Digital twinborn simulation control method for multi-energy-flow cooperation of source-network-load-storage integrated park

Abstract

The invention belongs to the technical field of simulation control, and particularly discloses a source-network-load-storage integrated park multi-energy-flow collaborative digital twinborn simulation control method, which comprises the following steps: obtaining real-time operation state data of a target park; performing hierarchical virtual mapping on the target park, and constructing a cooperative control digital twinborn body of the target park; performing multi-constraint coordination planning on the source network load storage resources of the target park to obtain a preliminary coordination control strategy of the target park; performing prospective deduction simulation on the preliminary cooperative control strategy, and identifying a safety out-of-limit event and a strategy conflict point in a simulation environment after simulation; performing simulation feedback optimization on the preliminary cooperative control strategy to obtain an optimization control strategy of the target park; and analyzing the optimization control strategy into a cooperative control instruction of the target park, and transmitting the cooperative control instruction to a control terminal of the target park. According to the invention, precision, security and high efficiency of multi-energy flow cooperative control of the source-network-load-storage integrated park can be realized.

Description

Technical Field

[0001] This invention belongs to the field of simulation control technology, and in particular relates to a digital twin simulation control method for multi-energy flow coordination in integrated power generation, grid, load and storage parks. Background Technology

[0002] Integrated energy-grid-load-storage parks combine various resources, including source-side power generation equipment, grid-side transmission facilities, load-side power terminals, and energy storage-side regulation devices. They serve as a core platform for achieving cascaded energy utilization and improving renewable energy consumption. Currently, the industry is gradually combining digital twin technology with multi-energy flow collaborative control technology. By collecting operational data from multiple devices, a virtual model of the park is constructed, and collaborative scheduling strategies are being developed to address the complex operating conditions and multi-dimensional control requirements related to economic efficiency, energy efficiency, and renewable energy consumption within the park.

[0003] However, existing control technologies still have many shortcomings in practical applications, making it difficult to support efficient and accurate multi-energy flow collaborative control. On the one hand, the diverse sources and heterogeneous formats of data in the park make it difficult for existing methods to efficiently complete communication protocol parsing, timing synchronization, and redundant data removal, resulting in insufficient accuracy and completeness of real-time operating status data, which cannot provide reliable support for collaborative decision-making. On the other hand, existing virtual mapping methods lack hierarchical design and fail to fully integrate the static topology characteristics and dynamic operating features of the park, making it difficult to accurately replicate the actual operating status of the park in the constructed digital twin. At the same time, in the control strategy formulation stage, existing multi-constraint coordination planning does not comprehensively analyze the adjustability of source, network, load, and storage resources, and lacks a forward-looking extrapolation and verification mechanism. The feedback optimization process lacks a quantitative calculation model, making it prone to safety limit violations and strategy conflicts, and unable to adapt to the comprehensive control objectives of the park. Summary of the Invention

[0004] The purpose of this invention is to provide a digital twin simulation control method for multi-energy flow coordination in integrated power generation, grid, load and storage parks. Through multi-source heterogeneous data fusion, hierarchical collaborative control digital twin construction, strategy forward-looking deduction and quantitative feedback optimization, the method achieves precise, safe and efficient multi-energy flow coordinated control in integrated power generation, grid, load and storage parks.

[0005] To achieve the above objectives, the technical solution adopted by this invention is a digital twin simulation control method for multi-energy flow coordination in an integrated source-grid-load-storage park, comprising the following steps: S1. Perform multi-source heterogeneous data fusion on the controllable operation setting parameters and status monitoring data of the target park to obtain the real-time operation status data of the target park; S2. Based on the network topology of the target park, perform hierarchical virtual mapping of the target park to construct a collaborative control digital twin of the target park; S3. Based on the collaborative control digital twin and real-time operation status data, multi-constraint coordination planning is carried out on the source-grid-load-storage resources of the target park to obtain the preliminary collaborative control strategy of the target park. S4. Conduct a forward-looking simulation of the preliminary collaborative control strategy to identify safety limit violations and strategy conflict points in the simulation environment after the simulation. S5. Based on safety violation events and policy conflict points, perform simulation feedback optimization on the preliminary collaborative control strategy to obtain the optimized control strategy for the target park. S6. Parse the optimized control strategy into collaborative control instructions for the target park and send the collaborative control instructions to the control terminal of the target park.

[0006] Preferably, in S1, the process of obtaining the real-time operating status data of the target park is as follows: Collect source-side equipment output plan parameters, grid-side topology connection parameters, load-side real-time power data, and energy storage equipment charging and discharging status parameters of the target park; The communication protocol is parsed to obtain multi-source data of the target park by analyzing the power output planning parameters of the source-side equipment, the topology connection parameters of the network side, the real-time power data of the load side, and the charging and discharging status parameters of the energy storage equipment. Based on a unified time series benchmark, the multi-source data is formatted and unified to obtain the synchronous multi-source data of the target park; Redundant data is removed from the synchronized multi-source data to obtain the real-time operational status data of the target park.

[0007] Preferably, in S2, the process of constructing a collaborative control digital twin of the target park is as follows: Analyze the network topology of the target campus and identify the power nodes, load nodes, energy storage nodes, and network branches within the target campus; A static topology layer for the target park is constructed using power nodes, load nodes, and energy storage nodes as static network nodes and network branches as relationships. By dynamically associating the nodes in the static topology layer with their characteristics, the dynamic behavior layer of the target park is obtained. Based on the static topology layer and dynamic behavior layer, and combined with the control objectives and security constraints of the target park, a collaborative control digital twin of the target park is constructed.

[0008] Preferably, by combining the control objectives and security constraints of the target park, a collaborative control digital twin of the target park is constructed, specifically as follows: The control objectives of the target park are analyzed by targeted feature deconstruction to obtain the economic indicators, energy efficiency indicators and renewable energy consumption indicators of the target park; Based on economic indicators, energy efficiency indicators, and renewable energy consumption indicators, the coordinated control mechanism of the target industrial park is analyzed to obtain the coordinated decision-making logic of the target industrial park. The security constraints of the target area are transformed into constraint rules to obtain the security verification logic of the target area. The coordination decision-making logic and security verification logic are functionally integrated to obtain the decision-making logic layer of the target park; The static topology layer, dynamic behavior layer, and decision logic layer are encapsulated and integrated in a programmatic manner to obtain a collaborative control digital twin of the target park.

[0009] Preferably, in S3, the process of obtaining the preliminary collaborative control strategy for the target park is as follows: Real-time operating status data is injected into the collaborative control digital twin to obtain the planning initial boundary conditions of the collaborative control digital twin; Based on the initial boundary conditions of the plan, the constraint range of the source-grid-load-storage resources of the target park is analyzed to obtain the set of resource adjustability and the set of operational constraints of the target park. Based on the set of adjustable resource capabilities and the set of operational constraints, an integrated coordinated decision-making process is made for source-grid-load-storage resources to obtain a preliminary collaborative control strategy for the target park.

[0010] Preferably, an integrated coordinated decision-making process is implemented for source-grid-load-storage resources to obtain a preliminary collaborative control strategy for the target park, specifically as follows: Based on the resource adjustability set and the operational constraint set, the operable boundary of source-grid-load-storage resources is traced to obtain the collaborative scheduling feasible domain of the target park; The feasible domain of coordinated scheduling is used as a constraint and input into the decision logic layer of the coordinated control digital twin to generate a preliminary coordinated instruction sequence for the target park. The initial coordination instruction sequence is encapsulated in a strategy format to obtain the initial collaborative control strategy for the target park.

[0011] Preferably, in S4, the process of performing a forward-looking simulation of the preliminary coordinated control strategy is as follows: Within the collaborative control digital twin, the initial instructions of the preliminary collaborative control strategy are parsed out, and the initial instructions are sequentially ordered to obtain the initial executable control instruction sequence of the collaborative control digital twin. According to the preset simulation step size, the collaborative control digital twin is driven to perform step-by-step simulation using a sequence of executable control instructions; During the step-by-step simulation, the state parameters of nodes and devices in the collaborative control digital twin are updated based on the device response and network coupling relationship of the collaborative control digital twin. After all the instructions in the executable control instruction sequence have been executed, the updated state parameters are integrated to obtain the full-cycle state evolution sequence of the cooperative control digital twin.

