Zero-carbon park source-grid-load-storage integrated collaborative scheduling method and system thereof

Abstract

This application discloses a method and system for integrated coordinated scheduling of energy sources, grid, load, and storage in a zero-carbon industrial park. The method includes: collecting real-time power generation of distributed photovoltaic systems, real-time operating current and voltage of permanent magnet motor clusters, and real-time state of charge of energy storage systems, and acquiring real-time electricity price signals and real-time carbon price signals; constructing a multi-objective optimization model based on real-time carbon price, the park's real-time remaining carbon allowance, real-time electricity price, and the collected data, with the park's comprehensive energy cost, carbon emissions, and green electricity consumption rate as optimization objectives; solving the model to generate a charging and discharging plan for the energy storage system, an operating strategy for the permanent magnet motor cluster, and power setting values ​​for the interaction between the microgrid and the external power grid, which serve as integrated coordinated scheduling instructions; issuing and executing the instructions for the current period; collecting status feedback data after execution; recalculating the park's real-time remaining carbon allowance based on the status feedback data, and returning to the data collection step in the next scheduling cycle.

Description

Technical Field

[0001] This application relates to the field of energy technology for zero-carbon industrial parks, specifically to a method and system for integrated coordinated scheduling of energy sources, grids, loads, and storage in zero-carbon industrial parks. Background Technology

[0002] Zero-carbon industrial parks, as crucial platforms for achieving "dual-carbon" goals, rely on the coordinated operation of distributed photovoltaic power generation on the source side, high-efficiency loads on the load side, energy storage systems on the storage side, and the park's microgrid to optimize both energy efficiency and low carbon emissions. Currently, in park energy management practices, distributed photovoltaic systems, energy storage systems, permanent magnet motor clusters, and microgrids are often managed by independent control systems, leading to obstacles in data exchange and command coordination between these systems. This relatively decentralized management approach makes it difficult for the park to adjust the operating strategies of each unit in a timely manner to achieve overall optimization when faced with real-time changes in electricity prices. Consequently, the green electricity absorption capacity and the economic efficiency of system operation still need improvement.

[0003] With the gradual maturation of the carbon trading market, incorporating carbon price signals into industrial park scheduling decisions has become a research hotspot. Existing research has attempted to include carbon trading prices in scheduling objectives by constructing multi-objective optimization models that balance energy costs and carbon emissions. However, most of these methods treat carbon allowances as static annual total indicators for ex-post calculation, failing to dynamically integrate the real-time carbon allowance consumption of the industrial park into the scheduling decision-making process. Furthermore, the adjustment potential of high-efficiency loads such as permanent magnet motors has not yet been correlated with the real-time remaining amount of carbon allowances, resulting in a certain degree of synergistic optimization space between energy cost control and carbon emission management in industrial parks. How to dynamically integrate the real-time remaining carbon allowances of industrial parks into scheduling decisions and deeply couple them with the energy efficiency regulation of permanent magnet motors has become a pressing technical problem to be solved in this field. Summary of the Invention

[0004] This application provides a method and system for integrated coordinated scheduling of source, grid, load and storage in zero-carbon industrial parks, which is used to at least solve the problems existing in the prior art.

[0005] A first aspect of this application provides a method for integrated coordinated scheduling of energy sources, grid, load, and storage in a zero-carbon industrial park, comprising the following steps: Step S1, collecting real-time power generation of distributed photovoltaic power on the source side, real-time operating current and voltage of permanent magnet motor clusters on the load side, and real-time state of charge of energy storage systems on the storage side, and obtaining real-time electricity price signals from the external power grid and real-time carbon price signals from the carbon trading market; Step S2, constructing a multi-objective optimization model based on the real-time carbon price signal, the park's real-time remaining carbon allowance, the real-time electricity price signal, and the real-time power generation, operating current, operating voltage, and state of charge, with the park's comprehensive energy consumption cost, park carbon emissions, and green electricity consumption rate as optimization objectives; Step S3, solving the multi-objective optimization model to generate the energy storage system for future periods. The integrated collaborative scheduling instructions are: 1) The charging and discharging power timing plan, the start / stop control commands and load adjustment commands of the permanent magnet motor cluster, and the interaction power setting values ​​between the park microgrid and the external power grid; 2) The instructions corresponding to the current time period in the integrated collaborative scheduling instructions are sent to the controller of the energy storage system, the inverter of the permanent magnet motor cluster, and the grid-connected interface controller of the park microgrid for execution; 3) The actual charging and discharging power returned by the controller of the energy storage system, the actual operating power returned by the inverter of the permanent magnet motor cluster, and the actual interaction power returned by the grid-connected interface controller are collected as status feedback data; 4) The remaining carbon quota in the park in real time is recalculated based on the status feedback data, and the data collection step S1 is returned when the next scheduling cycle arrives.

[0006] By incorporating real-time carbon price signals and the park's real-time remaining carbon allowances into the scheduling decision-making process, the scheduling model can dynamically respond to changes in the carbon trading market and the real-time consumption of the park's carbon allowances. This embodiment integrates carbon allowances as a real-time constraint into the optimization model, enabling the generated scheduling instructions to proactively adjust the operating strategies of each unit based on the remaining carbon allowances. Simultaneously, by collecting status feedback data and recalculating the real-time remaining carbon allowances, a rolling scheduling closed loop with dynamic carbon allowance updates is formed, allowing scheduling decisions to track the carbon allowance consumption progress in real time and avoiding the problem of carbon allowance management lagging behind actual energy consumption. Furthermore, this embodiment includes the operating parameters of permanent magnet motors in the data collection scope, creating conditions for establishing a correlation between their adjustment potential and the real-time remaining carbon allowances, which helps improve the synergy between energy cost control and carbon emission management.

[0007] Furthermore, the multi-objective optimization model constructed in step S2 includes a carbon quota constraint, which is used to limit the total predicted carbon emissions of the park in the future period to not exceed the sum of the park's real-time remaining carbon quota and the quota expected to be offset through carbon credits.

[0008] By setting carbon quota constraints, the park's real-time remaining carbon quotas are incorporated into the optimization model in the form of mathematical constraints, allowing the scheduling solution process to be directly affected by the remaining carbon quota amount. When the park has a relatively abundant real-time remaining carbon quota, the scheduling model has a relatively high tolerance for carbon emissions; when carbon quotas become scarce, the constraints force the optimization algorithm to seek scheduling schemes with lower carbon emissions. The expected amount of quotas offset through carbon credits provides the park with additional carbon emission buffer space, giving the carbon quota constraints a certain degree of flexibility and preventing excessive regulation due to short-term carbon quota shortages from affecting the park's normal production.

[0009] Further, step S3 generates start / stop control commands and load adjustment commands for the permanent magnet motor cluster, including: performing real-time energy efficiency assessment on individual permanent magnet motors in the cluster based on real-time operating current and real-time operating voltage to obtain the real-time energy efficiency value of each permanent magnet motor; comparing the real-time energy efficiency value with a preset energy efficiency threshold to identify inefficient operating motors whose real-time energy efficiency value is lower than the energy efficiency threshold; and generating shutdown commands or load reduction commands for inefficient operating motors based on the identification results of inefficient operating motors, real-time electricity price signals, real-time carbon price signals, and the remaining carbon allowances in the park.

[0010] By conducting real-time energy efficiency assessments of permanent magnet motors, scheduling decisions can obtain the real-time energy efficiency status of each motor, thereby correlating energy efficiency data with electricity price signals, carbon price signals, and carbon quota constraints. When the park's carbon quota becomes strained, the scheduling model can prioritize identifying and regulating motors with lower energy efficiency, minimizing the impact on the normal operation of high-efficiency motors while meeting carbon quota constraints. This approach establishes a direct link between the regulation potential of permanent magnet motors and the real-time consumption of carbon quotas, making energy efficiency data a crucial basis for scheduling decisions.

