A multi-source cooperative park carbon emission optimization control method and system

Abstract

This invention discloses a multi-source collaborative carbon emission optimization control method and system for industrial parks, specifically relating to the field of industrial park energy management technology. It addresses the problem that existing technologies fail to consider the cross-regional carbon transfer effect caused by load regulation, leading to increased overall system carbon emissions due to localized carbon reduction strategies. The method involves real-time collection of grid carbon intensity, distributed power generation output, and load demand data; generating carbon emission optimization instructions when the grid is high-carbon and experiencing power shortages; marking high-risk carbon transfer periods based on distribution network section congestion status and generator carbon emission ramp-up rates; dynamically correcting the carbon emission transfer factor of load regulation; constructing carbon-energy synergistic constraints in conjunction with distributed power generation absorption requirements; and generating and executing energy storage and load regulation instructions to achieve synergistic optimization of full absorption of clean energy in the industrial park and minimization of system-level carbon emissions.

Description

Technical Field

[0001] This invention relates to the field of energy management technology in industrial parks, and more specifically, to a multi-source collaborative method and system for optimizing and controlling carbon emissions in industrial parks. Background Technology

[0002] The increasing penetration rate of clean energy sources such as distributed photovoltaic and wind power in industrial parks necessitates that park energy management simultaneously consider both the full absorption of distributed power sources and the minimization of carbon emissions. Current mainstream solutions utilize static optimization models, constrained by historical average carbon emission factors, to coordinate resources such as energy storage and controllable loads to achieve energy balance within the park. When participating in distribution network interactions, the park typically aggregates into a single controllable unit to respond to dispatch commands. However, this model fails to consider real-time fluctuations in grid carbon intensity and the cross-regional carbon transfer effects caused by load regulation, making it difficult to accurately quantify the park's dynamic carbon footprint.

[0003] In existing technologies, the lack of a dynamic carbon flow transmission mechanism for the power grid and the coupling relationship between multi-source collaborative control in industrial parks means that local carbon reduction strategies (such as load reduction in industrial parks) may force the distribution network to call up high-carbon backup power sources, which may increase the overall carbon emissions of the system. This contradiction between the carbon optimization target and the actual system operation makes it impossible for industrial parks to achieve precise carbon emission control while ensuring the full consumption of clean energy. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides a multi-source collaborative carbon emission optimization control method and system for industrial parks to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A multi-source collaborative method for optimizing and controlling carbon emissions in industrial parks includes the following steps:

[0007] S1. Real-time collection of carbon intensity data of the power distribution network, output data of distributed power sources in the park, and load demand data;

[0008] S2. When the real-time difference between the distributed power generation output data and the load demand data indicates that there is a power deficit and the grid carbon intensity is higher than the preset threshold, a carbon emission optimization instruction is generated.

[0009] S3. When generating carbon emission optimization instructions, monitor the real-time blockage status and carbon emission ramp-up rate of key sections of the distribution network, and generate high-risk carbon transfer period markers.

[0010] S4. Based on carbon intensity data and high-risk periods for carbon transfer, calculate the carbon emission transfer factor corresponding to load adjustment behavior;

[0011] S5. Combining the carbon emission transfer factor and the requirements for distributed power consumption, construct carbon-energy synergistic constraints;

[0012] S6. Generate energy storage charging and discharging power commands and controllable load adjustment commands based on carbon-energy synergy constraints and send them to the park's energy management platform for execution.

[0013] Furthermore, real-time data collection includes carbon intensity data of the distribution network, distributed power generation output data within the park, and load demand data, including:

[0014] Carbon intensity data of the distribution network is obtained in real time from the power dispatch data platform;

[0015] The output data of distributed photovoltaic power generation units and wind power generation units in the park are collected by smart meters.

[0016] The load monitoring device collects real-time power data of controllable loads and real-time power data of uncontrollable loads from the load demand data.

[0017] The carbon intensity data of the distribution network, the output data of distributed photovoltaic power generation units, the output data of wind power generation units in the park, the real-time power data of controllable loads and the real-time power data of uncontrollable loads are bound together and stored with a unified timestamp.

[0018] Furthermore, when the real-time difference between distributed generation output data and load demand data indicates a power deficit and the grid carbon intensity exceeds a preset threshold, a carbon emission optimization instruction is generated, including:

[0019] Calculate the real-time difference between the sum of distributed power generation output data and the sum of load demand data;

[0020] When the real-time difference is less than zero, it is determined that there is a power deficit;

[0021] Simultaneously, the current value of carbon intensity data of the distribution network is obtained and compared with the preset carbon intensity threshold;

[0022] When the current value of the carbon intensity data of the distribution network is greater than the preset carbon intensity threshold and there is a power deficit, a carbon emission optimization instruction is generated.

[0023] Associate the carbon emission optimization instruction with the current timestamp.

[0024] Furthermore, when generating carbon emission optimization instructions, the real-time congestion status and carbon emission ramp-up rate of key sections of the distribution network are monitored to generate high-risk carbon transfer periods, including:

[0025] Obtain the real-time congestion status of key sections of the distribution network from the power grid dispatch automation system;

[0026] Read the carbon emission ramp rate of the standby unit from the generator set characteristic database;

[0027] When the real-time blocking status is blocked and the carbon emission ramp rate is greater than the preset ramp rate threshold, the current period is determined to be a high-risk period for carbon transfer.

[0028] Generate a marker for a high-risk period of carbon transfer that is linked to the current timestamp;

[0029] The timestamps of carbon emission optimization instructions are stored to mark high-risk periods of carbon transfer.

[0030] Furthermore, the real-time congestion status of key sections of the distribution network is obtained from the power grid dispatch automation system, including:

[0031] Power flow data of key sections are collected in real time through the SCADA interface of the scheduling automation system;

[0032] Compare the current flow data with the cross-sectional safe transmission limits;

[0033] When the absolute value of the power flow data exceeds the safe transmission limit, the real-time blocking state is determined to be a blocking state.

[0034] Otherwise, the real-time blocking state is determined to be a non-blocking state.

[0035] Furthermore, the carbon emission ramp rate of the standby unit is read from the generator set characteristic database, including:

[0036] The database is queried based on the standby unit number determined by the power grid dispatching instructions;

[0037] Extract the carbon emission coefficient per unit power change corresponding to the unit number;

[0038] Read the maximum power ramp rate of the unit under its current output status;

[0039] Multiplying the carbon emission coefficient per unit power change by the maximum power ramp rate yields the carbon emission ramp rate.

[0040] Furthermore, based on carbon intensity data and markers of high-risk periods for carbon transfer, the carbon emission transfer factor corresponding to load adjustment behavior is calculated, including:

[0041] Obtain the current value of carbon intensity data for the distribution network; read the status value marked during high-risk periods of carbon transfer;

[0042] When a high-risk period for carbon transfer is marked as active, the current value of the carbon intensity data of the distribution network is corrected using a preset weighting factor;

[0043] Multiply the corrected carbon intensity data by the unit load adjustment amount to obtain the carbon emission transfer factor corresponding to the load adjustment behavior;

[0044] When a high-risk period for carbon transfer is marked as inactive, the current value of the carbon intensity data of the distribution network is multiplied by the unit load adjustment amount to obtain the carbon emission transfer factor corresponding to the load adjustment behavior.

