorological forecast data, and the real-time energy price data, perform energy supply and demand forecasting processing to generate short-term energy supply and demand forecast data. S3. Based on the short-term energy supply and demand forecast data, perform energy system operation status assessment processing to generate system operation status assessment data; S4. When the system operation status assessment data indicates that there is a risk of energy supply and demand imbalance, the system performs adjustment strategy matching processing based on the short-term energy supply and demand forecast data and the preset multi-energy coupling adjustment strategy library to generate target adjustment strategy data. S5. Based on the target adjustment strategy data, perform coordinated scheduling processing on distributed energy storage devices, renewable energy power generation devices, and controllable loads to generate device coordinated scheduling instruction data. S6. Send the device collaborative scheduling instruction data to the corresponding execution terminal to perform dynamic energy adjustment processing, and simultaneously obtain the system operation status data after dynamic adjustment; S7. Based on the dynamically adjusted system operating status data, monitor the adjusted system operating status in real time, and perform closed-loop feedback optimization processing based on the monitoring data to update the preset multi-energy coupling adjustment strategy library.

2. The integrated energy dynamic adjustment method in a multi-energy coupling scenario according to claim 1, characterized in that: The steps involved in S1 to collect multi-energy load data, meteorological forecast data, and real-time energy price data of the integrated energy system are as follows: S11. Collect the multi-energy load data, including electrical load, thermal load, and cooling load, through a smart sensor network deployed on the user side; S12. Obtain the meteorological forecast data, including light intensity, wind speed, and ambient temperature, by accessing the meteorological service platform; S13. Obtain real-time energy price data, including peak-valley electricity prices and real-time electricity prices, through data interfaces with the electricity trading market and energy price platform.

3. The integrated energy dynamic adjustment method in a multi-energy coupling scenario according to claim 2, characterized in that: The generation of short-term energy supply and demand forecast data in S2 includes the following steps: S21. Input the multi-energy load data, the meteorological forecast data, and the real-time energy price data into the pre-trained integrated energy time series forecast model; S22. The integrated energy time-series prediction model is based on a long short-term memory network and combined with an attention mechanism to predict the energy supply and demand situation within a preset time window in the future, and generates the short-term energy supply and demand prediction data, which includes the difference between electricity supply and demand, heat supply and demand, and cold supply and demand at future times.

4. The integrated energy dynamic adjustment method in a multi-energy coupling scenario according to claim 3, characterized in that: The process of generating system operation status evaluation data in S3 includes the following steps: S31. Obtain the short-term energy supply and demand forecast data; S32. Compare the short-term energy supply and demand forecast data with preset safe operation thresholds, the preset safe operation thresholds including electricity supply and demand balance threshold, heat supply and demand balance threshold and cold energy supply and demand balance threshold. When any supply-demand difference in the short-term energy supply and demand forecast data exceeds the corresponding preset safe operation threshold, the system operation status assessment data is determined to indicate a risk of energy supply and demand imbalance. When all supply and demand gaps are within the corresponding preset safe operating thresholds, the system operating status assessment data is determined to be in normal operating status.

5. The integrated energy dynamic adjustment method in a multi-energy coupling scenario according to claim 4, characterized in that: The step of generating target adjustment strategy data in S4 includes the following steps: S41. Establish the preset multi-energy coupling regulation strategy library. The preset multi-energy coupling regulation strategy library contains multiple predefined regulation strategies. Each regulation strategy is associated with a specific set of energy supply and demand imbalance scenarios and corresponding equipment collaborative scheduling schemes. S42. When the system operation status assessment data indicates a risk of energy supply and demand imbalance, the short-term energy supply and demand forecast data is matched with the energy supply and demand imbalance scenario corresponding to each regulation strategy in the preset multi-energy coupling regulation strategy library. S43. Search for the adjustment strategy that best matches the short-term energy supply and demand forecast data, and determine it as the target adjustment strategy data.