[0011] Furthermore, when generating shutdown or load reduction commands based on the identification results of inefficient operating motors, the shutdown or load reduction operation is prioritized for the motor with the lowest real-time energy efficiency value among the inefficient operating motors, until the predicted total carbon emissions meet the carbon quota constraints.

[0012] When it is necessary to regulate inefficient motors, the regulation is performed sequentially from lowest to highest energy efficiency value, ensuring that the regulation operation achieves the expected carbon emission reduction effect with minimal impact on production. Motors with the lowest real-time energy efficiency value have relatively higher carbon emissions per unit of output; prioritizing the regulation of these motors can reduce interference with normal production in the industrial park while achieving the same carbon emission reduction. This sorting and regulation method makes the scheduling strategy under carbon quota constraints more refined and operable.

[0013] Furthermore, in step S3, a particle swarm optimization algorithm is used to solve the multi-objective optimization model, including: encoding the charging and discharging power of the energy storage system at each time node in the future period, the total operating power of the permanent magnet motor cluster at each time node in the future period, and the interaction power between the park microgrid and the external power grid at each time node in the future period into particle position vectors; using the optimization objective of the multi-objective optimization model as the evaluation basis for particle fitness, and searching for the globally optimal particle position that maximizes fitness by iteratively updating the particle velocity and position; decoding the globally optimal particle position to obtain the charging and discharging power timing plan, start-stop control commands and load adjustment commands, and interaction power setpoints.

[0014] The multi-objective optimization model is solved using particle swarm optimization, enabling collaborative optimization of the energy storage system's charging and discharging schedule, permanent magnet motor operation strategy, and microgrid interaction power within a unified solution framework. Decision variables for multiple scheduling objects are encoded as particle positions, allowing the optimization process to simultaneously consider the interactions and constraints between units, resulting in globally coordinated integrated scheduling instructions. An iterative search mechanism enables the algorithm to find feasible solutions that satisfy multi-objective requirements under complex constraints.

[0015] Furthermore, the real-time operating current and real-time operating voltage collected in step S1 also include the dispatchable capacity signal uploaded by the new energy vehicles connected to the park microgrid through the V2G terminal; the integrated collaborative scheduling instruction generated in step S3 also includes the charging and discharging scheduling instruction for the new energy vehicles, which is used to control the new energy vehicles to charge from the park microgrid when the real-time electricity price signal is at a valley value, and to discharge to the park microgrid when the real-time electricity price signal is at a peak value.

[0016] By incorporating the dispatchable capacity of new energy vehicles into the data collection scope, the park's microgrid can utilize the energy storage resources of these vehicles to participate in energy dispatch. During off-peak electricity prices, vehicles are guided to charge, satisfying the charging needs of car owners and absorbing surplus photovoltaic power generated at night. During peak electricity prices, vehicles are guided to discharge, providing additional power support to the park and reducing the need to purchase electricity from the external grid at higher prices. This V2G two-way energy interaction model expands the scope of dispatchable resources in the park, making new energy vehicles an important component of the coordinated dispatch of power generation, grid, load, and storage.

[0017] Furthermore, the method also includes a virtual power plant aggregation and control step, which is performed after step S3 and before step S4: aggregating distributed photovoltaic, energy storage systems and permanent magnet motor clusters into a virtual power plant; responding to frequency regulation or peak shaving instructions issued by the grid dispatch system using the regulation capacity reserved for participating in grid ancillary services, under the premise of meeting the load demand within the park, according to the integrated collaborative dispatch instructions; and including the ancillary service revenue generated from responding to frequency regulation or peak shaving instructions in the comprehensive energy cost of the park.

[0018] By aggregating distributed resources within the industrial park into virtual power plants, the park can participate in the grid ancillary services market as a whole. A certain amount of regulation capacity is reserved when generating dispatch instructions, preserving operational space for responding to grid demands. Revenue generated from responding to ancillary service demands is included in energy costs, allowing the park to reflect the economic returns of participating in grid interaction in its energy cost accounting. This model opens up additional revenue channels for the park without affecting its normal production.

[0019] Furthermore, step S6 is followed by an adaptive adjustment step for weight coefficients: based on the deviation between the state feedback data and the instruction portion corresponding to the current time period in the integrated collaborative scheduling instruction, the actual energy cost, actual carbon emissions, and actual green electricity consumption rate of the current scheduling cycle are calculated; based on the closeness of the actual energy cost, actual carbon emissions, and actual green electricity consumption rate to the optimization objectives, the weight coefficients corresponding to each optimization objective in the multi-objective optimization model are corrected online; and the corrected weight coefficients are used for model solving in the next scheduling cycle.

[0020] By collecting status feedback data and comparing it with scheduling instructions, the deviation information between the actual operating effect and the expected goal is obtained. The weight coefficients are then adjusted online based on the actual results, enabling the optimization model to dynamically adjust the relative importance of each optimization objective based on historical operating data. When a certain objective deviates from the expected value for a long period, the adaptive adjustment of the weight coefficients can make subsequent scheduling focus more on optimizing that objective, improving the adaptability of the multi-objective optimization model to the actual operating environment.

[0021] Furthermore, the remaining carbon allowance in the park in step S2 is dynamically determined by subtracting the carbon emissions already generated up to the current time from the total carbon allowance initially allocated to the park based on real-time power generation and status feedback data.

[0022] By using real-time power generation and status feedback data to calculate generated carbon emissions, the park's remaining carbon allowances can be dynamically updated based on actual energy consumption. Unlike static annual carbon allowance management, this dynamic update mechanism allows dispatch decisions to monitor carbon allowance consumption in real time, providing data support for the accurate construction of carbon allowance constraints. Real-time power generation reflects green energy consumption, while status feedback data reflects actual energy consumption; the combination of these two factors makes carbon emission accounting more closely reflect the park's actual operating status.

[0023] Furthermore, step S5 is followed by a carbon emission reduction data generation step: based on real-time power generation and status feedback data, and according to a preset carbon accounting methodology, the carbon emission reduction generated by the park within the completed scheduling cycle is calculated; the carbon emission reduction is associated with the corresponding scheduling cycle timestamp and equipment identifier to generate carbon emission reduction data records.

[0024] By generating carbon emission reduction data records at the end of each scheduling cycle, the park's carbon emission reduction achievements can be stored and managed in a structured form. The association between carbon emission reductions and timestamps and equipment identifiers provides a data foundation for subsequent carbon asset ownership and traceability. This periodic carbon emission reduction data generation method, matched with the rolling scheduling mechanism, ensures that carbon asset management remains synchronized with scheduling operations.

[0025] Furthermore, step S6 includes a carbon quota overrun warning and emergency control step: based on the real-time carbon price signal and the park's real-time remaining carbon quota, it is predicted whether there is a risk of carbon quota overrun at the end of the current scheduling cycle; if there is a risk of overrun, an emergency control instruction is generated and issued for execution. The emergency control instruction includes at least one of the following: a first instruction to control the energy storage system to discharge at the maximum allowable power, a second instruction to control all permanent magnet motor clusters to operate in the highest energy efficiency mode, and a third instruction to control the park's microgrid to fully absorb the real-time power generation of distributed photovoltaics locally.