[0012] Preferably, in S4, safety violation events and policy conflict points in the post-simulation environment are identified, specifically as follows: Based on the full-cycle state evolution sequence, the safety boundary compliance of key operating parameters in the collaborative control digital twin is detected, and events in which key operating parameters continuously exceed the preset safe operating range are identified. Events that exceed the preset safe operating range will be treated as safety violation events in the post-simulation environment. In the forward-looking simulation process, a logical consistency analysis is performed on the preliminary coordinated control strategy to obtain potential policy conflict points of the preliminary coordinated control strategy. The potential strategic conflict points are analyzed and the final judgment is made to obtain the strategic conflict points in the post-simulation environment.

[0013] Preferably, in S5, the process of obtaining the optimized control strategy for the target park is as follows: Based on the safety limit violation events and policy conflict points, abnormal parameters of the preliminary collaborative control strategy are traced to obtain the control parameters to be adjusted in the preliminary collaborative control strategy. A multi-objective trade-off tuning calculation is performed on the control parameter to be adjusted to obtain the adjusted control parameter. The calculation formula for the adjusted control parameter is as follows: ; In the formula, This represents the adjusted control parameter vector. This represents the initial control parameter vector consisting of the control parameters to be adjusted. This represents the safety limit deviation vector. Represents the target conflict difference vector. Represents the security sensitivity matrix. Represents the conflict reconciliation matrix. Indicates the weight of the security limit violation dimension. Indicates the weight of the conflict reconciliation dimension; Based on the adjusted control parameters, the preliminary collaborative control strategy is synchronously corrected and reconstructed to obtain the corrected collaborative control strategy for the target park. The revised collaborative control strategy is subjected to closed-loop verification and simulation. When the closed-loop verification and simulation confirms that the safety limit violation event and the conflict point with the strategy have been eliminated, the revised collaborative control strategy is output as the optimized control strategy for the target park.

[0014] Preferably, in step S6, the optimized control strategy is parsed into collaborative control instructions for the target area, and these instructions are then transmitted to the control terminal of the target area. Specifically: The timing parameters of the target park are obtained by analyzing the timing parameters of the optimized control strategy. The timing control parameters are encoded into instructions to obtain the standard control instructions for the target park. The standard control command set is sent to the control terminal of the target park through the communication interface of the collaborative control digital twin.

[0015] The present invention has the following beneficial effects: This invention improves the accuracy and completeness of real-time operational status data by performing protocol parsing, time synchronization, and redundancy removal on multi-source heterogeneous data from the target park, providing reliable support for collaborative control. It also employs hierarchical virtual mapping to construct a collaborative control digital twin that integrates static topology, dynamic behavior, and decision-making logic, accurately replicating the park's operational characteristics and ensuring the scientific nature of control strategy formulation.

[0016] This invention generates a preliminary strategy through multi-constraint coordination planning, identifies safety limit violation events and strategy conflict points through forward-looking extrapolation, and achieves precise tuning of control parameters by combining quantitative feedback optimization, thereby improving the safety and adaptability of the control strategy. Then, through time series analysis and efficient command encoding, the optimized strategy is issued to ensure the precise implementation of multi-energy flow collaborative control and improve the park's operational efficiency and overall benefits. Attached Figure Description

[0017] Figure 1 This is a schematic flowchart of the method of the present invention; Figure 2 This is a comparison curve of the system voltage before and after optimization in a certain park in Embodiment 1 of the present invention. Detailed Implementation

[0018] Example 1: As Figure 1 As shown, the digital twin simulation control method for multi-energy flow coordination in an integrated source-grid-load-storage park includes the following steps: S1. Perform multi-source heterogeneous data fusion on the controllable operation setting parameters and status monitoring data (key status monitoring data) of the target park to obtain the real-time operation status data of the target park; S2. Based on the network topology of the target park, perform hierarchical virtual mapping of the target park to construct a collaborative control digital twin of the target park; S3. Based on the collaborative control digital twin and real-time operation status data, multi-constraint coordination planning is carried out on the source-grid-load-storage resources of the target park to obtain the preliminary collaborative control strategy of the target park. S4. Conduct a forward-looking simulation of the preliminary collaborative control strategy to identify safety limit violations and strategy conflict points in the simulation environment after the simulation. S5. Based on safety violation events and policy conflict points, perform simulation feedback optimization on the preliminary collaborative control strategy to obtain the optimized control strategy for the target park. S6. Parse the optimized control strategy into collaborative control instructions for the target park and send the collaborative control instructions to the control terminal of the target park.

[0019] In S1, the real-time operational status data of the target park is obtained, including: Collect source-side equipment output plan parameters, grid-side topology connection parameters, load-side real-time power data, and energy storage equipment charging and discharging status parameters of the target park; The communication protocol is parsed to obtain multi-source data of the target park by analyzing the power output planning parameters of the source-side equipment, the topology connection parameters of the network side, the real-time power data of the load side, and the charging and discharging status parameters of the energy storage equipment. Based on a unified time series benchmark, the multi-source data is formatted and unified to obtain the synchronous multi-source data of the target park; Redundant data is removed from the synchronized multi-source data to obtain the real-time operational status data of the target park.

[0020] By deploying dedicated data acquisition equipment within the target park, the system directly acquires source-side equipment output planning parameters, network-side topology connection parameters, real-time load-side power data, and energy storage device charging and discharging status parameters. Source-side equipment output planning parameters are retrieved from the local control units of all source-side equipment within the park, including photovoltaic power stations, wind farms, and gas turbines. The retrieved information covers the planned output values ​​of the equipment for the next 24 hours, broken down by hour.

[0021] The network-side topology connection parameters were determined by conducting on-site verification of the park's power network facilities and combining the park's power system design blueprint. The physical connection methods between network-side components such as transformers, transmission lines, and switching equipment were recorded one by one, and the names and locations of other components connected to the input and output ends of each component were clarified.

[0022] Real-time power data on the load side is obtained by installing power sensors in the power supply circuits of all load equipment, including industrial production equipment, office equipment, and residential equipment, within the park. The sensors capture and record the power consumption of the load equipment in real time at a frequency of once per second.

[0023] The charging and discharging status parameters of the energy storage equipment are transmitted via a wired data transmission link established with the monitoring modules of all energy storage battery packs, energy storage converters and other energy storage equipment in the park. The current working status identifier of the equipment is read in real time, and the duration of the status is recorded. The working status identifier includes only three types: charging, discharging and standby.

[0024] The collected source-side equipment output plan parameters, grid-side topology connection parameters, load-side real-time power data, and energy storage device charging and discharging status parameters are parsed and processed using corresponding data communication protocols to obtain multi-source data.

[0025] For the power output plan parameters of the source-side equipment, the Modbus communication protocol is used for parsing. According to the data frame structure specified by the protocol, the equipment address function code and data segment check code are extracted from the transmitted raw data. Then, based on the meaning of the function code, the power output plan information contained in the data segment is interpreted.

[0026] For the network-side topology connection parameters, the IEC61850 communication protocol is used for parsing. Based on the object-oriented modeling specifications of this protocol, the original data is mapped into standard logical nodes and data objects, from which the connection relationship information of the network-side components is extracted.

[0027] For real-time power data on the load side, the TCP / IP communication protocol is used for parsing. By establishing a TCP connection between the client and the sensor data transmission module, the data packets sent by the sensor are received, and the data is split according to the data packet format specified by the IP protocol to extract the power value information.

[0028] For the charging and discharging status parameters of energy storage devices, the CAN bus communication protocol is used for parsing. According to the frame format of the CAN protocol, the identifiers and data fields in the data frames are identified. The working status identifier and duration information of the energy storage device are interpreted from the data field. All the effective information obtained after parsing is classified and integrated to form multi-source data of the target park.