[0045] Furthermore, combining the carbon emission transfer factor and the requirements for distributed power generation, carbon-energy synergistic constraints are constructed, including:

[0046] Obtain the carbon emission transfer factor corresponding to load adjustment behavior;

[0047] Read the real-time absorbable power from the output data of distributed power sources;

[0048] Establish a total load regulation constraint: the controllable load regulation amount shall not exceed the sum of the real-time absorbable electricity and the power deficit;

[0049] Establish carbon emission constraints: The total amount of carbon emissions transferred by all controllable load adjustment behaviors shall not exceed the preset carbon emission threshold;

[0050] The total load regulation constraint and carbon emission constraint are combined into a carbon-energy synergistic constraint condition.

[0051] Furthermore, based on the carbon-energy synergistic constraints, energy storage charging and discharging power commands and controllable load adjustment commands are generated and sent to the park's energy management platform for execution, including:

[0052] Invoke the total load regulation constraint and carbon emission constraint in the carbon-energy synergistic constraint conditions;

[0053] Determine the total controllable load regulation amount based on the total load regulation constraint;

[0054] Determine the priority order of controllable load adjustment based on carbon emission constraints;

[0055] Controllable load adjustment instructions are generated based on the total controllable load adjustment amount and priority order;

[0056] Generate energy storage charging and discharging power commands by combining the current state of charge of the energy storage with the maximum charging and discharging power.

[0057] The controllable load adjustment command and the energy storage charging and discharging power command are linked with timestamps and sent to the park's energy management platform.

[0058] On the other hand, the present invention provides a multi-source collaborative carbon emission optimization and control system for industrial parks, comprising the following modules:

[0059] The multi-source acquisition module is used to collect carbon intensity data of the power distribution network, output data of distributed power sources in the park, and load demand data in real time.

[0060] The optimization decision module is used to generate carbon emission optimization instructions when the real-time difference between distributed power generation output data and load demand data indicates a power deficit and the grid carbon intensity is higher than a preset threshold.

[0061] The congestion monitoring module is used to monitor the real-time congestion status and carbon emission ramp-up rate of key sections of the distribution network when carbon emission optimization instructions are generated, and to generate high-risk carbon transfer period markers.

[0062] The factor calculation module is used to calculate the carbon emission transfer factor corresponding to load adjustment behavior based on carbon intensity data and high-risk carbon transfer periods.

[0063] The constraint construction module is used to construct carbon-energy synergistic constraints by combining the carbon emission transfer factor and the requirements for distributed power generation.

[0064] The instruction execution module is used to generate energy storage charging and discharging power instructions and controllable load adjustment instructions based on carbon-energy synergy constraints and send them to the park's energy management platform for execution.

[0065] Compared with the prior art, the present invention has the following beneficial effects:

[0066] 1. By dynamically coupling the power grid carbon flow transfer mechanism with the multi-source collaborative control of the park, the cross-regional carbon transfer effect caused by load regulation behavior can be accurately quantified; based on the real-time monitoring of the blockage status of key sections of the distribution network and the carbon emission ramp-up rate of standby units, a carbon transfer high-risk period marking mechanism is constructed, and the carbon emission transfer factor of load regulation is dynamically corrected in combination with the power grid carbon intensity. This solves the contradiction of "local carbon reduction and global emission increase" caused by the traditional static optimization model ignoring the dynamic characteristics of power grid carbon flow, so that the park can effectively suppress the system-level carbon emission rebound caused by load regulation while ensuring the full absorption of distributed power sources.

[0067] 2. By constructing carbon-energy synergistic constraints, carbon emission transfer factors and distributed power consumption needs are coordinated in multiple dimensions to generate energy storage and load regulation instructions that take into account both grid congestion and carbon emission ramp-up risks. During high-carbon periods in the distribution network, carbon transfer risks are proactively avoided, and the park's energy dispatch is transformed from simple energy balance to carbon-energy synergistic optimization. Attached Figure Description

[0068] Figure 1 This is a flowchart of a multi-source collaborative carbon emission optimization control method for industrial parks according to the present invention;

[0069] Figure 2 This is a schematic diagram of the structure of a multi-source collaborative carbon emission optimization control system for industrial parks according to the present invention. Detailed Implementation

[0070] 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 of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0071] Example 1: Figure 1 This invention presents a multi-source synergistic carbon emission optimization and control method for industrial parks, which includes the following steps:

[0072] S1. Real-time collection of carbon intensity data of the power distribution network, output data of distributed power sources in the park, and load demand data;

[0073] S2. When the real-time difference between the distributed power generation output data and the load demand data indicates that there is a power deficit and the grid carbon intensity is higher than the preset threshold, a carbon emission optimization instruction is generated.

[0074] S3. When generating carbon emission optimization instructions, monitor the real-time blockage status and carbon emission ramp-up rate of key sections of the distribution network, and generate high-risk carbon transfer period markers.

[0075] S4. Based on carbon intensity data and high-risk periods for carbon transfer, calculate the carbon emission transfer factor corresponding to load adjustment behavior;

[0076] S5. Combining the carbon emission transfer factor and the requirements for distributed power consumption, construct carbon-energy synergistic constraints;

[0077] S6. Generate energy storage charging and discharging power commands and controllable load adjustment commands based on carbon-energy synergy constraints and send them to the park's energy management platform for execution.

[0078] S1. Real-time collection of carbon intensity data from the power distribution network, distributed power generation output data within the industrial park, and load demand data. Specific implementation details are as follows:

[0079] Carbon intensity data of the distribution network is obtained in real time from the power dispatch data platform by calling the platform's standard data interface. This interface is designed according to the general communication architecture of the power system, returning structured data messages at fixed time intervals. These messages contain a field identified as "grid carbon intensity." The data parsing unit extracts the value of this field as the carbon intensity data of the distribution network, with the unit being kilograms of carbon dioxide per kilowatt-hour. A retry mechanism is automatically triggered when the interface request fails; if consecutive failures exceed a set number, the most recently valid cached value is used.

[0080] Output data of distributed photovoltaic (PV) power generation units within the park is collected using smart meters that comply with electricity metering standards. Each PV unit has an independent metering device at its grid connection point, connected to the data acquisition unit via an industrial bus. The data acquisition unit periodically sends data request commands, which include the dedicated register address code for PV power data. The smart meters return instantaneous power values ​​in binary format, which the data acquisition unit converts to decimal values ​​and performs a unit conversion process to obtain the output data of the distributed PV unit in megawatts. The same technical principle is used to collect output data from the park's wind power generation units, the difference being the register address code for the wind power generation unit in the data request command.

[0081] Real-time power data for controllable loads is collected from load demand data using load monitoring devices, specifically monitoring terminals with two-way communication capabilities. These terminals are installed on the power supply circuits of adjustable electrical equipment within the park and periodically upload power measurements with device identification via a wireless communication protocol. Real-time power data for uncontrollable loads is collected using fixed-installation monitoring terminals, transmitting data via a wired communication protocol. All raw power data is in kilowatts (kW), which is then standardized at the data aggregation node, converted to megawatts (MW), and categorized with labels.

[0082] The carbon intensity data of the distribution network, the output data of distributed photovoltaic power generation units, the output data of wind power generation units in the park, and the real-time power data of controllable and uncontrollable loads are bound to a unified timestamp for storage, specifically using a time-series database management system. A high-precision clock synchronization module is deployed in the data receiving service to obtain a standard time reference source from the time synchronization server. When output data from a distributed photovoltaic power generation unit is received, the current precise time point T is recorded and a data record is generated; when processing other data sources synchronously, the same time point T is used to generate corresponding records. The storage engine establishes a data association structure indexed by time point T, ensuring that multi-dimensional data at the same time point can be queried through index association. Time synchronization deviation is controlled within a set threshold; when data delay exceeds the maximum allowable value, a data compensation process is initiated.