[0026] By predicting carbon allowance consumption for the next cycle after the scheduling cycle ends, the park can take countermeasures before the risk of exceeding carbon allowances arises. Emergency control commands are set separately for energy storage systems, permanent magnet motors, and photovoltaic (PV) integration, allowing the park to choose the most appropriate control method based on actual conditions. Maximum power discharge of energy storage systems can reduce carbon emissions from purchasing electricity from the external grid; operating permanent magnet motors in their highest efficiency mode can reduce carbon emissions per unit of output; and full integration of PV can increase the proportion of green electricity and reduce carbon emissions from fossil fuel substitution. This early warning and emergency response mechanism provides a safety net for the park's carbon emission management.

[0027] A second aspect of this application provides a zero-carbon industrial park integrated source-grid-load-storage coordinated scheduling system. This system includes: a data acquisition module, communicatively connected to the inverters of distributed photovoltaic systems, the frequency converters of permanent magnet motor clusters, and the controllers of energy storage systems, for acquiring real-time power generation, real-time operating current, real-time operating voltage, and real-time state of charge, and receiving real-time electricity price signals and real-time carbon price signals; an optimization solution module, communicatively connected to the data acquisition module, for constructing and solving a multi-objective optimization model based on the data acquired by the data acquisition module, and generating integrated coordinated scheduling instructions; an instruction execution module, communicatively connected to the optimization solution module and the controllers of the energy storage system, the frequency converters of the permanent magnet motor clusters, and the grid-connected interface controller of the industrial park microgrid, for issuing and executing the instruction portion corresponding to the current time period in the integrated coordinated scheduling instructions, and receiving returned status feedback data; and a carbon quota update module, communicatively connected to the instruction execution module and the optimization solution module, for recalculating the remaining real-time carbon quotas of the industrial park based on the status feedback data, and triggering the optimization solution module to re-solve the problem in each scheduling cycle.

[0028] By modularizing the functions corresponding to each step of the above method embodiments, the method can be deployed and run in actual hardware systems. The data acquisition module is responsible for interacting with field equipment to obtain real-time operating data and external signals; the optimization and solution module undertakes the core algorithm calculations and generates scheduling instructions; the instruction execution module is responsible for issuing instructions and collecting feedback; and the carbon quota update module realizes the dynamic update of carbon quotas and the triggering of scheduling cycles. The division of labor and cooperation among the modules enables the integrated source-grid-load-storage coordinated scheduling method to be implemented in actual parks, forming a complete closed loop from data acquisition to instruction execution to quota update. Attached Figure Description

[0029] Figure 1 A flowchart illustrating an integrated collaborative scheduling method for source-grid-load-storage systems in a zero-carbon industrial park, provided as an embodiment of this application; Figure 2 A schematic diagram of the framework of a zero-carbon industrial park source-grid-load-storage integrated collaborative scheduling system provided in this application embodiment; Figure 3 A schematic diagram illustrating the closed-loop process of carbon asset securitization and carbon benefit realization provided in this application embodiment; Figure 4 A schematic diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0030] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0031] In energy management practices within zero-carbon industrial parks, dispatch decisions typically revolve around optimizing energy costs, reducing electricity expenses through peak shaving and valley filling using energy storage systems and maximizing the utilization of distributed photovoltaic power. With the gradual maturation of the carbon trading market, park managers are beginning to focus on carbon emission control, making the incorporation of carbon price signals into dispatch targets a natural technological evolution. However, existing methods often treat carbon allowances as an annual total indicator for ex-post accounting—that is, calculating actual carbon emissions at the end of a settlement cycle and comparing them with the allocated carbon allowances. This ex-post accounting model prevents dispatch decisions from real-time monitoring of carbon allowance consumption; by the time insufficient carbon allowances are detected, the optimal adjustment opportunity has often been missed.

[0032] This application's in-depth analysis reveals that carbon allowance consumption is a dynamic, cumulative process closely related to the park's energy consumption behavior at every moment. If the remaining amount of carbon allowances can be tracked in real time during scheduling decisions and incorporated as a constraint into the optimization model, then scheduling instructions can proactively adjust the operating strategies of each unit based on the carbon allowance consumption progress. Simultaneously, as the park's main energy-consuming load, the operating efficiency of permanent magnet motors directly affects the carbon emissions per unit of output. If their energy efficiency status can be correlated with the real-time remaining amount of carbon allowances, lower-efficiency motors can be prioritized for control when carbon allowances are scarce, minimizing the impact on normal production while meeting carbon emission reduction requirements.

[0033] Based on the above analysis, this application proposes a technical approach that dynamically integrates real-time carbon quotas into scheduling decisions. By collecting real-time operational data from the source, grid, load, and storage sides, as well as external carbon price signals, a multi-objective optimization model incorporating carbon quota constraints is constructed, allowing the scheduling solution process to be directly influenced by the remaining carbon quota. After generating scheduling instructions, the remaining real-time carbon quota is recalculated by collecting execution feedback data, forming a rolling scheduling closed loop with dynamic carbon quota updates. This enables scheduling decisions to respond in real-time to changes in carbon quota consumption, transforming carbon quota management from ex-post accounting to ex-ante constraints and in-process control.

[0034] Please refer to Figure 1 , Figure 1 This application provides a method and system for integrated coordinated scheduling of source, grid, load and storage in a zero-carbon industrial park, comprising the following steps: Step S1: Collect real-time power generation of distributed photovoltaic power on the source side, real-time operating current and voltage of the permanent magnet motor cluster on the load side, and real-time state of charge of the energy storage system on the storage side, and obtain real-time electricity price signals from the external power grid and carbon price signals from the carbon trading market; Step S2: Based on the real-time carbon price signal, the remaining carbon quota in the park, the real-time electricity price signal, and the real-time power generation, operating current, operating voltage, and state of charge, construct a multi-objective optimization model with the park's comprehensive energy cost, carbon emissions, and green electricity consumption rate as optimization objectives; Step S3: Solve the multi-objective optimization model to generate a charging and discharging power timing plan for the energy storage system and the start-up and shutdown schedule of the permanent magnet motor cluster for future periods. Control commands, load regulation commands, and the power setting values ​​for interaction between the park's microgrid and the external power grid are used as integrated collaborative dispatch commands. Step S4: The command portion corresponding to the current time period in the integrated collaborative dispatch commands is sent to the controller of the energy storage system, the inverter of the permanent magnet motor cluster, and the grid-connected interface controller of the park's microgrid for execution. Step S5: The actual charging and discharging power returned by the controller of the energy storage system, the actual operating power returned by the inverter of the permanent magnet motor cluster, and the actual interactive power returned by the grid-connected interface controller are collected as status feedback data. Step S6: The remaining carbon quota in the park in real time is recalculated based on the status feedback data, and the data collection step S1 is returned when the next dispatch cycle arrives.

[0035] It is understandable that the real-time power generation of the source-side distributed photovoltaic (PV) system refers to the actual active power output of the PV system installed on building rooftops, parking lot rooftops, and other areas within the industrial park at the current moment. This value can be directly obtained through the power measurement unit built into the PV inverter, or it can be calculated by collecting and analyzing the output voltage and current of the PV modules. The unit of real-time power generation is usually kilowatts (kW) or megawatts (MW), and its value fluctuates dynamically with changes in irradiance, ambient temperature, and the operating status of the PV modules. For example, in an industrial park, a distributed PV system composed of polycrystalline silicon PV modules has a total installed capacity of 2 MW. During midday on a sunny day, the real-time power generation can reach approximately 1.8 MW. The real-time operating current and voltage of the load-side permanent magnet motor cluster refer to the actual current and voltage values ​​of the permanent magnet synchronous motors used to drive various mechanical equipment such as fans, pumps, and compressors within the park at the current moment. Compared to traditional asynchronous motors, permanent magnet motors have the characteristics of high efficiency and high power factor. Their operating current and voltage can be collected by sensors built into the motor controller or frequency converter. The unit of real-time operating current is Amperes (A), and the unit of real-time operating voltage is Volts (V). These two parameters can be used to calculate the real-time input power of the motor, and can also be combined with the motor's speed and load rate to further evaluate its operating efficiency. Taking a water pump motor in a certain industrial park as an example, its rated voltage is 380 volts, and the current is about 150 Amperes during normal operation. When the load changes, the current value will fluctuate within a certain range.