[0029] Beijing time is set as the unified time series benchmark, with milliseconds as the unit of time measurement. A unique timestamp is added to each data point in the multi-source dataset, with the timestamp value accurate to the number of milliseconds in Beijing time corresponding to the moment the data parsing is completed. A unified data format standard is established, where numerical data is uniformly represented in decimal, with the integer part retained to the units digit and the decimal part retained to two decimal places. Text data is uniformly encoded in UTF-8 format, with a character length not exceeding 256 characters.

[0030] According to the data format standard, all data from the multi-source data are converted to ensure that the format of each data item is completely consistent. Then, using timestamps as indexes, the source-side equipment output plan data, network-side topology connection data, load-side real-time power data, and energy storage device charging and discharging status data corresponding to the same timestamp are associated and combined to form a dataset with time series as the framework. This dataset is the synchronous multi-source data of the target park.

[0031] A pre-defined rule for determining redundant data is established: when two data entries have identical timestamps and every field of their content is identical, the data entry that appears after the rule is determined to be redundant.

[0032] Simultaneously, invalid data judgment rules are established. The value range of the power output plan parameters of the source-side equipment is limited to 0 to the rated output value of the equipment; data exceeding this range is judged as invalid data. The content of the network-side topology connection relationship parameters must be completely consistent with the connection relationships recorded in the park's power system design blueprint; inconsistencies are judged as invalid data. The value range of the load-side real-time power data is limited to 0 to the rated power value of the load equipment; data exceeding this range is judged as invalid data. The content of the energy storage device's charging and discharging status parameters can only be one of charging, discharging, or standby; any other content is judged as invalid data. Synchronous multi-source data is checked line by line, and redundant and invalid data are identified according to the above rules. This data is then completely removed from the synchronous multi-source data, and the remaining filtered dataset is the real-time operating status data of the target park.

[0033] By employing comprehensive and targeted data collection methods, we ensured the complete acquisition of source-side equipment output planning parameters, network-side topology connection parameters, load-side real-time power data, and energy storage device charging and discharging status parameters, providing comprehensive foundational data for subsequent data processing. By using communication protocols corresponding to various parameters for parsing, we ensured the effective interpretation of data from different sources, avoiding data loss or errors due to protocol incompatibility and ensuring the accuracy of multi-source data. Through unified time series benchmarks and data formats, we achieved synchronous integration of multi-source data, enabling different types of data to be systematically correlated according to timeline, eliminating barriers caused by differences in time series and format.

[0034] By employing clear rules for identifying redundant and invalid data, meaningless data was eliminated, ensuring the purity and validity of real-time operational status data. This allows the data to accurately and comprehensively reflect the actual operational status of the target park, providing reliable and high-quality data support for the subsequent construction of the collaborative control digital twin and the formulation of collaborative control strategies, thus ensuring the smooth progress of the entire digital twin simulation control method.

[0035] In S2, a collaborative control digital twin of the target park is constructed, including: Analyze the network topology of the target campus and identify the power nodes, load nodes, energy storage nodes, and network branches within the target campus; A static topology layer for the target park is constructed using power nodes, load nodes, and energy storage nodes as static network nodes and network branches as relationships. By dynamically associating the nodes in the static topology layer with their characteristics, the dynamic behavior layer of the target park is obtained. Based on the static topology layer and dynamic behavior layer, and combining the control objectives and security constraints of the target park, a collaborative control digital twin of the target park is constructed, specifically as follows: The control objectives of the target park are analyzed by targeted feature deconstruction to obtain the economic indicators, energy efficiency indicators and renewable energy consumption indicators of the target park; Based on economic indicators, energy efficiency indicators, and renewable energy consumption indicators, the coordinated control mechanism of the target industrial park is analyzed to obtain the coordinated decision-making logic of the target industrial park. The security constraints of the target area are transformed into constraint rules to obtain the security verification logic of the target area. The coordination decision-making logic and security verification logic are functionally integrated to obtain the decision-making logic layer of the target park; The static topology layer, dynamic behavior layer, and decision logic layer are encapsulated and integrated in a programmatic manner to obtain a collaborative control digital twin of the target park.

[0036] By reviewing the detailed design documents and equipment installation drawings of the target park's power system, and combining this with on-site surveys and verification, a comprehensive analysis of the target park's network topology was conducted. Based on equipment function and energy flow, the following was identified: power nodes are the locations of photovoltaic power stations, wind farms, gas turbines, and other equipment capable of generating electricity within the park; load nodes are the areas where terminal equipment consuming electricity, such as industrial production equipment, commercial office equipment, and residential electrical equipment, is located; energy storage nodes are the installation areas of energy storage equipment such as energy storage battery packs and energy storage converters; and network branches are the power transmission and conversion facilities connecting each node, such as transmission lines, transformers, and switching equipment. The specific location, equipment model, and network branch length, rated capacity, and connection method of each node were recorded to ensure accurate identification of all power nodes, load nodes, energy storage nodes, and network branches within the target park.

[0037] Using graphical modeling tools, the identified power nodes, load nodes, and energy storage nodes are treated as independent static network node modeling units. Each unit includes static attribute information such as node name, equipment composition, and physical location. Using network branches as connecting links, and based on the actual connection relationships determined by on-site surveys and design documents, corresponding connections between each static network node are established in the modeling environment. For example, power nodes are connected to transformer branches via specifically numbered transmission line branches, and then from the transformer branches to the load nodes and energy storage nodes. This fully recreates the physical connection architecture between each node, ultimately forming a static topology layer that is completely consistent with the actual network topology of the target park.

[0038] Technical parameter manuals, factory test reports, and historical operating data for each device corresponding to each node in the static topology layer are collected. Dynamic operating characteristic parameters for each node are extracted, including the output response delay time and maximum output adjustment rate for power nodes, the power fluctuation cycle and maximum load change for load nodes, and the charging and discharging efficiency and charging and discharging power adjustment range for energy storage nodes. By establishing a mapping relationship between node static attributes and dynamic characteristic parameters—for example, linking the rated power parameters of power nodes to the dynamic law of actual output changes with sunlight and wind speed, and linking the equipment type of load nodes to the dynamic characteristics of power consumption at different times—these relationships are integrated into the static topology layer, enabling static nodes to simulate dynamic operating states, thus forming a dynamic behavior layer for the target park.

[0039] After clarifying the core control objectives of the target park, a target decomposition method is used to deconstruct its targeted characteristics. Economic indicators focus on optimizing the overall operating cost of the park, specifically encompassing quantifiable cost components such as source-side equipment operation and maintenance costs, grid purchase costs, and energy storage equipment depreciation costs, with the total annual operating cost as the core evaluation criterion. Energy efficiency indicators focus on maximizing comprehensive energy utilization efficiency, reflected in the conversion efficiency, transmission efficiency, and end-use efficiency of multiple energy forms such as electricity and heat within the park, calculated as the weighted average of the efficiency of each energy link as the energy efficiency evaluation standard. Renewable energy consumption indicators aim to improve the utilization rate of renewable energy, explicitly setting a threshold of 80% for renewable energy power generation in the park's total electricity consumption, using this as the core evaluation standard. Through the above deconstruction process, the economic, energy efficiency, and renewable energy consumption indicators of the target park are accurately obtained.

[0040] Based on the three core indicators obtained from the deconstruction, this paper analyzes the interrelationships and collaborative optimization paths among these indicators, and elucidates the collaborative control mechanism of the target industrial park. When renewable energy generation is sufficient, priority is given to using renewable energy to reduce grid power purchases and lower operating costs. Simultaneously, energy storage devices are adjusted for charging to improve renewable energy absorption. When load demand is high, the discharge of energy storage devices and the expansion of conventional power sources are coordinated to ensure stable power supply while avoiding energy waste and improving energy efficiency. These control scenarios and corresponding operating rules are systematically reviewed, clarifying the priority and control measures of each indicator under different operating conditions, forming a complete set of coordinated decision-making logic to guide the generation of subsequent control strategies.