[0083] The carbon intensity data acquisition process for the distribution network includes a validity verification step. When the value returned by the interface exceeds a reasonable range, it is automatically replaced with the historical average. The reasonable range is determined based on the carbon emission benchmarks of the power industry; for example, the typical range for some regional power grids is between 0.5 kJ and 1.1 kJ of CO2 per kilowatt-hour. The output data acquisition process for distributed photovoltaic power generation units includes range verification. When the measured value exceeds the rated capacity of the physical equipment, an anomaly handling process is initiated. The real-time power data transmission process for controllable loads employs a data integrity verification mechanism. If the verification fails, a data retransmission process is initiated.

[0084] The data interface call process implements a secure authentication mechanism, verifying the requester's identity through digital certificates. The physical connection between the metering equipment and the data acquisition unit conforms to industrial bus deployment specifications, and the bus length does not exceed the maximum value allowed by the technical standard. The wireless communication of the load monitoring device adopts encrypted transmission technology, and the network topology design meets signal coverage requirements. The time-series database system is configured with a data retention policy, automatically cleaning up overdue historical data. The time synchronization system is configured with primary and backup clock sources to ensure the reliability of time reference.

[0085] The data storage structure strictly adheres to the principle of terminology consistency. The output data field name for distributed photovoltaic (PV) power generation units is fixed as "PV Power Data," the output data field name for wind power generation units in the industrial park is fixed as "Wind Power Data," the real-time power data field name for controllable loads is fixed as "Controllable Load Power," the real-time power data field name for uncontrollable loads is fixed as "Uncontrollable Load Power," and the carbon intensity data field name for the distribution network is fixed as "Grid Carbon Intensity." All numerical fields are stored in floating-point format, and the timestamp field stores time information accurate to milliseconds. The data write interface implements a flow control strategy, activating a buffer queue mechanism when the data rate exceeds processing capacity.

[0086] The output data conversion process for distributed photovoltaic (PV) power generation units includes data processing steps, dividing the original metered value by 1000 to convert kilowatts to megawatts. For example, an original measurement of 1500 kilowatts is converted to 1.5 megawatts. The output data of the wind power generation units in the park uses the same conversion factor. The unit conversion processing of load power data is performed uniformly at the data aggregation node to avoid accuracy loss caused by multiple conversions. The timestamp binding process uses nanosecond-level time resolution, for example, recording the complete time format "2024-08-20T14:30:45.123456789Z".

[0087] In the data validity verification process, the historical average is calculated using a sliding time window method, for example, taking the arithmetic mean of valid data within the most recent 5 minutes. The range verification threshold is set according to the equipment's technical parameters, for example, not exceeding 1.2 times the rated capacity. The communication bus length limit is set based on signal attenuation characteristics, for example, an RS-485 bus not exceeding 1200 meters. The time synchronization deviation threshold is determined according to the data application scenario, for example, not exceeding 200 milliseconds. The buffer queue depth is configured according to the system's processing capacity, for example, accommodating a maximum of 10,000 data points.

[0088] S2. When the real-time difference between distributed generation output data and load demand data indicates a power deficit and the grid carbon intensity is higher than a preset threshold, a carbon emission optimization instruction is generated, specifically as follows:

[0089] The real-time difference between the sum of distributed generation output data and the sum of load demand data is calculated. Specifically, the output data of the currently bound distributed photovoltaic (PV) power generation units, the output data of the park's wind power generation units, the real-time power data of controllable loads, and the real-time power data of uncontrollable loads are obtained from the time-series database. The sum of distributed generation output data equals the sum of the output data of the distributed PV power generation units and the output data of the park's wind power generation units. The sum of load demand data equals the sum of the real-time power data of controllable loads and the real-time power data of uncontrollable loads. The real-time difference is obtained by subtracting the sum of the load demand data from the sum of the distributed generation output data. The calculation result is in megawatts (MW). The data acquisition time window is 500 milliseconds to ensure that all data are synchronous measurements within the same acquisition period.

[0090] When the real-time difference is less than zero, a power deficit is determined, and the numerical comparison logic is executed. The comparator receives the real-time difference input; when it detects that the value is less than zero, it generates a Boolean status flag "Power Deficit Status" and sets it to true. The absolute value of the power deficit is equal to the negative of the real-time difference, and this value is simultaneously recorded in the process variable register. When the real-time difference is greater than or equal to zero, the "Power Deficit Status" flag is set to false, and the value in the process variable register is cleared.

[0091] Simultaneously, the current carbon intensity data of the distribution network is acquired and compared with a preset carbon intensity threshold, achieved through a parallel processing mechanism. While calculating the real-time difference, the current carbon intensity data of the distribution network at the same timestamp is read from the time-series database. The preset carbon intensity threshold is stored in a configurable parameter table, and this threshold is dynamically adjusted according to the carbon emission level of the power grid, for example, set to 0.8 kg of CO2 per kilowatt-hour. The comparator compares the current carbon intensity data of the distribution network with the preset carbon intensity threshold. When the current carbon intensity data of the distribution network is greater than the preset carbon intensity threshold, a "high carbon state" flag is generated and set to true.

[0092] When the current value of the carbon intensity data of the distribution network is greater than the preset carbon intensity threshold and there is a power deficit, a carbon emission optimization instruction is generated, implemented using logical AND conditional judgment. The state detector receives a "high carbon state" flag and a "power deficit state" flag as input. When both flags are true, the carbon emission optimization instruction generator is triggered. The carbon emission optimization instruction is a data structure containing an instruction code and a timestamp, with the instruction code fixed as "CO_OPT_ACTIVE". The time deviation between the generation time and the data acquisition time does not exceed 50 milliseconds.

[0093] The carbon emission optimization command is associated with the current timestamp, specifically implemented through a time binding module. The timestamp generator obtains the current Coordinated Universal Time (UTC) from the BeiDou time system, accurate to the millisecond level. The data encapsulator combines the carbon emission optimization command and the current timestamp into a structured data packet, with the packet format defined as [Command Identifier: CO_OPT_ACTIVE, Timestamp: 2024-08-20T14:30:45.123Z]. This data packet is written to the command queue for further processing.

[0094] The preset carbon intensity threshold is determined based on the grid's carbon emission characteristics, for example, by referencing 1.2 times the grid's average carbon intensity over the past 30 days. The threshold update cycle is configurable, for example, automatically calculated and updated every 24 hours. Manual adjustment of the threshold is supported when the grid operating mode changes, with the adjustment range limited to between 0.5,000 and 1.5 kg of CO2 per kilowatt-hour. The threshold is stored in non-volatile memory, and the data is not lost after a power outage.

[0095] The real-time difference calculation process includes data validity checks. If any input data exceeds a reasonable range—for example, the output data of a distributed photovoltaic power generation unit exceeds 1.2 times its rated capacity—the calculation is paused and a data verification process is triggered. The reasonableness range of the data is set according to the equipment's technical parameters; for example, the output data of a wind power generation unit should not exceed 105% of its designed maximum output. Abnormal data is automatically replaced with the most recent valid value, and alarm events are recorded.

[0096] A jitter prevention mechanism is implemented in the power deficit determination process. A power deficit is only determined to exist when the real-time difference fluctuates near zero, for example, when it is less than zero twice out of three consecutive checks. The determination result is maintained for a minimum duration, such as at least one second, to avoid false judgments caused by instantaneous fluctuations. The jitter prevention parameters can be configured according to grid stability requirements.