[0036] The real-time State of Charge (SOC) of an energy storage system refers to the percentage of the battery's current remaining capacity relative to its rated capacity. This parameter is crucial for the operation and control of the energy storage system, reflecting the current available power level. Real-time SOC can be directly read from the Battery Management System (BMS), and its value ranges from 0% to 100%, where 0% indicates a fully discharged battery and 100% indicates a fully charged battery. For example, a 1 MWh lithium iron phosphate energy storage system may reach 95% SOC after overnight charging, but this may drop to around 30% after daytime discharge. Real-time electricity price signals from the external power grid refer to the electricity market or grid company's information on current and future electricity purchase and sales prices. This signal can be obtained through a data interface with the grid dispatch system or subscribed to from an electricity trading platform. Real-time electricity price signals typically include time-of-use pricing information, such as peak-hour prices, off-peak prices, and flat-hour prices, expressed in yuan per kilowatt-hour. Taking the electricity price for industrial and commercial use in a certain region as an example, the peak electricity price may be 1.2 yuan per kilowatt-hour, the off-peak electricity price may be 0.3 yuan per kilowatt-hour, and the normal electricity price may be 0.7 yuan per kilowatt-hour.

[0037] The real-time carbon price signal in the carbon trading market refers to the current transaction price of carbon emission allowances or certified emission reductions in the carbon trading market. This signal can be obtained through a data interface with the carbon trading platform and reflects the market's pricing level for carbon emission rights. The unit of the real-time carbon price signal is usually yuan per ton or yuan per kilogram, and its value fluctuates due to policy regulation, market supply and demand, and macroeconomic conditions. For example, in a certain region's carbon trading market, the real-time carbon price of carbon emission allowances may be 58 yuan per ton. The real-time remaining carbon allowances of a park refer to the amount of carbon allowances initially allocated to the park in the current carbon emission management cycle, minus the carbon emissions generated up to the current moment. This parameter reflects the carbon allowance space that the park can continue to emit in the future and is a key constraint to consider in scheduling decisions. The unit of carbon allowances is usually tons or kilograms, and its value gradually decreases as the park's energy consumption activities continue. It is understood that if the park obtains additional offset allowances through the carbon inclusion mechanism, these additional allowances can also be included in the total available carbon allowances.

[0038] A multi-objective optimization model is a mathematical optimization model characterized by the need to simultaneously consider multiple conflicting optimization objectives and seek a balance among them. In the technical solution of this application, the optimization objectives of the multi-objective optimization model include minimizing the overall energy cost of the industrial park, minimizing the carbon emissions of the industrial park, and maximizing the green electricity consumption rate. The overall energy cost of the industrial park refers to the total cost of purchasing electricity from the external grid, minus potential revenue from electricity sales and ancillary services; the carbon emissions of the industrial park refer to the indirect carbon emissions corresponding to the purchase of electricity from the external grid; and the green electricity consumption rate refers to the proportion of distributed photovoltaic power generation actually consumed by the industrial park to the total photovoltaic power generation. These three objectives are somewhat correlated; for example, increasing green electricity consumption can simultaneously reduce energy costs and carbon emissions, but may be affected by fluctuations in photovoltaic output and limitations in energy storage capacity. The future time period refers to a preset time interval extending into the future from the current moment. The length of this time period can be set according to scheduling needs, for example, it can be set to the next 4 hours, the next 24 hours, or the next 48 hours. In the technical solution of this application, the future time period is used to plan the operation strategies of energy storage systems, permanent magnet motor clusters, and microgrids over a future period of time, making scheduling decisions forward-looking. For example, if the future time period is set to 4 hours and the scheduling cycle is 15 minutes, then the future time period contains 16 time nodes, each node corresponding to a 15-minute scheduling interval.

[0039] In the embodiments of this application, the construction of a multi-objective optimization model is the core link in realizing the integrated coordinated scheduling of power generation, grid, load, and storage. This model divides a future preset time period into T scheduling cycles, each cycle corresponding to a time node. It quantifies the three optimization objectives—comprehensive energy consumption cost of the industrial park, carbon emissions of the industrial park, and green electricity consumption rate—through mathematical expressions. The comprehensive energy consumption cost of the industrial park is represented as the sum of the products of the power purchased from the external grid Pgrid(t) and the corresponding real-time electricity price Priceelec(t) in each scheduling cycle, i.e., ΣPgrid(t)·Priceelec(t). The carbon emissions of the industrial park are represented as the sum of the products of the power purchased Pgrid(t), the grid carbon emission factor EFgrid, and the real-time carbon price Pricecarbon(t) in each scheduling cycle, ΣPgrid(t)·EFgrid·Pricecarbon(t). The green electricity consumption rate is characterized by the ratio of the actual photovoltaic power consumed in the park, Ppv,used(t), to the total photovoltaic power generated, Ppv,gen(t), i.e., ΣPpv,used(t) / ΣPpv,gen(t). The three optimization objectives are balanced using weight coefficients w1, w2, and w3, allowing the model to dynamically adjust the importance of each objective based on the actual needs of the park. The objective function of the multi-objective optimization model is specifically expressed as:

[0040] Where T represents the number of scheduling cycles included in the future preset time period, Pgrid(t) represents the power purchased by the park from the external grid in the t-th scheduling cycle, Priceelec(t) represents the real-time electricity price signal in the t-th scheduling cycle, EFgrid represents the carbon emission factor of the external grid, Pricecarbon(t) represents the real-time carbon price signal in the t-th scheduling cycle, Ppv,used(t) represents the actual photovoltaic power consumed by the park in the t-th scheduling cycle, Ppv,gen(t) represents the real-time power generation of distributed photovoltaics in the t-th scheduling cycle, and w1, w2, and w3 are the weighting coefficients corresponding to each objective. In this objective function, the first term, energy cost, and the second term, carbon cost, have the same unit of measurement and can be directly added together; the third term, green electricity consumption rate, is a dimensionless ratio, and its comprehensive balance with the cost term is achieved through weighting coefficients.

[0041] Based on the multi-objective optimization model, this application further introduces a carbon quota constraint, using the park's real-time remaining carbon quota as a hard boundary for scheduling decisions. The carbon quota constraint limits the park's predicted total carbon emissions over future periods to no more than the sum of the park's real-time remaining carbon quota and the quota expected to be offset through carbon credits. The total carbon emissions are calculated based on the purchased power Pgrid(t) and the grid carbon emission factor EFgrid for each scheduling cycle, i.e., the cumulative sum of carbon emission rates for each cycle. The carbon quota constraint is specifically expressed as follows: Qremaining represents the park's remaining carbon allowances, measured in kgCO2. It is dynamically determined by subtracting the carbon emissions already generated up to the current time, calculated based on real-time power generation and status feedback data, from the park's initially allocated total carbon allowances. Qoffset(T) represents the amount of allowances expected to be offset through carbon credits within a preset future period, also measured in kgCO2. This can be estimated based on the carbon emission reductions corresponding to available credits in the carbon credit pool, for example... ,in Let be the total available credits in the carbon credit pool, and f be the conversion function between credits and carbon emission reductions. This constraint allows the scheduling solution process to be influenced in real-time by the remaining carbon allowances. When the park has ample remaining carbon allowances, the constraint's effect on the scheduling scheme is relatively weak, allowing the optimization algorithm to focus more on reducing energy costs. When carbon allowances become scarce, the constraint actively narrows the range of feasible solutions, forcing the optimization algorithm to seek scheduling schemes with lower carbon emissions. This transforms carbon allowance management from post-event statistics to pre-event constraints and in-event control.