[0041] We identified various safety constraints for the target industrial park's power operation, including node voltage constraints, system frequency constraints, network branch transmission power constraints, and equipment operating temperature constraints. Each safety constraint was then transformed into a clear and verifiable rule. For example, the node voltage constraint was transformed into "the operating voltage of all power supply nodes and load nodes must be maintained within ±5% of the rated voltage of 380V," the system frequency constraint into "the operating frequency of the park's power system must be stable within the range of 50Hz ±0.2Hz," and the network branch transmission power constraint into "the actual transmission power of each transmission line must not exceed its rated transmission power of 1000kW." Through this rule transformation, abstract safety constraints were converted into concrete and executable safety verification logic.

[0042] A logic fusion algorithm is employed to integrate coordination decision-making logic and security verification logic. Security verification logic is set as a pre-judgment condition; every control command generated by the coordination decision-making logic must first be verified by the security verification logic. If the control command does not violate any security constraints after execution, it is a valid command; if it violates a security constraint, the feedback is sent to the coordination decision-making logic for command adjustment until a control command that meets security requirements is generated. Through this fusion approach of pre-verification and dynamic feedback, the two logics are organically combined to form a decision logic layer that integrates control decision-making and security assurance functions.

[0043] Modular encapsulation technology is employed to encapsulate the static topology layer, dynamic behavior layer, and decision logic layer into independent functional modules, with standardized data interaction interfaces designed for each module. The static topology layer module provides the dynamic behavior layer module with basic node information and connection relationship data. Based on this basic data, the dynamic behavior layer module outputs dynamic node operating status data. The decision logic layer module receives the output data from the dynamic behavior layer module, combines it with its own logic to generate control commands, and feeds them back to the first two modules. These three modules are integrated through interfaces to construct a unified simulation model capable of comprehensively replicating the target campus's network structure, dynamic operating characteristics, and decision control mechanisms—essentially a collaborative control digital twin of the target campus.

[0044] By employing a step-by-step analysis, layered construction, and logical fusion approach, the network topology and operational characteristics of the target campus were accurately replicated. The static topology layer ensured the consistency of the physical structure, the dynamic behavior layer achieved dynamic simulation of the operational state, and the decision logic layer balanced the achievement of control objectives with the satisfaction of safety constraints. The constructed collaborative control digital twin accurately reflects the actual operational state of the target campus, providing a high-precision virtual simulation platform for subsequent multi-constraint coordination planning of source-network-load-storage resources, simulation and optimization of control strategies. This ensures the scientific validity and feasibility of subsequent control strategy formulation. Furthermore, the standardized encapsulation and integration methods enhance the stability and scalability of the digital twin, guaranteeing the efficient advancement of the entire digital twin simulation control method.

[0045] In S3, the preliminary collaborative control strategy for the target park is obtained, including: Real-time operating status data is injected into the collaborative control digital twin to obtain the planning initial boundary conditions of the collaborative control digital twin; Based on the initial boundary conditions of the plan, the constraint range of the source-grid-load-storage resources of the target park is analyzed to obtain the set of resource adjustability and the set of operational constraints of the target park. Based on the set of adjustable resource capabilities and the set of operational constraints, an integrated coordinated decision-making process is performed on the source-grid-load-storage resources to obtain a preliminary collaborative control strategy for the target park, specifically: Based on the resource adjustability set and the operational constraint set, the operable boundary of source-grid-load-storage resources is traced to obtain the collaborative scheduling feasible domain of the target park; The feasible domain of coordinated scheduling is used as a constraint and input into the decision logic layer of the coordinated control digital twin to generate a preliminary coordinated instruction sequence for the target park. The initial coordination instruction sequence is encapsulated in a strategy format to obtain the initial collaborative control strategy for the target park.

[0046] Real-time operational status data is injected through a dedicated data input interface of the collaborative control digital twin. This data includes source-side equipment output plan data, network-side topology connection data, load-side real-time power data, and energy storage device charging and discharging status data. During the injection process, data is matched one-to-one with the power nodes, load nodes, energy storage nodes, and network branches in the static topology layer of the collaborative control digital twin. For example, source-side equipment output plan data is associated with the corresponding power nodes, and load-side real-time power data is associated with the corresponding load nodes.

[0047] The data injection trigger mechanism is set to execute once every 5 minutes to ensure that the injected data is synchronized with the actual operating status of the target park. After the injection is completed, the collaborative control digital twin automatically records the current status parameters, data associations and timestamp information of each node and device. These information together constitute the planning start boundary conditions of the collaborative control digital twin and clarify the initial state benchmark for subsequent coordinated planning.

[0048] Based on the status parameters of each node and equipment recorded in the initial boundary conditions of the plan, the constraint range of source, grid, load and storage resources in the target park is analyzed one by one. For source-side resources, based on the rated power of the equipment, historical operating data, and current output status, the maximum output value, minimum output value, and output adjustment rate range of each source-side equipment are determined, forming source-side resource adjustability data (resource adjustability set). For network-side resources, based on the rated transmission capacity and allowable voltage fluctuation range of the network branches, the upper limit of transmission power and voltage control range of each network branch are determined, forming network-side resource adjustability data. For load-side resources, combined with the technical parameters and operating requirements of the load equipment, load types that can be reduced, load types that can be transferred, and non-adjustable load types are classified, and the maximum reduction amount and transfer time window of the adjustable load are determined, forming load-side resource adjustability data. For energy storage resources, based on the remaining capacity, charging and discharging efficiency, and rated charging and discharging power of the energy storage equipment, the maximum charging power, maximum discharging power, upper limit of charging time, and upper limit of discharging time of the energy storage equipment are determined, forming energy storage resource adjustability data. The above-mentioned resource adjustability data from all sides are integrated to obtain the resource adjustability set of the target park.

[0049] Meanwhile, the physical constraints and rules of the source-grid-load-storage resource operation process are sorted out. The constraints on the source side equipment operation include the continuous operation time of the equipment not exceeding 72 hours and the start-stop interval not less than 30 minutes. The constraints on the grid side operation include the node voltage being maintained at 380V±5% and the transmission power of the network branch not exceeding the rated capacity. The constraints on the load side operation include the power supply interruption time of important loads not exceeding 10 seconds. The constraints on the energy storage equipment operation include the cumulative number of charge-discharge cycles not exceeding the design life and the operating temperature being maintained between 0℃ and 45℃. These constraints are compiled into the set of operation constraint conditions for the target park.

[0050] Based on the resource adjustability set and the operational constraint set, the boundary traversal method is used to trace the operational boundary of source-grid-load-storage resources.

[0051] For source-side resources, based on the maximum and minimum output values ​​of the resource adjustability set, and combined with the equipment start-stop interval and continuous operation duration limit of the operation constraint set, the feasible output range of source-side equipment in different time periods is determined; For network-side resources, based on the transmission power range and voltage control range of the resource adjustability concentration, and combined with the safe operation requirements of the operation constraint concentration, determine the feasible range of transmission power and node voltage of network branches; For load-side resources, based on the load reduction capacity and transfer time window of the concentrated resource adjustability, and combined with the power supply requirements of important loads under concentrated operational constraints, the feasible methods and scope of load adjustment are determined. For energy storage resources, based on the charging and discharging power and duration limits of the concentrated resource adjustability, and combined with the temperature and cycle number requirements of the concentrated operational constraints, the feasible power range and time range for charging and discharging of energy storage equipment are determined.

[0052] By cross-comparing the feasible ranges of resources on each side, the common feasible range under all constraints is taken to form the collaborative scheduling feasible domain of the target park. This feasible domain clarifies the scheduling range of source, grid, load and storage resources under the premise of meeting all constraints.