6. The integrated energy dynamic adjustment method in a multi-energy coupling scenario according to claim 5, characterized in that: The steps involved in generating device collaborative scheduling instruction data in S5 are as follows: S51. Analyze the target adjustment strategy data to obtain the required equipment collaborative scheduling scheme. The equipment collaborative scheduling scheme includes the power setpoints and start / stop status of phase change thermal storage devices, phase change cold storage devices, electrochemical energy storage devices, air source heat pumps, photovoltaic power generation equipment and wind power generation equipment. S52. Based on the device collaborative scheduling scheme, generate the device collaborative scheduling instruction data, which includes power adjustment instructions, charging and discharging instructions and start / stop instructions for each device.

7. The integrated energy dynamic adjustment method in a multi-energy coupling scenario according to claim 6, characterized in that: The dynamic energy adjustment process in step S6, and the simultaneous acquisition of the dynamically adjusted system operating status data, includes the following steps: S61. The device collaborative scheduling instruction data is sent to the controller of each execution terminal through the Internet of Things communication protocol; S62. The controllers of each execution terminal control the corresponding distributed energy storage devices, renewable energy power generation devices and controllable loads to perform actions according to the received instructions, so as to realize the dynamic regulation of energy. The dynamic regulation includes peak shaving and valley filling, promoting the local consumption of renewable energy and multi-energy complementarity. S63. During the dynamic adjustment process, the system operation status data after dynamic adjustment is acquired and recorded in real time as the basis for closed-loop feedback optimization. The system operation status data includes the real-time power of each device, the state of charge of the energy storage device, and the actual balance between energy supply and demand.

8. The integrated energy dynamic adjustment method in a multi-energy coupling scenario according to claim 7, characterized in that: Updating the preset multi-energy coupling adjustment strategy library in S7 includes the following steps: S71. Based on the dynamically adjusted system operating status data, monitor the actual supply and demand balance of the system and the equipment operating status in real time. S72. Compare and analyze the monitored system operation status data with the short-term energy supply and demand forecast data, and calculate the adjustment deviation; S73. When the adjustment deviation exceeds a preset deviation threshold, closed-loop feedback optimization is triggered. Using a reinforcement learning model, the preset multi-energy coupling adjustment strategy library is updated online according to the adjustment deviation to optimize the subsequent adjustment strategy matching process.

9. A comprehensive energy dynamic regulation system for multi-energy coupling scenarios, used to implement the comprehensive energy dynamic regulation method for multi-energy coupling scenarios as described in any one of claims 1-8, characterized in that: The system includes a data acquisition and sensing module, an energy prediction and analysis module, a strategy matching and decision-making module, a collaborative scheduling and control module, and a feedback optimization and learning module.

10. A comprehensive energy dynamic regulation system for multi-energy coupling scenarios according to claim 9, characterized in that: The system specifically includes: The data acquisition and sensing module collects multi-energy load data through intelligent sensor networks and multi-energy meters, receives meteorological forecast data through meteorological service interfaces, receives real-time energy price data through energy trading platform interfaces, and outputs the above three types of data to the energy forecasting and analysis module. The energy forecasting and analysis module receives the multi-energy load data, meteorological forecast data and real-time energy price data, performs energy supply and demand forecasting processing through the time-series forecasting unit, and performs operational status assessment processing through the status assessment unit, and outputs short-term energy supply and demand forecast data and system operational status assessment data. The strategy matching and decision-making module receives the short-term energy supply and demand forecast data and the system operation status assessment data, performs matching processing with the preset multi-energy coupling regulation strategy library through the strategy matching unit, and outputs the target regulation strategy data. The collaborative scheduling and control module receives the target adjustment strategy data, generates equipment collaborative scheduling instruction data through the instruction generation unit, sends it to the execution terminal through the communication interface unit, and obtains the dynamically adjusted system operation status data. The feedback optimization and learning module receives the dynamically adjusted system operation status data, compares and analyzes it with the short-term energy supply and demand forecast data through the deviation calculation unit, and updates the preset multi-energy coupling regulation strategy library online through the strategy update unit.