[0097] The carbon emission optimization instruction generation process includes priority handling logic. When multiple instruction generation conditions are met simultaneously, the carbon emission optimization instruction has the highest execution priority. The instruction data packet includes a version identifier, such as version number V1.2, to ensure subsequent system compatibility. Instruction transmission employs a cyclic redundancy check mechanism with a check bit length of 16 bits.

[0098] The timestamp association process implements clock synchronization calibration. Every 60 seconds, the time is compared with the time source, and the local clock is automatically corrected when the deviation exceeds 10 milliseconds. The timestamp format strictly adheres to the ISO 8601 standard and includes a time zone identifier. A timestamp index is created during data storage, supporting fast retrieval of instruction records by time range.

[0099] The exception handling mechanism covers the entire process: when the real-time difference calculation times out (e.g., it takes more than 100 milliseconds to complete), the result of the previous period is forcibly used; when carbon intensity data acquisition fails, a sliding window average is temporarily used as a substitute; when timestamp synchronization fails, a local high-precision crystal oscillator is enabled to maintain the timing. All exception events are logged in detail and system alarms are generated.

[0100] Data storage employs a dual-backup mechanism, with carbon emission optimization instructions simultaneously written to both an in-memory database and solid-state storage. The in-memory database stores instruction records for the most recent 24 hours, while the solid-state storage stores 90 days of historical data. Data backup is performed hourly, and the backup process does not affect real-time instruction generation. Access to the data interface is subject to access control, allowing only authorized terminals to query instruction history.

[0101] The frequency of instruction generation is controlled by the system timer, with a minimum interval of 200 milliseconds to prevent high-frequency instructions from impacting subsequent system operations. When multiple carbon emission optimization instructions are generated simultaneously, only the latest instruction is retained, and historical instructions are automatically marked as invalid. Invalid instructions can still be queried but will no longer participate in control logic operations.

[0102] S3. When generating carbon emission optimization instructions, monitor the real-time congestion status and carbon emission ramp-up rate of key sections of the distribution network, and generate high-risk carbon transfer period markers. The specific implementation is as follows:

[0103] Real-time congestion status of key sections of the distribution network is obtained from the power grid dispatch automation system, specifically by accessing the monitoring interface of the dispatch automation system via a power data acquisition protocol. This monitoring interface outputs power flow data values ​​for the key sections twice per second, with the data unit being megawatts. A dedicated communication channel is established during acquisition, using a request-response model to send section identification codes and receive structured data packets containing timestamps and power flow values. After parsing the data packets, the power flow value field is extracted, and the data acquisition time point is recorded.

[0104] After real-time power flow data of key sections is collected through the monitoring interface of the dispatch automation system, it is immediately compared with the section's safe transmission limits. The section's safe transmission limits are stored in a local parameter database, which is automatically synchronized daily with the latest limit data released by the power grid dispatching department. The comparison process performs a numerical comparison, arithmetically comparing the absolute value of the power flow data with the corresponding section's safe transmission limit. The limit unit is megawatts, consistent with the unit of the power flow data.

[0105] When the absolute value of the power flow data exceeds the safe transmission limit, the real-time blocking state is determined to be blocked, a status identifier "Blocked" is generated, and stored in the status register. If the absolute value of the power flow data is less than or equal to the safe transmission limit, the real-time blocking state is determined to be non-blocked, and a status identifier "Unblocked" is generated. The determination result is stored in conjunction with the data acquisition time point, and the time synchronization accuracy is controlled within 50 milliseconds.

[0106] To retrieve the carbon emission ramp-up rate of standby generator units from the generator set characteristic database, the first step is to obtain a list of standby generator unit numbers specified in the grid dispatch command. The generator unit numbers use a 12-digit numerical encoding rule, for example, "370201001002". Based on this list, a query request is initiated to the generator set characteristic database. The query command includes the generator unit number and parameter type identifier.

[0107] Extract the carbon emission coefficient per unit power change corresponding to the unit number, specifically by accessing the unit's static parameter table in the database. The static parameter table stores the approved carbon emission intensity per unit of power generation for each unit, determined through unit type testing. For example, the coefficient for a coal-fired unit is 0.85 kg of carbon dioxide per kilowatt-hour. The data is temporarily stored in a cache after being read.

[0108] The maximum power ramp rate of the unit under its current output status is read by accessing the unit's operating status table in real time. This table records the unit's current output percentage and its corresponding maximum ramp capacity. For example, a unit's maximum ramp rate at 80% output is 9 MW per minute. The ramp rate value is dynamically updated based on the unit's real-time operating conditions.

[0109] Multiply the carbon emission coefficient per unit power change by the maximum power ramp rate to obtain the carbon emission ramp rate. Perform floating-point multiplication. For example, to multiply 0.85 kg of carbon dioxide per kilowatt-hour by 9 megawatts per minute, first convert the units and then calculate: 0.85 × (9 × 1000 / 60) = 127.5 kg of carbon dioxide per minute. Save the calculation result as a floating-point variable.

[0110] When the real-time congestion status is "blocked" and the carbon emission ramp-up rate exceeds a preset ramp-up rate threshold, the current period is determined to be a high-risk period for carbon transfer. The logic judgment unit receives the real-time congestion status identifier and the carbon emission ramp-up rate value. When the status identifier is "blocked" and the ramp-up rate value is greater than the preset threshold, it outputs a high-risk period determination signal. The preset threshold is set according to the grid regulation capacity, for example, 100 kg of carbon dioxide per minute.

[0111] A high-risk carbon transfer period marker is generated and bound to the current timestamp. The marker uses a binary encoding format. A marker value of 1 indicates that the high-risk period is active, and 0 indicates that it is not active. After receiving the high-risk period determination signal, the marker generator immediately obtains the current Coordinated Universal Time from the time synchronization system and generates a data record in the format [marker value: 1, timestamp: 2024-08-20T14:35:12.456Z].

[0112] The high-risk carbon transfer period is associated with the timestamp of the carbon emission optimization instruction, which is stored through a time alignment module. The carbon emission optimization instruction queue generated in step S2 is read, and the instruction record closest to the current time is found. When the time deviation is less than 200 milliseconds, the high-risk carbon transfer period is marked and written to an extended field of that instruction record. This is stored in a dedicated partition of the distributed database.

[0113] The cross-sectional safety transmission limit update mechanism includes version management. Each synchronization operation generates a new version number, such as V2.3.1. Historical versions are retained for 30 days for audit traceability. When the received limit data is abnormal, such as a negative value or exceeding 10,000 MW, it automatically reverts to the previous valid version and issues an alarm.

[0114] The unit characteristic database query implements timeout control. The maximum waiting time for a single query is set to 300 milliseconds, after which it will automatically retry. If it fails three times consecutively, it will switch to the backup database node. Query results are subject to range validation; for example, a carbon emission coefficient per unit power change exceeding 2.0 kg of carbon dioxide per kilowatt-hour is considered invalid.

[0115] The calculation of carbon emission ramp rate includes unit consistency processing. When the carbon emission factor per unit power change is in kilograms of carbon dioxide per kilowatt, and the maximum power ramp rate is in megawatts per minute, unit conversion is required: first convert the ramp rate to kilowatts per minute (multiply by 1000), and then multiply by the factor value to obtain the amount of kilograms of carbon dioxide per minute.

[0116] A mechanism to prevent false triggering is set up for high-risk period determination. An activation flag is only generated when the blocked state persists for more than a set duration, such as 10 seconds, and the number of times the ramp rate exceeds the limit reaches a threshold proportion (e.g., exceeding the limit 4 out of 5 times). The determination parameters can be adjusted through the configuration interface.