[0042] For solving the aforementioned multi-objective optimization model, this application employs a particle swarm optimization algorithm for iterative optimization. The particle swarm optimization algorithm searches for the optimal solution in the solution space by simulating the foraging behavior of bird flocks. In the solution process, the decision variables first need to be encoded as particle position vectors. The decision variables include the charging and discharging power Pstorage(t) of the energy storage system in each scheduling cycle within a future preset time period, the total operating power Pmotor(t) of the permanent magnet motor cluster in each scheduling cycle within the future time period, and the interaction power Pgrid(t) between the microgrid in the park and the external power grid. The position vector of each particle can be represented as: Wherein, Pstorage(t): the charging and discharging power of the energy storage system in the t-th scheduling cycle, with positive values ​​indicating discharging and negative values ​​indicating charging, in kW. Pmotor(t): the total operating power of the permanent magnet motor cluster in the t-th scheduling cycle, in kW. Pgrid(t): the interaction power between the park's microgrid and the external power grid in the t-th scheduling cycle, with positive values ​​indicating electricity purchase and negative values ​​indicating electricity sale, in kW.

[0043] The particle's velocity vector Vi represents the rate of change of position and is initially randomly generated. The fitness of each particle is evaluated based on the objective function value F of the multi-objective optimization model; a smaller F value indicates a better fitness. In each iteration, each particle records its own historical best position pbest, and the entire particle swarm records its global best position gbest. The particle's velocity and position are updated iteratively according to the following formula: Where ω is the inertia weight, controlling the degree to which the particle inherits the current velocity; c1 and c2 are learning factors, adjusting the step size of the particle's learning towards the individual optimum and the global optimum, respectively; r1 and r2 are random numbers in the interval [0,1], increasing the randomness of the search. During the iteration process, the values ​​of the decision variables also need to satisfy the power balance constraint: in For other unadjustable loads within the park, this balance constraint ensures the physical feasibility of the generated scheduling scheme. Through multiple iterations, the algorithm terminates when the preset maximum number of iterations is reached or the convergence condition is met, outputting the globally optimal particle position gbest. Decoding gbest yields the charging and discharging power timing plan of the energy storage system. Operation strategy of permanent magnet motor cluster and the interaction power setting between the park's microgrid and the external power grid. This forms a complete integrated collaborative scheduling command.

[0044] The charging and discharging power timing plan of an energy storage system refers to the planned sequence of charging and discharging power values ​​at various time points within a future period. A positive charging power indicates that the energy storage system is absorbing energy from the microgrid for charging, while a positive discharging power indicates that the energy storage system is releasing energy back to the microgrid. Charging and discharging cannot occur simultaneously. This timing plan is an important component of the scheduling instructions, guiding the energy storage system controller to perform charging and discharging operations according to the planned power. The start-stop control instructions and load regulation instructions for permanent magnet motor clusters refer to control instructions regarding the operating status and load level of each motor in the cluster. Start-stop control instructions are used to control the starting or stopping of motors, such as issuing a stop command to an inefficiently operating motor; load regulation instructions are used to adjust the operating load of the motors, such as adjusting the motor speed through a frequency converter to change its load rate. These instructions collectively constitute the operating strategy of the permanent magnet motor cluster.

[0045] The interaction power setpoint between the microgrid in the park and the external power grid refers to the planned active power value to be purchased from or sold to the external power grid per unit time at the connection point between the microgrid and the external power grid. A positive setpoint indicates power purchase from the external power grid, while a negative setpoint indicates power sale to the external power grid. The interaction power setpoint is a crucial parameter for coordinating the balance of energy sources, loads, and storage within the park and its relationship with the external power grid. Integrated coordinated dispatch instructions refer to a unified instruction set that integrates the charging and discharging power timing plans of the aforementioned energy storage system, the start-stop control instructions and load adjustment instructions of the permanent magnet motor cluster, and the interaction power setpoint between the microgrid in the park and the external power grid. This instruction set contains the complete operating strategies for each unit in the future time period, reflecting the coordinated relationship between the energy sources, grid, load, and storage. Status feedback data refers to the actual operating parameters returned by each device controller after executing the dispatch instructions, including the actual charging and discharging power returned by the energy storage system controller, the actual operating power returned by the inverter of the permanent magnet motor cluster, and the actual interaction power returned by the grid connection interface controller. This feedback data reflects the execution effect of the dispatch instructions and is an important basis for verifying the correctness of the instructions and updating the system status. The scheduling cycle refers to the time interval between two consecutive scheduling operations, which can be set to, for example, 15 minutes. At the beginning of each scheduling cycle, the system re-collects data and performs optimization solutions to generate new scheduling instructions. During each scheduling cycle, each device operates according to the instructions for the current time period. This periodic scheduling method enables the system to respond promptly to changes in its operating status.

[0046] This application's embodiments incorporate real-time carbon price signals and the park's real-time remaining carbon allowances into the scheduling decision-making process, enabling the scheduling model to dynamically respond to changes in the carbon trading market and the real-time consumption of the park's carbon allowances. This application's embodiments integrate carbon allowances as a real-time constraint into the optimization model, allowing the generated scheduling instructions to proactively adjust the operating strategies of each unit based on the remaining carbon allowances. Simultaneously, by collecting status feedback data and recalculating the real-time remaining carbon allowances, a rolling scheduling closed loop with dynamic carbon allowance updates is formed, enabling scheduling decisions to track the carbon allowance consumption progress in real time and avoiding the problem of carbon allowance management lagging behind actual energy consumption. Furthermore, this application's embodiments include the operating parameters of permanent magnet motors in the data collection scope, creating conditions for establishing a correlation between their adjustment potential and the real-time remaining carbon allowances, which helps improve the synergy between energy cost control and carbon emission management.

[0047] In some embodiments disclosed in this application, the multi-objective optimization model constructed in step S2 includes a carbon quota constraint condition, which is used to limit the total predicted carbon emissions of the park in the future period to not exceed the sum of the park's real-time remaining carbon quota and the quota expected to be offset through carbon credits.

[0048] Carbon quota constraints are mathematical restrictions that must be satisfied when constructing a multi-objective optimization model. Their role is to transform the real-time remaining carbon quota of the industrial park into a hard boundary in the optimization process. This constraint can be described mathematically, for example, limiting the cumulative carbon emissions at each time point within a future period to no more than a certain upper limit. It is understandable that the introduction of carbon quota constraints requires the optimization model to always treat the real-time remaining carbon quota as an insurmountable bottom line in the process of seeking the lowest energy cost, minimum carbon emissions, and highest green electricity consumption rate. The predicted total carbon emissions within the future period refer to the sum of indirect carbon emissions corresponding to the park's purchase of electricity from the external grid within a preset future time period, calculated based on the dispatch instructions generated during the optimization process. This predicted value is based on the expected results of dispatch decisions and may differ somewhat from the actual carbon emissions after execution. The calculation of the predicted total carbon emissions needs to consider the purchased power and the grid carbon emission factor at each time point; the higher the purchased power or the higher the grid carbon emission factor, the greater the predicted total carbon emissions.