[0053] The feasible domain for coordinated scheduling is input to the decision logic layer in the coordinated control digital twin via a data interface. The decision logic layer pre-integrates a coordinated control mechanism based on economic indicators, energy efficiency indicators, and renewable energy consumption indicators. The decision logic layer first determines the current operating scenario of the target park. The scenario is divided based on the proportion of renewable energy output: a proportion higher than 60% is a high consumption scenario, a proportion between 30% and 60% is a balanced scenario, and a proportion lower than 30% is a low consumption scenario.

[0054] In high-consumption scenarios, the decision-making logic layer prioritizes renewable energy consumption, generating instructions including controlling energy storage devices to charge at maximum feasible power, reducing conventional source-side equipment to minimum feasible output, and prioritizing renewable energy power supply for load-side equipment. In balanced scenarios, with energy efficiency as the core, generated instructions include optimizing network branch transmission paths to reduce losses, adjusting the charging and discharging of energy storage devices according to load fluctuations, and matching the output of source-side equipment to load demand. In low-consumption scenarios, with optimal economic efficiency as the goal, generated instructions include conventional source-side equipment operating at the lowest possible cost, charging energy storage devices during off-peak hours and discharging during peak hours, and adjusting transferable loads to operate during off-peak hours. These instructions are arranged chronologically, with each instruction specifying the executing entity, action, time, and parameter range, forming a preliminary coordination instruction sequence for the target park.

[0055] A unified preliminary collaborative control strategy format standard was established, comprising six modules: strategy identifier, generation time, effective duration, applicable scenarios, instruction sequence list, and effective conditions. The strategy identifier is a unique 16-digit code consisting of year, month, date, and sequence number; the generation time is accurate to the second; the effective duration is set to 24 hours, calculated from the generation time; the applicable scenarios clearly correspond to one or more of three scenarios: high absorption, balanced, and low absorption; the instruction sequence list lists all instructions in the preliminary coordination instruction sequence in chronological order, with each instruction including the execution object ID, action type, execution time node, parameter range, and execution result verification standard; the effective conditions are set as follows: the total load rate of the park is between 60% and 90%, there are no major equipment maintenance plans, and the fluctuation of renewable energy output does not exceed 20%.

[0056] According to the standard, the preliminary coordination instruction sequence is organized and supplemented, and the adjustment trigger condition is clarified as follows: when the real-time monitoring shows that the fluctuation of renewable energy output exceeds 20%, the load change rate exceeds 15%, or the equipment operating status is abnormal, the strategy adjustment process is automatically triggered to ensure that the packaged strategy information is complete, structurally standardized, and highly executable, and finally the preliminary collaborative control strategy of the target park is obtained.

[0057] By accurately injecting real-time operational status data into the digital twin, the initial boundary conditions of the planning were ensured to be highly consistent with the actual operational status of the park, providing a reliable initial benchmark for subsequent coordinated planning. A comprehensive analysis of the constraints on source-grid-load-storage resources was conducted, clearly defining the resource adjustability and operational constraints, thus avoiding strategy infeasibility issues caused by information gaps during the planning process.

[0058] By tracing the operational boundary, the feasible domain for coordinated scheduling is clearly defined, establishing a safe and reasonable scope for scheduling decisions and ensuring the feasibility of the strategy. The preliminary coordination instruction sequence generated based on the decision logic layer is highly targeted, meeting the needs of different operational scenarios in the park while balancing economic efficiency, energy efficiency, and renewable energy consumption targets. The strategy is encapsulated using a standardized format, resulting in a clear and complete structure for the preliminary coordinated control strategy. This facilitates subsequent forward-looking simulations and feedback optimization, laying a solid foundation for the entire coordinated control process and enhancing the scientific rigor and efficiency of the source-grid-load-storage resource coordination planning.

[0059] In S4, a forward-looking simulation of the preliminary coordinated control strategy is conducted, including: Within the collaborative control digital twin, the initial instructions of the preliminary collaborative control strategy are parsed out, and the initial instructions are sequentially ordered to obtain the initial executable control instruction sequence of the collaborative control digital twin. According to the preset simulation step size, the collaborative control digital twin is driven to perform step-by-step simulation using a sequence of executable control instructions; During the step-by-step simulation, the state parameters of nodes and devices in the collaborative control digital twin are updated based on the device response and network coupling relationship of the collaborative control digital twin. After all the instructions in the executable control instruction sequence have been executed, the updated state parameters are integrated to obtain the full-cycle state evolution sequence of the cooperative control digital twin.

[0060] Identify safety violation events and policy conflict points in the post-simulation environment, including: Based on the full-cycle state evolution sequence, the safety boundary compliance of key operating parameters in the collaborative control digital twin is detected, and events in which key operating parameters continuously exceed the preset safe operating range are identified. Events that exceed the preset safe operating range will be treated as safety violation events in the post-simulation environment. In the forward-looking simulation process, a logical consistency analysis is performed on the preliminary coordinated control strategy to obtain potential policy conflict points of the preliminary coordinated control strategy. The potential strategic conflict points are analyzed and the final judgment is made to obtain the strategic conflict points in the post-simulation environment.

[0061] Within the collaborative control digital twin, the initial collaborative control strategy is decomposed into instructions, extracting all the independent operation instructions contained therein. Each instruction is parsed to extract core information such as the node or device ID corresponding to the executing subject, the specific execution action, the preset execution time, and the parameter range corresponding to the action. Then, they are arranged in ascending order according to the timestamp of the preset execution time in all instructions to ensure the temporal logic of instruction execution is coherent, ultimately forming an initial executable control instruction sequence of the collaborative control digital twin arranged in an orderly time sequence.

[0062] The preset simulation step size is 1 minute. This time interval is used as the smallest time unit for step-by-step simulation. The instructions in the initial executable control instruction sequence are allocated according to the simulation step size. Each simulation step size corresponds to the execution of one or more instructions that need to be executed within that time unit. The corresponding nodes and devices in the collaborative control digital twin are driven by the instructions to simulate the actions according to the instructions, thereby realizing the step-by-step simulation process of the collaborative control digital twin driven by the executable control instruction sequence.

[0063] In the step-by-step simulation process of each simulation step, based on the preset equipment response models in the collaborative control digital twin, such as the response delay time of the power output adjustment of the source-side equipment, the efficiency curve of the charging and discharging of the energy storage equipment, the response characteristics of the power change of the load equipment, and the network coupling relationship model, such as the correlation between node voltage and branch transmission power, and the mutual influence law of power flow between different nodes, the state changes of nodes and equipment are calculated according to the instructions executed in the current step, such as the increase or decrease of the power output of the power node, the change of the remaining capacity of the energy storage node, and the adjustment value of the transmission power of the network branch. The original state parameters are replaced with the calculated new state parameters to complete the update of the state parameters of nodes and equipment in the collaborative control digital twin.

[0064] After all instructions in the initial executable control instruction sequence have been executed according to the preset simulation step size, the updated state parameters of all nodes and devices under each simulation step size are collected, including the real-time output of power nodes, the real-time power of load nodes, the remaining capacity and charging / discharging status of energy storage nodes, the transmission power and voltage of network branches, and the operating temperature of devices. These parameters are then sorted and arranged according to the time sequence of the simulation step size to form a dataset covering the state change process of all nodes and devices throughout the entire simulation cycle. This dataset is the full-cycle state evolution sequence of the collaborative control digital twin.

[0065] The key operating parameters in the collaborative control digital twin are clearly defined, including node voltage, network branch transmission power, equipment operating temperature, and system operating frequency. Safe operating ranges for each key operating parameter are preset: node voltage safe operating range is 380V±5%, network branch transmission power safe operating range is no more than the rated transmission power of that branch, equipment operating temperature safe operating range is 0℃-45℃, and system operating frequency safe operating range is 50Hz±0.2Hz. Based on the full-cycle state evolution sequence, the value of each key operating parameter is extracted step by step and compared with the corresponding preset safe operating range to determine whether the parameter value is within the safe range. If the value of a key operating parameter exceeds the corresponding preset safe operating range for three or more consecutive simulation steps, this situation is recorded as an independent event.