[0117] The timestamp binding process implements clock drift compensation. Every 60 seconds, the deviation between the local clock and the time source is detected. When the accumulated deviation exceeds 5 milliseconds, a compensation value is added to the timestamp. The millisecond portion of the timestamp is fixed to three decimal places, for example, ".456".

[0118] Data storage employs a dual-channel write mechanism. The primary channel writes to a real-time database for immediate access, while the backup channel writes to a blockchain-based evidence storage system for auditing and traceability. The blockchain evidence storage system generates a hash value for every 10 records, ensuring data immutability.

[0119] Anomaly handling covers the entire process: when cross-sectional power flow data is invalid, the most recent valid value is used as a replacement and a data quality identifier is added; when the unit ramp rate is missing, the default value of the same type of unit is used; when the timestamp deviation exceeds the limit, the manual time synchronization process is initiated. All abnormal events generate a unique tracking code and are stored in the log system.

[0120] The high-risk period tagged storage structure includes lifecycle management. Active tags are refreshed every 5 seconds, and automatically invalidated if there is no refresh for more than 60 seconds. Storage partitions implement an automatic cleanup mechanism; invalidated tags are retained for 24 hours before being transferred to archive storage. Archive data compression is set to 50% to save space.

[0121] The communication security mechanism includes transmission encryption and authentication. Access to the dispatch automation system uses two-way digital certificate authentication, with certificates automatically renewed every 90 days. Database query channels implement field-level encryption, and sensitive parameters such as unit numbers are transmitted using the AES-256 algorithm. Access logs fully record the operator's identity and the operation time.

[0122] S4. Based on carbon intensity data and high-risk periods for carbon transfer, calculate the carbon emission transfer factor corresponding to load adjustment behavior. The specific implementation is as follows:

[0123] The current value of the carbon intensity data for the distribution network is obtained by reading the latest timestamp-corresponding field value of the grid carbon intensity data from the time-series database established in S1. The read operation is performed every 200 milliseconds, prioritizing data records with a time deviation of less than 50 milliseconds. When multiple records with similar times exist, the data with the closest timestamp is selected as the current value. The data unit is fixed at kilograms of carbon dioxide per kilowatt-hour, with a value range limited to 0.1 to 1.5. If the value exceeds this range, a data verification process is initiated.

[0124] The status value of markers indicating high-risk periods for carbon transfer is read by accessing the marker storage area generated by S3. The marker storage area uses a key-value pair data structure, where the key is a timestamp (accurate to milliseconds) and the value is a Boolean status identifier. During a query, the marker record with the smallest time difference is searched based on the current system time. The status value returns 0 or 1, where 1 represents an active state and 0 represents an inactive state. The status retrieval latency is controlled within 10 milliseconds.

[0125] When a high-risk period for carbon transfer is marked as active, the current value of the distribution network's carbon intensity data is corrected using a preset weighting factor. The weighting factor is stored in a configurable parameter table, and its value is set according to the severity of grid congestion. For example, 1.1 is used for mild congestion, and 1.3 is used for severe congestion. The correction calculation performs floating-point multiplication: Correction value = Original carbon intensity × Weighting factor. The calculation result is rounded to three decimal places; for example, the original value of 0.85 multiplied by 1.2 equals 1.020.

[0126] Multiplying the corrected carbon intensity data by the unit load adjustment amount yields the carbon emission transfer factor corresponding to the load adjustment behavior. The unit load adjustment amount is defined as the change in load power of 1 MW, and this value is a constant. The multiplication operation is performed using double-precision floating-point calculation: Carbon emission transfer factor = Corrected carbon intensity × 1. The result is in kilograms of carbon dioxide per megawatt; for example, 1.020 × 1 = 1.020 kilograms of carbon dioxide per megawatt.

[0127] When a high-risk period for carbon transfer is marked as inactive, the current value of the distribution network's carbon intensity data is multiplied by the unit load adjustment amount to directly obtain the carbon emission transfer factor corresponding to the load adjustment behavior. The calculation process is independent of the active state path. For example, when the current carbon intensity value is 0.82, the carbon emission transfer factor = 0.82 × 1 = 0.82 kg of carbon dioxide per megawatt. The calculation result is stored in a dedicated field in the real-time database.

[0128] The weighting factor is set based on historical grid congestion data analysis. A mapping relationship is established between congestion level and correction coefficient: when the congestion duration exceeds 10 minutes, the weighting factor increases by 0.1; when the congestion power exceeds the limit by 20%, the weighting factor increases by 0.15. The maximum cumulative correction coefficient does not exceed 1.5. The parameter table is automatically updated weekly and supports manual adjustment.

[0129] The baseline value for unit load regulation is set at 1 MW, which corresponds to the power regulation capacity of typical load regulation equipment. Proportional coefficients can be configured for specific load types; for example, electric vehicle charging stations are calculated at 0.8, and air conditioning group control systems at 1.2. These proportional coefficients are stored in the equipment feature library and are associated with the load equipment identification code when retrieved.

[0130] The calculation process is monitored in real time: when the carbon emission transfer factor calculation result exceeds a preset threshold (e.g., 2.0 kg CO2 per megawatt), a high carbon emission alarm is triggered and the event log is recorded. The threshold is set according to the peak carbon emission intensity of the region and supports dynamic adjustment from 1.0 to 3.0 kg CO2 per megawatt.

[0131] The state transition process includes a debouncing mechanism. A path transition is only executed if the high-risk flag state remains stable for a set duration (e.g., 5 seconds). Instantaneous state changes do not trigger changes to the calculation path, avoiding frequent transitions that could cause result oscillations. The state retention time parameter can be configured online.

[0132] Data storage employs a hierarchical structure: raw carbon intensity values ​​are stored in the base layer, corrected values ​​in the processing layer, and the final carbon emission transfer factor in the result layer. Data across all layers is linked via a unified timestamp, supporting end-to-end traceability. Storage periods are configured as follows: 30 days for the base layer, 7 days for the processing layer, and 365 days for the result layer.

[0133] The anomaly handling mechanism includes: when carbon intensity data is missing, the moving average of the three most recent valid values ​​is used as a substitute; when the weighting factor parameter is abnormal, it automatically reverts to the default value of 1.0; when the flag state is invalid, it is forced to be processed as inactive. All abnormal events generate a unique error code and record the occurrence time.

[0134] Data smoothing is performed before outputting the calculation results. A sliding window mean filter is used, with a window size of 5 consecutive sampling points. For example, if the results of 5 consecutive calculations are [1.02, 1.03, 1.05, 1.02, 1.04], the final output value is 1.032. The filtering parameters can be adjusted according to the load type.

[0135] The security mechanism includes data integrity verification. A 32-bit cyclic redundancy check (CRC) code is generated and appended to the result record after each calculation. A recalculation is triggered if the check fails; after three consecutive failures, the output is frozen and a fault is reported. The computing service runs in a secure container environment, and memory access is protected by boundaries.

[0136] Version management ensures traceability: each parameter update generates a new version number, such as CF_VER2.3. Calculation result records include the input data version and the calculation program version. Historical version data is retained for 180 days, and rollback queries by time range are supported.

[0137] S5. Combining the carbon emission transfer factor and the requirements for distributed power generation, construct carbon-energy synergistic constraints, specifically implemented as follows:

[0138] The system retrieves the carbon emission transfer factor corresponding to load adjustment behavior, specifically reading the latest timestamped carbon emission transfer factor value from the S4 calculation result storage area. A time alignment mechanism is used during reading, selecting valid records with a time deviation of less than 100 milliseconds based on the current system time. When multiple load types have carbon emission transfer factors, they are extracted according to the controllable load equipment identification code; for example, the factor value of 1.25 kg CO2 per megawatt is extracted from the air conditioning group control system. The data acquisition frequency is synchronized with the load adjustment cycle, defaulting to reading once every 10 seconds.