[0049] The park's real-time remaining carbon allowances refer to the total initial carbon allowances allocated to the park during the current management cycle, minus the actual carbon emissions generated up to the current moment. This parameter gradually decreases as the park's energy activities continue, reflecting the carbon allowance space the park can continue to emit in the future. The amount of allowances expected to be offset through carbon credits refers to the amount of carbon allowances that are expected to be offset in the future using carbon credits accumulated by park employees through green behaviors. Carbon credits are an incentive tool set up by the park to encourage employees to engage in low-carbon behaviors such as green travel and energy-efficient office work. A certain number of credits corresponds to a certain amount of carbon emission reductions, which, after verification, can be used to offset the park's carbon emission allowances. Understandably, the introduction of expected offset allowances provides some flexibility to carbon allowance constraints, allowing the park to obtain additional emission allowances by activating carbon credits when carbon allowances are tight.

[0050] Carbon credits refer to the quantified points earned by park employees through low-carbon behaviors such as green cycling, charging new energy vehicles, and energy-efficient office practices. After verification by the platform, these credits can be converted into corresponding carbon emission reductions according to a preset conversion relationship, which can be used to offset the carbon quota gap of companies in the park. For example, an employee who accumulates 1,000 carbon credits through green commuting over a year can provide the park with 10 kilograms of carbon emission reduction capacity, based on the conversion relationship of 100 credits to 1 kilogram of carbon emission reduction.

[0051] This application embodiment incorporates the park's real-time remaining carbon allowances into the optimization model through carbon quota constraints, making the scheduling solution process directly affected by the remaining carbon allowances. When the park has ample real-time remaining carbon allowances, the scheduling model has a relatively high tolerance for carbon emissions; when carbon allowances become scarce, the constraints force the optimization algorithm to seek scheduling schemes with lower carbon emissions. The allowances expected to be offset through carbon credits provide the park with additional carbon emission buffer space, giving the carbon allowance constraints a certain degree of flexibility and preventing excessive regulation due to short-term carbon allowance shortages from affecting the park's normal production.

[0052] In some embodiments disclosed in this application, step S3, generating start / stop control commands and load adjustment commands for the permanent magnet motor cluster, includes: performing real-time energy efficiency assessment on individual permanent magnet motors in the permanent magnet motor cluster based on real-time operating current and real-time operating voltage to obtain the real-time energy efficiency value of each permanent magnet motor; comparing the real-time energy efficiency value with a preset energy efficiency threshold to identify inefficient operating motors whose real-time energy efficiency value is lower than the energy efficiency threshold; and generating shutdown commands or load reduction commands for inefficient operating motors based on the identification results of inefficient operating motors, real-time electricity price signals, real-time carbon price signals, and the remaining carbon allowances in the park in real-time.

[0053] As can be understood, real-time energy efficiency (REE) refers to the percentage of a permanent magnet motor's efficiency in converting input electrical energy into output mechanical energy at the current operating moment. This value is calculated by taking the motor's input power from the real-time collected operating current and voltage, then extrapolating the output power based on the motor's speed and load rate, and finally obtaining the REE as the ratio of output power to input power. REE reflects the motor's energy conversion efficiency under current operating conditions; a higher value indicates more economical motor operation and lower carbon emissions. For example, a permanent magnet motor with a rated REE of 95% might see its REE drop to around 88% when the load rate decreases to 30%.

[0054] The energy efficiency threshold is a reference standard value used to judge whether a permanent magnet motor is operating well. This threshold can be preset based on the motor's rated energy efficiency parameters, industry standards, or historical operating data from the industrial park. Its purpose is to distinguish between high-efficiency and low-efficiency motors. Understandably, the setting of the energy efficiency threshold needs to consider the differences in characteristics between different types of motors. For example, for newly commissioned high-efficiency permanent magnet motors, the energy efficiency threshold can be set at around 94%; for motors with a longer service life, the energy efficiency threshold can be appropriately relaxed to around 90%.

[0055] Inefficient motors refer to permanent magnet motors whose real-time energy efficiency value is lower than the preset energy efficiency threshold. These motors consume more electrical energy per unit of output during operation, resulting in relatively higher carbon emissions, thus possessing significant potential for energy conservation and carbon reduction. Identifying inefficient motors is a prerequisite for subsequent control operations. By accurately locating equipment with low energy efficiency, control resources can be focused on the most valuable areas for optimization.

[0056] A shutdown command is a control signal that stops the permanent magnet motor from operating. When an inefficiently running motor has a minor impact on production or the carbon emission cost of its continued operation is too high, a shutdown command can be issued to it, causing it to exit operation and reducing the overall carbon emissions of the industrial park. A load reduction command is a control signal that reduces the operating load of a permanent magnet motor. Unlike a shutdown command, a load reduction command does not change the motor's operating state, but rather adjusts the motor's speed or output power through equipment such as frequency converters, allowing it to operate at a lower load level. Load reduction commands can reduce motor energy consumption and carbon emissions while meeting some production needs, making it a gentler control method than shutdown.

[0057] This application embodiment assesses the energy efficiency of permanent magnet motors in real time, enabling scheduling decisions to obtain the real-time energy efficiency status of each motor. This allows for the correlation of energy efficiency data with electricity price signals, carbon price signals, and carbon quota constraints. When carbon quotas in the industrial park become strained, the scheduling model can prioritize identifying and regulating motors with lower energy efficiency, minimizing the impact on the normal operation of high-efficiency motors while meeting carbon quota constraints. This approach establishes a direct link between the regulation potential of permanent magnet motors and the real-time consumption of carbon quotas, making energy efficiency data a crucial basis for scheduling decisions.

[0058] In some embodiments disclosed in this application, when generating a shutdown command or load reduction command based on the identification result of inefficient operating motors, the shutdown or load reduction operation is performed on the motor with the lowest real-time energy efficiency value among the inefficient operating motors first, until the predicted total carbon emissions meet the carbon quota constraint conditions.

[0059] It's understandable that the motor with the lowest real-time energy efficiency value refers to the motor that ranks first among the inefficient operating motors, sorted from lowest to highest real-time energy efficiency value. The lower the energy efficiency value, the more electrical energy the motor consumes per unit of output, and the higher its corresponding carbon emissions. For example, in a certain industrial park with three inefficient motors, their real-time energy efficiency values ​​are 85%, 88%, and 90%, respectively. The motor with an energy efficiency value of 85% is the motor with the lowest real-time energy efficiency value. Performing shutdown or load reduction operations refers to the specific control actions taken on the motor according to control commands. Shutdown operations are used to completely stop the motor's operation, suitable for scenarios with minimal impact on production or where rapid carbon emission reduction is urgently needed. Load reduction operations are used to reduce the motor's operating load, for example, by reducing the motor speed from its rated speed to 70% using a frequency converter, allowing it to continue operating at a lower load level. It's understandable that choosing between shutdown and load reduction requires a comprehensive consideration of the motor's importance to production and the need to reduce carbon emissions.

[0060] The projected total carbon emissions refer to the sum of carbon emissions expected to be generated by the park within a predetermined future time period, calculated based on the current dispatch plan. This figure is based on the anticipated results of dispatch instructions and is related to the power purchase capacity and grid carbon emission factor at each time point in the dispatch plan. The carbon quota constraint refers to the restriction that the projected total carbon emissions of the park in the future period cannot exceed the upper limit of available carbon quotas. The upper limit of available carbon quotas includes the park's real-time remaining carbon quotas and the quotas expected to be offset through carbon credits.

[0061] In this embodiment, when it is necessary to regulate inefficient motors, the regulation is performed sequentially from lowest to highest energy efficiency value. This ensures that the regulation operation achieves the expected carbon emission reduction effect with minimal impact on production. The motor with the lowest real-time energy efficiency value corresponds to a relatively high carbon emission per unit output. Prioritizing the regulation of these motors can reduce interference with normal production in the industrial park while achieving the same carbon emission reduction. This sorting and regulation method makes the scheduling strategy under carbon quota constraints more refined and operable.