[0066] All events where key operating parameters continuously exceed the preset safe operating range are uniformly classified as safety limit violation events in the post-simulation environment. Each safety limit violation event must record in detail the key operating parameter type involved, the numerical range exceeding the safe range, the number of simulation steps continuously exceeding the range, the corresponding node or device ID, and other core information to ensure the traceability of safety limit violation events.

[0067] During each simulation step of the forward-looking simulation, a logical consistency analysis is performed on the instructions in the preliminary coordinated control strategy. The analysis dimensions include time consistency, action consistency, and parameter consistency. Time consistency checks whether the same node or device receives multiple instructions with different time requirements in the same simulation step. Action consistency checks whether the same node or device receives contradictory action instructions in the same simulation step, such as receiving charging and discharging instructions simultaneously. Parameter consistency checks whether the parameter settings in the instructions conflict with the technical parameters of the node or device, such as charging and discharging power parameters exceeding the rated charging and discharging power of the device. Through the above multi-dimensional analysis, the combination of instructions with logical contradictions in the preliminary coordinated control strategy is identified, which are the potential strategy conflict points of the preliminary coordinated control strategy.

[0068] For each identified potential policy conflict point, a specific conflict simulation is initiated within the collaborative control digital twin. The simulation step size range containing the instructions involved in the potential conflict point is isolated, and only all relevant instructions within this range are loaded for individual simulation. The changes in the operating status of nodes and devices in the collaborative control digital twin are observed to determine whether the instruction combination corresponding to the potential conflict point will cause the device to fail to perform actions normally, the system to operate abnormally, or a safety limit violation event to occur. If the specific simulation results show that the above problems will occur, the potential policy conflict point is determined to be a real conflict, and it is identified as a policy conflict point in the post-simulation environment. The simulation step size where the conflict occurs, the content of the instructions involved, and the specific operational problems caused are recorded.

[0069] By conducting forward-looking simulations step by step, a full-cycle, refined simulation of the execution process of the initial collaborative control strategy was achieved, ensuring a high degree of consistency between the simulation results and the actual operating scenario. By clarifying the preset safety range and continuous limit-crossing judgment criteria for key operating parameters, as well as multi-dimensional logical consistency analysis and specific conflict deduction, safety limit-crossing events and strategy conflict points in the simulation environment were comprehensively and accurately identified. This provided clear and specific optimization targets for the subsequent simulation feedback optimization of the initial collaborative control strategy, avoiding the risk of control strategy execution failure due to undiscovered potential safety hazards and strategy contradictions. It ensured the safety, rationality, and executability of the subsequent optimized control strategy, further improving the reliability of the entire digital twin simulation control method.

[0070] In S5, the optimized control strategy for the target park is obtained, including: Based on the safety limit violation events and policy conflict points, abnormal parameters of the preliminary collaborative control strategy are traced to obtain the control parameters to be adjusted in the preliminary collaborative control strategy. A multi-objective trade-off tuning calculation is performed on the control parameter to be adjusted to obtain the adjusted control parameter. The calculation formula for the adjusted control parameter is as follows: ; In the formula, This represents the adjusted control parameter vector. This represents the initial control parameter vector consisting of the control parameters to be adjusted. This represents the safety limit deviation vector. Represents the target conflict difference vector. Represents the security sensitivity matrix. Represents the conflict reconciliation matrix. Indicates the weight of the security limit violation dimension. Indicates the weight of the conflict reconciliation dimension; Based on the adjusted control parameters, the preliminary collaborative control strategy is synchronously corrected and reconstructed to obtain the corrected collaborative control strategy for the target park. The revised collaborative control strategy is subjected to closed-loop verification and simulation. When the closed-loop verification and simulation confirms that the safety limit violation event and the conflict point with the strategy have been eliminated, the revised collaborative control strategy is output as the optimized control strategy for the target park.

[0071] Each identified safety violation event and each policy conflict point is individually associated and matched with the instructions in the preliminary collaborative control strategy. For each safety violation event, the key operating parameters and corresponding control instructions involved are analyzed to identify the specific control items in the instruction that directly regulate the key operating parameters. For each policy conflict point, the relevant control items of the same or related control objects involved in the conflicting instructions are analyzed. By tracking the specific values ​​of these control items in the preliminary collaborative control strategy, all control items that have a direct causal relationship with the safety violation event or policy conflict point are screened out. These control items are then integrated to form the control parameters to be adjusted in the preliminary collaborative control strategy.

[0072] First, define the initial set of control parameters, which represents the specific combinations of values ​​for the selected control parameters to be adjusted. Next, determine the safe over-limit deviation data. For each control parameter to be adjusted, calculate the difference between the actual operating value of that parameter and the preset safe operating range boundary value for each over-limit event, ensuring that each over-limit situation has a clear record of the difference.

[0073] Next, target conflict difference data is determined. For each control parameter to be adjusted, the specific difference in the parameter value in the conflicting instructions is calculated, and the difference for each conflict is fully recorded. Then, safety sensitivity-related impact data is constructed. By analyzing the response characteristics of the devices and the coupling relationship with the network in the collaborative control digital twin, combined with historical operating data of the target park and multiple simulation data, the probability change of the corresponding key operating parameters exceeding the preset safe operating range when each control parameter to be adjusted changes by a fixed unit is statistically analyzed. These values ​​are organized according to the correspondence between control parameters and limit-breaking events to form system impact data. This data can accurately reflect the impact of different control parameters on safety limit-breaking events.

[0074] Simultaneously, data on the degree of mitigation related to conflict reconciliation is constructed. By analyzing the collaborative control mechanism and security verification logic of the target park, and combining data from past cases of handling strategic conflict points, the specific numerical reduction in the severity of strategic conflict is calculated when each control parameter to be adjusted changes by a fixed unit. These numerical values ​​are then organized according to the correspondence between control parameters and conflict points to form systematic mitigation data. This data can clearly demonstrate the effect of different control parameters on resolving strategic conflicts.

[0075] Next, weight allocation was set. Based on the target park's primary objective of safe operation while also considering the control requirements for resolving strategy conflicts, the weight related to safety exceeding limits was set at 60%, and the weight related to strategy conflict reconciliation was set at 40%, ensuring that system safety was prioritized during the adjustment process. The adjustment amount related to safety exceeding limits was calculated by multiplying the weight of safety exceeding limits by the corresponding safety sensitivity impact value, and then multiplying by the corresponding safety exceeding limit deviation value for that control parameter, to obtain the individual adjustment amount for each control parameter to be adjusted for safety exceeding limits.

[0076] Calculate the adjustment amount related to strategy conflict by multiplying the weight related to strategy conflict reconciliation by the corresponding conflict reconciliation and mitigation degree value, and then multiplying by the target conflict difference value corresponding to the control parameter. This yields the individual adjustment amount for each control parameter to be adjusted for strategy conflict. Add the individual adjustment amount for safety exceeding limits to the individual adjustment amount for strategy conflict for each control parameter to obtain the total adjustment amount for that parameter. Finally, subtract the corresponding total adjustment amount from the initial value of each control parameter to obtain the adjusted value of each control parameter. All adjusted values ​​are then integrated to form the adjusted control parameter.

[0077] Each value in the adjusted control parameters is precisely matched with the corresponding control item in the initial collaborative control strategy. The original values ​​of the control parameters to be adjusted in the initial collaborative control strategy are replaced one by one, ensuring that all instructions in the strategy involving that control parameter are synchronously updated. During the replacement process, the overall structure, timing logic, and applicable scenario framework of the initial collaborative control strategy remain unchanged. Only the instruction content related to the adjusted control parameters is modified. Simultaneously, it is checked whether there are new logical contradictions between the modified instructions. If new minor logical deviations occur, local fine-tuning is performed based on the core requirements of the adjusted control parameters to ensure that all instructions are sequentially coherent, logically consistent, and parameter-matched, ultimately forming a structurally complete and effective corrected collaborative control strategy.