[0139] The system reads the real-time absorbable power from the distributed power generation output data and calculates it through integration. It retrieves the output data of distributed photovoltaic (PV) power generation units and wind power generation units within the park, spanning the previous 5 minutes from the S1 time-series database. The power output value per minute is integrated over time: Real-time absorbable power = Σ(PV output + Wind power output) × 1 minute, with the result in megawatt-hours (MWh). For example, if the output value for 5 consecutive minutes is [2.1, 2.3, 2.0, 2.2, 2.4] MWh, then the real-time absorbable power = (2.1 + 2.3 + 2.0 + 2.2 + 2.4) × 1 / 60 = 0.183 MWh.

[0140] Establish a total load regulation constraint: the controllable load regulation amount shall not exceed the sum of the real-time absorbable electricity and the power deficit. The power deficit value is taken from the absolute value data in the S2 process variable register. The constraint expression is: Σ controllable load regulation amount ≤ (real-time absorbable electricity + |power deficit|). The unit of controllable load regulation amount is megawatts (MW), and the real-time absorbable electricity needs to be converted to megawatts (multiplied by 60), for example, 0.183 MWh × 60 = 10.98 MW. The upper limit of the constraint is stored in the constraint condition register with two decimal places.

[0141] Establish carbon emission constraints: The total carbon emission transfer resulting from all controllable load adjustments must not exceed a preset carbon emission threshold. The formula for calculating the total carbon emission transfer is: Σ(single type of controllable load adjustment amount × corresponding carbon emission transfer factor). The preset carbon emission threshold is set based on historical emission peaks, for example, 80% of the maximum daily emissions over the most recent 30 days. The constraint expression is: Σ(adjustment amount × factor) ≤ preset carbon emission threshold, with the unit uniformly expressed as kilograms of carbon dioxide.

[0142] The total load regulation constraint and carbon emission constraint are combined into a carbon-energy synergistic constraint, encapsulated using structured data. The combination process generates a data package containing the following fields: constraint identifier, total load regulation upper limit, total carbon emission upper limit, and effective timestamp. For example, the data package format is: [Constraint ID:CT_001, Load Limit: 15.6MW, Carbon Emission Limit: 1200kgCO2, Timestamp: 2024-08-20T14:40:00.000Z]. The data package is written to the constraint library and marked as the currently valid version.

[0143] The real-time absorbable power calculation employs a sliding time window mechanism. The window size is configurable, with a default of 5 minutes and an allowable range of 1 to 15 minutes. The integral calculation uses the trapezoidal rule: the output value for a given minute is (starting value + ending value) / 2. For example, if the output is 2.1 MW in the first minute and 2.3 MW in the second minute, then the contribution for that minute is (2.1 + 2.3) / 2 × 1 / 60 = 0.0367 MWh. The window sliding step is fixed at 1 minute.

[0144] Power deficit handling involves unit conversion. The power deficit unit in the S2 register is megawatts (MW), while the real-time absorbable power unit is megawatt-hours (MWh). Before combining them, the units must be unified: multiply the real-time absorbable power by 60 to convert it into an equivalent megawatt value (based on a 5-minute time window). For example, 10 MWh × 60 = 600 MW equivalent value, which, when added to the 50 MW power deficit, gives an upper limit of 650 MW.

[0145] The preset carbon emission threshold dynamic adjustment mechanism includes three sources: a base value of 120% of the average daily emissions of the previous month, a floating value adjusted according to the daily carbon intensity change rate (e.g., if carbon intensity increases by 10%, the threshold is lowered by 5%), and a manual correction value that allows maintenance personnel to adjust within a range of ±20%. The final threshold is the weighted average of the three values, with a weighting configuration of 6:3:1.

[0146] The constraint combination process implements logical verification. When the upper limit of the total load adjustment is less than 10% of the current controllable total load power, it is automatically relaxed to 20%; when the upper limit of carbon emission constraints is lower than 100 kg of carbon dioxide, it is forcibly raised to the basic threshold. The verification rules avoid overly strict constraints that could lead to an infeasible solution.

[0147] The data encapsulation format is clearly defined: the upper limit field for total load regulation is named "load_limit" and its data type is a floating-point number (unit: megawatt); the upper limit field for total carbon emissions is named "carbon_limit" and its data type is an integer (unit: kilogram of carbon dioxide); the timestamp field uses the ISO 8601 format and is accurate to milliseconds. The data packet size is fixed at 128 bytes.

[0148] The anomaly handling mechanism includes: automatically resetting the calculated real-time absorbable power to zero and recording the anomaly event when it is negative; using the average value of the most recent 5 minutes as a substitute when power deficit data is invalid; and calculating based on the maximum factor value of the same type of load when the carbon emission transfer factor is missing. All abnormal operations generate audit logs.

[0149] Version management ensures traceability: each constraint update generates a sequence number, such as CT_VER0052. Historical versions are retained for the most recent 100 records, and queries by time range are supported. The currently active version is maintained with dual replicas in memory, automatically switching in case of primary replica failure.

[0150] The security mechanism includes data signing: each constraint data packet is appended with a 256-bit elliptic curve digital signature, and the verification key is stored in a hardware security module. If signature verification fails, the data packet is discarded and an alarm is triggered. Access to the constraint library is subject to role-based access control, and modification operations require two-factor authentication.

[0151] The parameter adjustment interface provides a web-based configuration page: the adjustment range for the total load regulation constraint is limited to 0 to 150% of the current total load of the park; the carbon emission constraint threshold is set between 500,000 and 5,000 kg of carbon dioxide; the time window parameter is set via a slider control (1 to 15 minutes). All modifications require secondary confirmation to take effect.

[0152] The constraint's effective period is synchronized with the power grid dispatching period. It remains effective for 30 minutes by default, and the next cycle calculation begins 5 minutes before its expiration. In special circumstances, immediate refresh is supported, with a minimum refresh interval of 3 minutes to avoid control oscillations caused by frequent changes.

[0153] S6. Generate energy storage charging and discharging power commands and controllable load adjustment commands based on carbon-energy synergistic constraints and issue them to the park's energy management platform for execution. The specific implementation is as follows:

[0154] The system invokes the load regulation and carbon emission constraints from the carbon-energy coordinated constraints, specifically reading the latest effective version of the data package from the S5 constraint library. During the data package parsing process, key fields are extracted: the upper limit of the load regulation is identified as "load_limit", and the upper limit of the carbon emission is identified as "carbon_limit". For example, when parsing the data package [constraint ID:CT_001, load_limit:15.6, carbon_limit:1200], the upper limit of load is 15.6 MW and the upper limit of carbon emission is 1200 kg of carbon dioxide. The invocation frequency is synchronized with the constraint update cycle, and it is executed once every 30 minutes by default.

[0155] The total controllable load regulation is determined based on the total load regulation constraint, using a direct mapping method of constraint values. The total load regulation is equal to the upper limit value in the total load regulation constraint, which serves as the baseline value for the total regulation. For example, when the "load_limit" field value is 15.6 MW, the total controllable load regulation is determined to be 15.6 MW. This value is written to the total allocation register, with precision retained to two decimal places.