[0062] In some embodiments disclosed in this application, step S3 uses a particle swarm optimization algorithm to solve the multi-objective optimization model, including: encoding the charging and discharging power of the energy storage system at each time node in the future period, the total operating power of the permanent magnet motor cluster at each time node in the future period, and the interaction power between the park microgrid and the external power grid at each time node in the future period into particle position vectors; using the optimization objective of the multi-objective optimization model as the evaluation basis for particle fitness, iteratively updating the velocity and position of the particles to search for the globally optimal particle position that maximizes fitness; decoding the globally optimal particle position to obtain the charging and discharging power timing plan, start-stop control commands and load adjustment commands, and interaction power setpoints.

[0063] It is understandable that Particle Swarm Optimization (PSO) is an intelligent optimization method that simulates the collective behavior of flocks of birds or schools of fish, searching for the optimal solution through a group of particles moving in the solution space. Each particle represents a candidate solution, has a current position and velocity, and adjusts its direction of motion based on its best position and the best position found by the entire swarm. It is also understandable that PSO is suitable for optimization problems with continuous variables, and can handle multiple objectives appropriately when solving multi-objective optimization models. The particle position vector is the data structure representing candidate solutions in the algorithm, composed of decision variables arranged in a fixed order. In this application, the particle position vector contains a sequence of values ​​for the charging and discharging power of the energy storage system at each time node in the future period, the total operating power of the permanent magnet motor cluster at each time node in the future period, and the interaction power between the microgrid in the park and the external power grid at each time node in the future period. For example, if the future period is divided into 16 time nodes, then the position vector needs to contain 16 charging and discharging power values, 16 total operating power values, and 16 interaction power values, for a total of 48 variables.

[0064] Each time node within a future period refers to a discrete moment or time interval into which a future period is evenly divided. Each node corresponds to a scheduling interval, used to represent the average power or state within that time interval. For example, if the scheduling cycle is 15 minutes and the future period is set to 4 hours, then the future period contains 16 time nodes, each node representing a 15-minute time interval. Charging and discharging power refers to the charging or discharging power value of the energy storage system at a certain time node. It can usually be represented by a single variable, with a positive value representing charging and a negative value representing discharging; alternatively, two variables can be used to represent charging and discharging power respectively, but they must not both be positive simultaneously.

[0065] Total operating power refers to the sum of the active power of all operating motors in the permanent magnet motor cluster at a certain time point. This value reflects the overall power load of the cluster, and its magnitude is affected by the motor start-stop status and load adjustment commands. Interaction power refers to the exchange power between the microgrid in the park and the external power grid at a certain time point. Positive values ​​indicate purchasing electricity from the external power grid, while negative values ​​indicate selling electricity to the external power grid. The magnitude of interaction power is jointly affected by internal load, photovoltaic output, and energy storage regulation. Particle fitness is a quantitative indicator for evaluating the quality of a particle's position. In this application, the optimization objective of the multi-objective optimization model is used as the basis for fitness evaluation, that is, energy cost, carbon emissions, and green electricity consumption rate are combined into a fitness value in a certain way. The better the fitness, the better the scheduling scheme. Iterative update is the core process of the particle swarm optimization algorithm. In each iteration, each particle calculates a new velocity based on its own historical best position and the group's historical best position, and then updates its current position based on the velocity. After multiple iterations, the particle swarm gradually gathers towards the optimal solution region.

[0066] The globally optimal particle position refers to the particle position with the best fitness found in the swarm after iteration, representing the best scheduling scheme obtained by the algorithm. Decoding is the process of restoring the particle position vector to the actual scheduling command. Based on the correspondence during encoding, the charging and discharging power of the energy storage system at each time node, the total operating power of the permanent magnet motor cluster at each time node, and the interactive power of the microgrid at each time node are extracted from the globally optimal particle position, forming the charging and discharging power timing plan, operating strategy, and interactive power setpoint, respectively.

[0067] This application employs a particle swarm optimization algorithm to solve a multi-objective optimization model, enabling collaborative optimization of the energy storage system's charging and discharging schedule, the permanent magnet motor's operating strategy, and the microgrid's interactive power within a unified solution framework. By encoding the decision variables of multiple scheduling objects as particle positions, the optimization process can simultaneously consider the mutual influence and constraints between units, resulting in a globally coordinated integrated scheduling instruction. An iterative search mechanism allows the algorithm to find feasible solutions that satisfy multi-objective requirements under complex constraints.

[0068] In some embodiments disclosed in this application, the real-time operating current and real-time operating voltage collected in step S1 also include the dispatchable capacity signal uploaded by the new energy vehicles connected to the park microgrid via the V2G terminal; the integrated collaborative scheduling instruction generated in step S3 also includes the charging and discharging scheduling instruction for the new energy vehicles, which is used to control the new energy vehicles to charge from the park microgrid when the real-time electricity price signal is at a valley value, and to discharge to the park microgrid when the real-time electricity price signal is at a peak value.

[0069] Understandably, the dispatchable capacity signal refers to the battery power information reported by new energy vehicles to the park's microgrid via V2G terminals, which can be used to participate in charge and discharge scheduling. V2G is an abbreviation for Vehicle-to-Grid, representing the technology that enables bidirectional energy flow between electric vehicles and the power grid. The dispatchable capacity signal typically includes information such as the vehicle's current remaining battery power, dischargeable battery power, and expected stay time. For example, a new energy logistics vehicle enters the park at 10:00 AM, reports a dispatchable capacity of 30 kWh via its V2G terminal, and expects to stay for 4 hours.

[0070] Charging and discharging scheduling commands are control signals used to regulate the direction and magnitude of energy exchange between new energy vehicles and the park's microgrid. These commands can instruct vehicles to charge from the microgrid during specific time periods or to discharge to the microgrid. The charging and discharging scheduling commands need to interact with the vehicle's V2G terminal, which controls the charging and discharging process of the vehicle's battery according to the commands. A real-time electricity price signal at its lowest point indicates that the current electricity price is at a relatively low level for the day, typically corresponding to off-peak hours at night or periods of surplus power generation from new energy sources. A signal at its highest point indicates that the current electricity price is at a relatively high level for the day, typically corresponding to peak electricity consumption periods in the morning and evening on weekdays.

[0071] This application's embodiments incorporate the dispatchable capacity of new energy vehicles into the data collection scope, enabling the park's microgrid to utilize the vehicles' energy storage resources for energy dispatch. During off-peak electricity prices, vehicles are guided to charge, satisfying owners' charging needs while also absorbing surplus photovoltaic power generated at night. During peak electricity prices, vehicles are guided to discharge, providing additional power support to the park and reducing the need to purchase electricity from the external grid at higher prices. This V2G bidirectional energy interaction mode expands the scope of dispatchable resources within the park, making new energy vehicles an important component of the coordinated dispatch of power generation, grid, load, and storage.

[0072] In some embodiments disclosed in this application, the method further includes a virtual power plant aggregation and control step, which is performed after step S3 and before step S4: aggregating distributed photovoltaic, energy storage systems and permanent magnet motor clusters into a virtual power plant; responding to frequency regulation or peak shaving instructions issued by the grid dispatch system using the regulation capacity reserved for participating in grid ancillary services, under the premise of meeting the load demand within the park, according to the integrated collaborative dispatch instructions; and including the ancillary service revenue generated from responding to the frequency regulation or peak shaving instructions into the comprehensive energy cost of the park.

[0073] Understandably, a virtual power plant is a technological model that aggregates distributed resources such as distributed power sources, energy storage systems, and adjustable loads into a unified whole through advanced control, metering, and communication technologies, enabling them to participate in grid operation and electricity market management. In this application, the virtual power plant is composed of distributed photovoltaic, energy storage systems, and permanent magnet motor clusters within a park, and can accept grid dispatch instructions and provide ancillary services.