[0078] The revised cooperative control strategy is input into the cooperative control digital twin. Following a preset one-minute simulation step, the instructions in the strategy are sequentially ordered to form an executable control instruction sequence. This sequence drives the digital twin to perform step-by-step simulations. During each simulation step, based on the preset device response model and network coupling relationship model in the digital twin, the state parameters of each node and device are updated. After all instructions have been executed, the entire lifecycle state evolution sequence is integrated.

[0079] Based on this sequence, safety boundary compliance checks are performed on key operating parameters to confirm whether there are any events that continuously exceed the preset safe operating range. Simultaneously, logical consistency analysis is performed on the instructions in the strategy to confirm whether there are any potential strategy conflict points. If the detection results show that there are no safety limit violations or strategy conflict points, the closed-loop verification simulation is considered successful. If related problems still exist, the abnormal parameter tracing step is returned to, and the relevant parameters in the corrected collaborative control strategy are re-traced, tuned, and corrected until the closed-loop verification simulation confirms that all safety limit violations and strategy conflict points have been eliminated. At this point, the corrected collaborative control strategy is output as the optimized control strategy for the target park.

[0080] By precisely tracing the sources of abnormal parameters, the core control parameters leading to safety exceedances and strategy conflicts in the initial collaborative control strategy were identified, providing a clear target for subsequent optimization and avoiding blind adjustments. During the multi-objective trade-off tuning calculation, data on the degree of impact related to safety sensitivity and the degree of mitigation related to conflict reconciliation were introduced. The former, through statistical analysis of historical and simulation data, precisely quantified the impact of control parameters on safe operation, while the latter, through analysis of the control mechanism and summarization of historical cases, clarified the effect of control parameters on conflict resolution. Combined with fixed weights set based on the priority of control objectives, the needs for safety assurance and conflict resolution were transformed into specific, calculable adjustment quantities. This ensures that the adjusted control parameters not only prioritize system safety operation requirements but also efficiently resolve strategy conflicts, while simultaneously considering economic efficiency, energy efficiency, and renewable energy consumption targets.

[0081] The synchronous correction and reconstruction process ensures the consistency and integrity of strategy adjustments, avoiding new logical contradictions caused by local modifications and ensuring the structural stability and command effectiveness of the corrected collaborative control strategy. Closed-loop verification and simulation, through a mechanism of cyclical detection and repeated adjustments, completely eliminates safety limit violations and strategy conflict points, ensuring the optimized control strategy possesses extremely high executability and reliability. The entire simulation feedback optimization process transforms abstract parameter adjustment logic into concrete and operable implementation steps, compensating for the shortcomings of the initial collaborative control strategy, significantly improving the accuracy and adaptability of the control strategy, and providing a solid guarantee for the safe and efficient implementation of multi-energy flow collaborative control of source-grid-load-storage in the target park.

[0082] In S6, the optimized control strategy is parsed into collaborative control instructions for the target area, and these instructions are transmitted to the control terminal of the target area, including: The timing parameters of the target park are obtained by analyzing the timing parameters of the optimized control strategy. The timing control parameters are encoded into instructions to obtain the standard control instructions for the target park. The standard control command set is sent to the control terminal of the target park through the communication interface of the collaborative control digital twin.

[0083] The optimized control strategy is broken down sentence by sentence, and all valid control instructions contained therein are extracted. Each instruction clearly defines the unique identifier of the node or equipment in the target park corresponding to the executing entity, the specific execution action, the preset execution timestamp, the specific numerical range corresponding to the action, and the execution result verification standard. All instructions are sorted in ascending order of execution timestamp. At the same time, it is checked whether there are multiple conflicting instructions under the same execution entity at the same timestamp to ensure the continuity of the timing logic. The core information of each instruction after sorting, such as the executing entity identifier, execution action, timestamp, numerical range, and verification standard, is integrated into structured data to form the timing control parameters of the target park.

[0084] A unified standard control instruction coding rule was established. This rule specifies that the instruction code consists of five parts: the execution subject identifier segment, the execution action code segment, the timestamp code segment, the value range code segment, and the check code segment. The execution subject identifier segment uses an 8-bit binary number, corresponding to a unique identifier for all nodes and devices within the target area; the execution action code segment uses a 4-bit binary number, with each execution action corresponding to a unique binary code value; the timestamp code segment uses a 16-bit binary number, accurate to the second; the value range code segment uses a 24-bit binary number, converting the value range into the corresponding binary number according to a preset ratio; and the check code segment uses an 8-bit binary number, obtained by XORing the first four segments. Following this coding rule, each structured data item in the timing control parameters is converted into a fixed-length binary code, with each code corresponding to an independent standard control instruction. All standard control instructions are integrated to form a standard control instruction set.

[0085] The collaborative control digital twin is configured with a dedicated Ethernet communication interface, which supports the TCP / IP communication protocol. The default communication transmission rate is 115200bps, and the data transmission format is 8 data bits, 1 stop bit, and no parity bit. Standard control command sets are encapsulated into data packets according to TCP / IP protocol requirements. Each data packet includes a header, command set fragment, packet trailer, and CRC32 checksum. The header clearly identifies the data packet sequence number and the total number of packets. A command issuance trigger mechanism is set up to check for new standard control command sets every minute. When a new command set is detected, a connection request is initiated to the control terminal in the target area via the communication interface. After the control terminal responds to the connection, it sends the encapsulated command data packets sequentially according to the data packet sequence number.

[0086] After receiving each data packet, the control terminal verifies the integrity of the data packet using a CRC32 checksum. Once the verification is successful, it sends back a reception confirmation signal. The collaborative control digital twin receives the confirmation signal and then sends the next data packet. This process continues until all data packets have been sent, completing the distribution of the standard control command set to the target park control terminal.

[0087] By accurately analyzing the timing parameters of the optimized control strategy, it was ensured that the timing control parameters were completely consistent with the core requirements of the optimized control strategy, and that the timing logic was coherent and conflict-free, providing a structured and standardized data foundation for subsequent instruction encoding. Unified encoding rules transformed the timing control parameters into standard control instructions that the control terminal could directly recognize, avoiding the terminal's inability to parse instructions due to inconsistent instruction formats, and improving instruction compatibility and executability. The adoption of an Ethernet communication interface supporting the TCP / IP protocol and a data packet encapsulation, CRC32 checksum, and acknowledgment feedback mechanism ensured the integrity, accuracy, and reliability of the standard control instruction set during transmission, preventing instruction loss or tampering. This enabled the optimized control strategy to be accurately and efficiently transmitted to the control terminal and executed, completing the closed-loop process of the entire digital twin simulation control and guaranteeing the accuracy and real-time performance of the multi-energy flow coordinated control of the target park's source-grid-load-storage systems.

[0088] The method of this embodiment is applied to an integrated source-grid-load-storage park scenario under a low-voltage power distribution system. The comparison curves of the system node voltage optimization effect using the collaborative control digital twin scheme proposed in this embodiment are shown below. Figure 2 As shown, Figure 2 The red dashed line represents the system node voltage state before the implementation of the solution, which fluctuated significantly (voltage values ​​oscillated wildly between 380V and 395V), posing a risk of deviating from the stable operating range of the equipment. The blue solid line represents the voltage curve after the implementation of the solution. Relying on the collaborative control digital twin for dynamic sensing and precise coordinated regulation of the source-grid-load-storage equipment, the node voltage fluctuations significantly converged, and the operational stability was significantly improved. This curve visually verifies the practical effectiveness of the method in this embodiment in optimizing system voltage characteristics and ensuring stable power supply.

[0089] Example 2: A digital twin simulation control device for multi-energy flow coordination in an integrated power generation, grid, load, and storage park, comprising: One or more processors; Memory, used to store one or more computer programs; When one or more programs are executed by one or more processors, the one or more processors execute the method in Example 1.

[0090] Example 3: A computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, cause the processor to perform the method in Example 1.