[0156] The priority order of controllable load adjustments is determined based on carbon emission constraints using a sorting algorithm. First, a list of carbon emission transfer factors for all controllable load devices is retrieved from the S4 storage area. The sorting rule is based on ascending order of carbon emission transfer factors: devices with smaller factor values ​​have higher priority. For example, device A has a factor of 0.82, device B has a factor of 1.15, and device C has a factor of 1.03, generating a priority sequence [A, C, B]. Devices with the same factor value are then sorted a second time based on their response speed, with shorter response times given priority.

[0157] Controllable load adjustment instructions are generated based on the total controllable load adjustment amount and priority order, using a sequential allocation strategy. The adjustment amount is allocated starting with the highest priority device, using the following formula: Single device allocation amount = min(device adjustable capacity, remaining total capacity). During the allocation process, the remaining total capacity is updated in real time: Total base value - Allocated amount. For example, if the total capacity is 15.6 MW, device A with an adjustable capacity of 8 MW will be allocated 8 MW, leaving 7.6 MW remaining; device C with an adjustable capacity of 5 MW will be allocated 5 MW, leaving 2.6 MW remaining; and device B with an adjustable capacity of 3 MW will be allocated 2.6 MW. The generated instruction format is [Device ID: Power Value], such as [A:8.0,C:5.0,B:2.6].

[0158] The system combines the current state of charge (SOC) of the energy storage system with the maximum charge / discharge power to generate charge / discharge power commands, performing multi-condition judgments. First, it reads the current SOC percentage from the energy storage monitoring system, which is collected in real-time by the battery management system. The charge / discharge logic includes three modes: when the SOC is below 20%, a charging command is generated, with power value = min(power deficit × 0.5, maximum charging power); when the SOC is above 80%, a discharging command is generated, with power value = min(power deficit × 0.7, maximum discharging power); when the SOC is between 20% and 80%, the power value = power deficit × 0.6 and is limited to the charge / discharge power range. For example, with an SOC of 75%, a power deficit of 10 MW, and a maximum discharging power of 8 MW, a discharging command of 7 MW is generated (min(10 × 0.7, 8) = 7).

[0159] Controllable load adjustment commands and energy storage charging / discharging power commands are linked with timestamps and sent to the park's energy management platform via a data encapsulation protocol. The timestamps are obtained from the BeiDou time synchronization system with millisecond accuracy. The command encapsulation format is defined as: {Timestamp: "2024-08-20T14:45:30.500Z", Load Command: [{A:8.0},{C:5.0},{B:2.6}], Energy Storage Command: {Mode:"Discharge", Power: 7.0}}. The transmission channel uses a message queue, with each message appended with a 16-bit cyclic redundancy check code.

[0160] The priority ranking is dynamically adjusted by weight. Based on the carbon emission transfer factor ranking, a weight is introduced for equipment adjustability: equipment with an adjustable capacity greater than 5 MW has a weight coefficient of 1.2, and equipment with an adjustable capacity less than 1 MW has a coefficient of 0.8. The final priority score = factor value × weight coefficient, and equipment is ranked in ascending order of score. For example, equipment B has a factor of 1.15 × weight 1.2 = 1.38, and equipment C has a factor of 1.03 × 1 = 1.03; therefore, the order is adjusted so that C takes precedence over B.

[0161] Energy storage command generation sets protection strategies: discharging is prohibited when the state of charge is below 15%, and charging is prohibited when it is above 90%; the rate of change of charging and discharging power is limited to 5% of the rated power per minute; each mode switch requires an interval of at least 10 minutes. Protection parameters are stored in the safety configuration table and can be modified online.

[0162] Data compression is implemented during the instruction encapsulation process. The load instruction list uses differential encoding: the first device records the complete ID and power value, while subsequent devices only record the incremental power value. For example, [A:8.0,C:+5.0,B:+2.6] is encoded as "A:8.0,C:5.0,B:2.6" and compressed to "0A8.0C5.0B2.6". The compression rate can reach 30%, reducing transmission bandwidth usage.

[0163] The delivery mechanism includes reliability guarantees: the message queue implements an acknowledgment and retransmission mechanism, and the receiver must return an ACK signal within 200 milliseconds; if no acknowledgment is received three times consecutively, the backup transmission channel will be switched; all delivery commands are cached locally for 24 hours and support breakpoint resumption.

[0164] Total allocation is followed by redistribution of surplus capacity. After the initial allocation, if the remaining total capacity is greater than the minimum controllable load adjustment unit (e.g., 0.1 MW), a secondary allocation is performed according to priority: Single device supplementary amount = min(remaining adjustable capacity of device, remaining total capacity). The secondary allocation instruction is marked as a supplementary instruction and is bound to the same timestamp as the main instruction.

[0165] Timestamp synchronization employs phase-locked loop (PLL) technology. The command generation time T0 is compared with the time source; if the deviation exceeds 5 milliseconds, a compensation value Δt is added to the timestamp. For example, if the local clock is detected to be 3 milliseconds fast, the timestamp is recorded as T0-0.003. The compensation value is calibrated hourly.

[0166] Anomaly handling covers key scenarios: When state of charge data fails, it forces entry into safe mode (charging / discharging power ≤ 20% of rated value); when a load device is offline, it automatically skips the device and reallocates the total power; when the sending channel is interrupted, it enables the SMS backup channel to send simplified instructions. All abnormal operations are recorded in a detailed event log.

[0167] The security mechanism includes digital signatures: each instruction is signed using the sender's private key to generate an ECDSA signature, which is then verified by the receiver using a pre-set public key. Instructions failing to sign are immediately invalidated and trigger a security audit. Key pairs are rotated monthly, with old keys retained for 30 days for historical data verification.

[0168] Command lifecycle management: The default validity period of issued commands is 30 minutes. Commands that are not executed within the time limit will automatically expire. Commands that have been executed are marked as completed and moved to the history database. Conflict command handling follows the time priority principle. If a later command conflicts with an unexecuted command, it will be automatically rejected.

[0169] Example 2: Figure 2 A schematic diagram of a multi-source collaborative carbon emission optimization control system for industrial parks is provided. This multi-source collaborative carbon emission optimization control system for industrial parks includes the following modules:

[0170] The multi-source acquisition module is used to collect carbon intensity data of the power distribution network, output data of distributed power sources in the park, and load demand data in real time.

[0171] The optimization decision module is used to generate carbon emission optimization instructions when the real-time difference between distributed power generation output data and load demand data indicates a power deficit and the grid carbon intensity is higher than a preset threshold.

[0172] The congestion monitoring module is used to monitor the real-time congestion status and carbon emission ramp-up rate of key sections of the distribution network when carbon emission optimization instructions are generated, and to generate high-risk carbon transfer period markers.

[0173] The factor calculation module is used to calculate the carbon emission transfer factor corresponding to load adjustment behavior based on carbon intensity data and high-risk carbon transfer periods.

[0174] The constraint construction module is used to construct carbon-energy synergistic constraints by combining the carbon emission transfer factor and the requirements for distributed power generation.

[0175] The instruction execution module is used to generate energy storage charging and discharging power instructions and controllable load adjustment instructions based on carbon-energy synergy constraints and send them to the park's energy management platform for execution.

[0176] All calculations involved in the embodiments are dimensionless numerical calculations, and the preset parameters and thresholds in the calculations are set by those skilled in the art according to the actual situation.