Claims

1. A digital twin simulation control method for multi-energy flow coordination in an integrated power generation, grid, load, and storage park, characterized by the following steps: include: S1. Perform multi-source heterogeneous data fusion on the controllable operation setting parameters and status monitoring data of the target park to obtain the real-time operation status data of the target park; S2. Based on the network topology of the target park, perform hierarchical virtual mapping of the target park to construct a collaborative control digital twin of the target park; S3. Based on the collaborative control digital twin and real-time operation status data, multi-constraint coordination planning is carried out on the source-grid-load-storage resources of the target park to obtain the preliminary collaborative control strategy of the target park. S4. Conduct a forward-looking simulation of the preliminary collaborative control strategy to identify safety limit violations and strategy conflict points in the simulation environment after the simulation. S5. Based on safety violation events and policy conflict points, perform simulation feedback optimization on the preliminary collaborative control strategy to obtain the optimized control strategy for the target park. S6. Parse the optimized control strategy into collaborative control instructions for the target park and send the collaborative control instructions to the control terminal of the target park.

2. The digital twin simulation control method for multi-energy flow coordination in an integrated source-grid-load-storage park according to claim 1, characterized in that, In S1, the process of obtaining the real-time operating status data of the target park is as follows: Collect source-side equipment output plan parameters, grid-side topology connection parameters, load-side real-time power data, and energy storage equipment charging and discharging status parameters of the target park; The communication protocol is parsed to obtain multi-source data of the target park by analyzing the power output planning parameters of the source-side equipment, the topology connection parameters of the network side, the real-time power data of the load side, and the charging and discharging status parameters of the energy storage equipment. Based on a unified time series benchmark, the multi-source data is formatted and unified to obtain the synchronous multi-source data of the target park; Redundant data is removed from the synchronized multi-source data to obtain the real-time operational status data of the target park.

3. The digital twin simulation control method for multi-energy flow coordination in an integrated source-grid-load-storage park according to claim 1, characterized in that, In S2, the process of constructing a collaborative control digital twin of the target park is as follows: Analyze the network topology of the target campus and identify the power nodes, load nodes, energy storage nodes, and network branches within the target campus; A static topology layer for the target park is constructed using power nodes, load nodes, and energy storage nodes as static network nodes and network branches as relationships. By dynamically associating the nodes in the static topology layer with their characteristics, the dynamic behavior layer of the target park is obtained. Based on the static topology layer and dynamic behavior layer, and combined with the control objectives and security constraints of the target park, a collaborative control digital twin of the target park is constructed.

4. The digital twin simulation control method for multi-energy flow coordination in an integrated source-grid-load-storage park according to claim 3, characterized in that, Based on the control objectives and security constraints of the target park, a collaborative control digital twin of the target park is constructed, specifically as follows: The control objectives of the target park are analyzed by targeted feature deconstruction to obtain the economic indicators, energy efficiency indicators and renewable energy consumption indicators of the target park; Based on economic indicators, energy efficiency indicators, and renewable energy consumption indicators, the coordinated control mechanism of the target industrial park is analyzed to obtain the coordinated decision-making logic of the target industrial park. The security constraints of the target area are transformed into constraint rules to obtain the security verification logic of the target area. The coordination decision-making logic and security verification logic are functionally integrated to obtain the decision-making logic layer of the target park; The static topology layer, dynamic behavior layer, and decision logic layer are encapsulated and integrated in a programmatic manner to obtain a collaborative control digital twin of the target park.

5. The digital twin simulation control method for multi-energy flow coordination in an integrated source-grid-load-storage park according to claim 1, characterized in that, In S3, the process of obtaining the preliminary collaborative control strategy for the target park is as follows: Real-time operating status data is injected into the collaborative control digital twin to obtain the planning initial boundary conditions of the collaborative control digital twin; Based on the initial boundary conditions of the plan, the constraint range of the source-grid-load-storage resources of the target park is analyzed to obtain the set of resource adjustability and the set of operational constraints of the target park. Based on the set of adjustable resource capabilities and the set of operational constraints, an integrated coordinated decision-making process is made for source-grid-load-storage resources to obtain a preliminary collaborative control strategy for the target park.

6. The digital twin simulation control method for multi-energy flow coordination in an integrated source-grid-load-storage park according to claim 5, characterized in that, By conducting integrated coordination and decision-making on the source-grid-load-storage resources, a preliminary collaborative control strategy for the target park is obtained, specifically as follows: Based on the resource adjustability set and the operational constraint set, the operable boundary of source-grid-load-storage resources is traced to obtain the collaborative scheduling feasible domain of the target park; The feasible domain of coordinated scheduling is used as a constraint and input into the decision logic layer of the coordinated control digital twin to generate a preliminary coordinated instruction sequence for the target park. The initial coordination instruction sequence is encapsulated in a strategy format to obtain the initial collaborative control strategy for the target park.

7. The digital twin simulation control method for multi-energy flow coordination in an integrated source-grid-load-storage park according to claim 1, characterized in that, In S4, the process of prospective simulation of the preliminary coordinated control strategy is as follows: Within the collaborative control digital twin, the initial instructions of the preliminary collaborative control strategy are parsed out, and the initial instructions are sequentially ordered to obtain the initial executable control instruction sequence of the collaborative control digital twin. According to the preset simulation step size, the collaborative control digital twin is driven to perform step-by-step simulation using a sequence of executable control instructions; During the step-by-step simulation, the state parameters of nodes and devices in the collaborative control digital twin are updated based on the device response and network coupling relationship of the collaborative control digital twin. After all the instructions in the executable control instruction sequence have been executed, the updated state parameters are integrated to obtain the full-cycle state evolution sequence of the cooperative control digital twin.

8. The digital twin simulation control method for multi-energy flow coordination in an integrated source-grid-load-storage park according to claim 7, characterized in that, In S4, safety violation events and policy conflict points in the post-simulation environment are identified, specifically: Based on the full-cycle state evolution sequence, the safety boundary compliance of key operating parameters in the collaborative control digital twin is detected, and events in which key operating parameters continuously exceed the preset safe operating range are identified. Events that exceed the preset safe operating range will be treated as safety violation events in the post-simulation environment. In the forward-looking simulation process, a logical consistency analysis is performed on the preliminary coordinated control strategy to obtain potential policy conflict points of the preliminary coordinated control strategy. The potential strategic conflict points are analyzed and the final judgment is made to obtain the strategic conflict points in the post-simulation environment.

9. The digital twin simulation control method for multi-energy flow coordination in an integrated source-grid-load-storage park according to claim 1, characterized in that, In S5, the process of obtaining the optimal control strategy for the target park is as follows: Based on the safety limit violation events and policy conflict points, abnormal parameters of the preliminary collaborative control strategy are traced to obtain the control parameters to be adjusted in the preliminary collaborative control strategy. A multi-objective trade-off tuning calculation is performed on the control parameter to be adjusted to obtain the adjusted control parameter. The calculation formula for the adjusted control parameter is as follows: ； In the formula, This represents the adjusted control parameter vector. This represents the initial control parameter vector consisting of the control parameters to be adjusted. This represents the safety limit deviation vector. Represents the target conflict difference vector. Represents the security sensitivity matrix. Represents the conflict reconciliation matrix. Indicates the weight of the security limit violation dimension. Indicates the weight of the conflict reconciliation dimension; Based on the adjusted control parameters, the preliminary collaborative control strategy is synchronously corrected and reconstructed to obtain the corrected collaborative control strategy for the target park. The revised collaborative control strategy is subjected to closed-loop verification and simulation. When the closed-loop verification and simulation confirms that the safety limit violation event and the conflict point with the strategy have been eliminated, the revised collaborative control strategy is output as the optimized control strategy for the target park.

10. The digital twin simulation control method for multi-energy flow coordination in an integrated source-grid-load-storage park according to claim 1, characterized in that, In S6, the optimized control strategy is parsed into collaborative control instructions for the target area, and these instructions are then transmitted to the control terminal of the target area. Specifically: The timing parameters of the target park are obtained by analyzing the timing parameters of the optimized control strategy. The timing control parameters are encoded into instructions to obtain the standard control instructions for the target park. The standard control command set is sent to the control terminal of the target park through the communication interface of the collaborative control digital twin.

Images