[0177] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0178] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and inventive constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0179] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0180] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0181] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0182] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A multi-source collaborative method for optimizing and controlling carbon emissions in industrial parks, characterized in that, Includes the following steps: S1. Real-time collection of carbon intensity data of the power distribution network, output data of distributed power sources in the park, and load demand data; S2. When the real-time difference between the distributed power generation output data and the load demand data indicates that there is a power deficit and the grid carbon intensity is higher than the preset threshold, a carbon emission optimization instruction is generated. S3. When generating carbon emission optimization instructions, monitor the real-time blockage status and carbon emission ramp-up rate of key sections of the distribution network, and generate high-risk carbon transfer period markers. S4. Based on carbon intensity data and high-risk periods for carbon transfer, calculate the carbon emission transfer factor corresponding to load adjustment behavior; S5. Combining the carbon emission transfer factor and the requirements for distributed power consumption, construct carbon-energy synergistic constraints; S6. Generate energy storage charging and discharging power commands and controllable load adjustment commands based on carbon-energy synergy constraints and send them to the park's energy management platform for execution.

2. The multi-source collaborative carbon emission optimization and control method for industrial parks according to claim 1, characterized in that, Real-time collection of carbon intensity data from the power distribution network, distributed power generation output data within the industrial park, and load demand data, including: Carbon intensity data of the distribution network is obtained in real time from the power dispatch data platform; The output data of distributed photovoltaic power generation units and wind power generation units in the park are collected by smart meters. The load monitoring device collects real-time power data of controllable loads and real-time power data of uncontrollable loads from the load demand data. The carbon intensity data of the distribution network, the output data of distributed photovoltaic power generation units, the output data of wind power generation units in the park, the real-time power data of controllable loads and the real-time power data of uncontrollable loads are bound together and stored with a unified timestamp.

3. The multi-source collaborative carbon emission optimization and control method for industrial parks according to claim 2, characterized in that, When the real-time difference between distributed generation output data and load demand data indicates a power deficit and the grid carbon intensity exceeds a preset threshold, a carbon emission optimization instruction is generated, including: Calculate the real-time difference between the sum of distributed power generation output data and the sum of load demand data; When the real-time difference is less than zero, it is determined that there is a power deficit; Simultaneously, the current value of carbon intensity data of the distribution network is obtained and compared with the preset carbon intensity threshold; When the current value of the carbon intensity data of the distribution network is greater than the preset carbon intensity threshold and there is a power deficit, a carbon emission optimization instruction is generated. Associate the carbon emission optimization instruction with the current timestamp.

4. The multi-source synergistic carbon emission optimization and control method for industrial parks according to claim 3, characterized in that, When generating carbon emission optimization instructions, the real-time congestion status and carbon emission ramp-up rate of key sections of the distribution network are monitored to generate high-risk carbon transfer periods, including: Obtain the real-time congestion status of key sections of the distribution network from the power grid dispatch automation system; Read the carbon emission ramp rate of the standby unit from the generator set characteristic database; When the real-time blocking status is blocked and the carbon emission ramp rate is greater than the preset ramp rate threshold, the current period is determined to be a high-risk period for carbon transfer. Generate a marker for a high-risk period of carbon transfer that is linked to the current timestamp; The timestamps of carbon emission optimization instructions are stored to mark high-risk periods of carbon transfer.

5. The multi-source collaborative carbon emission optimization and control method for industrial parks according to claim 4, characterized in that, Obtain the real-time congestion status of key sections of the distribution network from the power grid dispatch automation system, including: Power flow data of key sections are collected in real time through the SCADA interface of the scheduling automation system; Compare the current flow data with the cross-sectional safe transmission limits; When the absolute value of the power flow data exceeds the safe transmission limit, the real-time blocking state is determined to be a blocking state. Otherwise, the real-time blocking state is determined to be a non-blocking state.

6. The multi-source synergistic carbon emission optimization and control method for industrial parks according to claim 4, characterized in that, Read the carbon emission ramp rate of the standby unit from the generator set characteristic database, including: The database is queried based on the standby unit number determined by the power grid dispatching instructions; Extract the carbon emission coefficient per unit power change corresponding to the unit number; Read the maximum power ramp rate of the unit under its current output status; Multiplying the carbon emission coefficient per unit power change by the maximum power ramp rate yields the carbon emission ramp rate.

7. The multi-source synergistic carbon emission optimization and control method for industrial parks according to claim 4, characterized in that, Based on carbon intensity data and markers of high-risk periods for carbon transfer, the carbon emission transfer factor corresponding to load adjustment behavior is calculated, including: Obtain the current value of carbon intensity data for the distribution network; read the status value marked during high-risk periods of carbon transfer; When a high-risk period for carbon transfer is marked as active, the current value of the carbon intensity data of the distribution network is corrected using a preset weighting factor; Multiply the corrected carbon intensity data by the unit load adjustment amount to obtain the carbon emission transfer factor corresponding to the load adjustment behavior; When a high-risk period for carbon transfer is marked as inactive, the current value of the carbon intensity data of the distribution network is multiplied by the unit load adjustment amount to obtain the carbon emission transfer factor corresponding to the load adjustment behavior.

8. The multi-source collaborative carbon emission optimization and control method for industrial parks according to claim 7, characterized in that, Combining the carbon emission transfer factor and the requirements for distributed power generation, carbon-energy synergistic constraints are constructed, including: Obtain the carbon emission transfer factor corresponding to load adjustment behavior; Read the real-time absorbable power from the output data of distributed power sources; Establish a total load regulation constraint: the controllable load regulation amount shall not exceed the sum of the real-time absorbable electricity and the power deficit; Establish carbon emission constraints: The total amount of carbon emissions transferred by all controllable load adjustment behaviors shall not exceed the preset carbon emission threshold; The total load regulation constraint and carbon emission constraint are combined into a carbon-energy synergistic constraint condition.

9. The multi-source collaborative carbon emission optimization and control method for industrial parks according to claim 8, characterized in that, Based on the carbon-energy synergistic constraints, energy storage charging and discharging power commands and controllable load adjustment commands are generated and sent to the park's energy management platform for execution, including: Invoke the total load regulation constraint and carbon emission constraint in the carbon-energy synergistic constraint conditions; Determine the total controllable load regulation amount based on the total load regulation constraint; Determine the priority order of controllable load adjustment based on carbon emission constraints; Controllable load adjustment instructions are generated based on the total controllable load adjustment amount and priority order; Generate energy storage charging and discharging power commands by combining the current state of charge of the energy storage with the maximum charging and discharging power. The controllable load adjustment command and the energy storage charging and discharging power command are linked with timestamps and sent to the park's energy management platform.

10. A multi-source collaborative carbon emission optimization control system for industrial parks, used to implement the multi-source collaborative carbon emission optimization control method for industrial parks as described in any one of claims 1-9, characterized in that, Includes the following modules: The multi-source acquisition module is used to collect carbon intensity data of the power distribution network, output data of distributed power sources in the park, and load demand data in real time. The optimization decision module is used to generate carbon emission optimization instructions when the real-time difference between distributed power generation output data and load demand data indicates a power deficit and the grid carbon intensity is higher than a preset threshold. The congestion monitoring module is used to monitor the real-time congestion status and carbon emission ramp-up rate of key sections of the distribution network when carbon emission optimization instructions are generated, and to generate high-risk carbon transfer period markers. The factor calculation module is used to calculate the carbon emission transfer factor corresponding to load adjustment behavior based on carbon intensity data and high-risk carbon transfer periods. The constraint construction module is used to construct carbon-energy synergistic constraints by combining the carbon emission transfer factor and the requirements for distributed power generation. The instruction execution module is used to generate energy storage charging and discharging power instructions and controllable load adjustment instructions based on carbon-energy synergy constraints and send them to the park's energy management platform for execution.

Images