[0074] Aggregation refers to integrating multiple independent distributed resources through a unified control platform, making them appear as a single, dispatchable power plant. Aggregated resources can leverage economies of scale to meet the grid's access requirements for regulation capacity. Regulation capacity refers to the power regulation capability reserved by the virtual power plant to respond to grid ancillary service demands. Regulation capacity can be upward regulation capacity (increasing generation or decreasing consumption) or downward regulation capacity (decreasing generation or increasing consumption). For example, an energy storage system reserves 200 kW of discharge capacity as upward regulation capacity, and a permanent magnet motor cluster reserves 100 kW of load reduction capacity as downward regulation capacity. Grid ancillary services refer to services provided to maintain the safe and stable operation of the power system and ensure power quality, mainly including frequency regulation, peak shaving, reserve, and reactive power regulation. Frequency regulation services are used to address power fluctuations on the order of seconds to minutes, maintaining grid frequency stability; peak shaving services are used to address hourly load changes, alleviating peak-valley pressure.

[0075] Frequency regulation commands are control signals issued by the power grid dispatching system to regulate power and maintain frequency stability. Frequency regulation commands typically require controlled resources to respond within seconds to tens of seconds, demanding high regulation precision. Peak shaving commands are control signals issued by the power grid dispatching system to alleviate load peak-valley differences. Peak shaving commands typically have response times on the order of minutes, with longer regulation durations. Ancillary service revenue refers to the economic return obtained by a virtual power plant after participating in the power grid ancillary service market and responding to frequency regulation or peak shaving commands. Revenue can be calculated based on regulation capacity, regulation volume, and regulation effect. The comprehensive energy cost of a power plant park refers to the total cost of electricity purchased from the external power grid, minus potential electricity sales revenue and ancillary service revenue. Including ancillary service revenue in energy costs reflects the economic contribution of participating in power grid interaction in cost accounting.

[0076] This application's embodiment aggregates distributed resources within the industrial park into virtual power plants, enabling the park to participate in the grid ancillary services market as a whole. A certain amount of regulation capacity is reserved when generating dispatch instructions, preserving operational space for responding to grid demands. Revenue generated from responding to ancillary service demands is included in energy costs, allowing the park to reflect the economic returns of participating in grid interaction in its energy cost accounting. This model opens up additional revenue channels for the park without affecting its normal production.

[0077] In some embodiments disclosed in this application, step S6 is followed by a weight coefficient adaptive adjustment step: based on the deviation between the state feedback data and the instruction portion corresponding to the current time period in the integrated collaborative scheduling instruction, the actual energy cost, actual carbon emissions, and actual green electricity consumption rate of the current scheduling cycle are calculated; based on the closeness of the actual energy cost, actual carbon emissions, and actual green electricity consumption rate to the optimization target, the weight coefficients corresponding to each optimization target in the multi-objective optimization model are corrected online; and the corrected weight coefficients are used for model solving in the next scheduling cycle.

[0078] Understandably, deviation refers to the difference between the status feedback data and the instruction portion of the integrated collaborative scheduling command corresponding to the current time period. Deviation can be expressed as a numerical difference, such as the difference between the actual energy storage discharge power and the commanded discharge power; it can also be expressed as a relative error. The calculation method for deviation can be selected according to actual needs, such as using absolute deviation or relative deviation. For example, if the energy storage discharge command is 500 kW and the actual discharge power is 485 kW, then the absolute deviation is -15 kW, and the relative deviation is -3%.

[0079] The actual energy cost for the current dispatch cycle refers to the energy expenses actually incurred by the park during the recently completed dispatch cycle, calculated based on the actual interactive power in the status feedback data and the real-time electricity price signal for the corresponding period. The calculation of the actual energy cost can consider the electricity purchase cost minus the electricity sales revenue. If the park participates in the ancillary services market, the confirmed ancillary service revenue can also be deducted. The actual carbon emissions for the current dispatch cycle refer to the indirect carbon emissions actually generated by the park during the recently completed dispatch cycle, calculated based on the actual interactive power in the status feedback data and the grid carbon emission factor. The grid carbon emission factor refers to the carbon dioxide emissions corresponding to a unit of purchased electricity, usually published periodically by the power grid company or environmental protection department. For example, if the park's actual electricity purchase in a certain dispatch cycle is 1000 kWh and the grid carbon emission factor is 0.5 kg / kWh, then the actual carbon emissions are 500 kg. The actual green electricity consumption rate for the current dispatch cycle refers to the proportion of photovoltaic electricity actually consumed by the park during the recently completed dispatch cycle to the total photovoltaic power generation, calculated based on the actual photovoltaic consumption in the status feedback data and the real-time power generation. The actual photovoltaic (PV) power absorption capacity can be obtained by integrating the PV power actually flowing to the load in the park's microgrid. The actual green electricity absorption rate can be calculated by dividing the actual PV absorption capacity by the total PV power generation and then multiplying by 100%.

[0080] Proximity refers to the degree of conformity between actual energy costs, actual carbon emissions, and actual green energy consumption rates and the optimization targets. Proximity can be measured by the difference between actual and target values; a smaller difference indicates better proximity. For example, if the expected energy cost is 1000 yuan and the actual energy cost is 1050 yuan, the difference is 50 yuan, which can be considered a good degree of proximity. Online correction refers to the process of dynamically adjusting weight coefficients based on actual feedback data during system operation. Online correction can be performed after each scheduling cycle or every few cycles. The basis for correction is the degree of proximity between the actual operating results and the optimization targets. When the actual performance of a certain target deviates significantly from the expectation, its corresponding weight coefficient is appropriately adjusted.

[0081] This application embodiment obtains deviation information between the actual operating effect and the expected goal by collecting status feedback data and comparing it with scheduling instructions. The weight coefficients are then adjusted online based on the actual results, enabling the optimization model to dynamically adjust the relative importance of each optimization objective based on historical operating data. When a certain objective deviates from the expected value for a long period, the adaptive adjustment of the weight coefficients can make subsequent scheduling focus more on optimizing that objective, improving the adaptability of the multi-objective optimization model to the actual operating environment.

[0082] In some embodiments disclosed in this application, the remaining carbon allowance in the park in real time in step S2 is dynamically determined by subtracting the carbon emissions already generated up to the current time from the total amount of carbon allowance initially allocated to the park based on real-time power generation and status feedback data.

[0083] It is understood that the initial carbon allowance allocated to a park refers to the carbon emission quota that is verified and allocated to the park by the carbon trading authority or relevant institutions at the beginning of the current carbon emission management cycle, based on the park's historical emission levels, industry benchmarks, or policy requirements. This total amount is usually measured in tons or kilograms and represents the upper limit of the total amount of carbon dioxide that the park can legally emit throughout the entire management cycle. The initial allowance can be allocated free of charge or obtained through auction, depending on the rules of the carbon trading market.

[0084] Carbon emissions already generated refer to the cumulative actual carbon emissions calculated based on real-time power generation and status feedback data within the completed dispatch cycles up to the current moment. Calculating carbon emissions already generated requires summing the actual carbon emissions for each completed dispatch cycle. The calculation of actual carbon emissions is based on the actual interactive power in the status feedback data, i.e., the actual power purchased by the park from the external power grid, converted using the grid's carbon emission factor. It is understood that carbon emissions already generated do not include expected future carbon emissions; they only reflect actual carbon emissions that have already occurred.

[0085] Dynamic determination means that the remaining carbon allowances in a park are not calculated once and fixed, but are continuously updated as energy consumption activities in the park progress. Each time a scheduling cycle is completed and new status feedback data is obtained, the system recalculates the remaining carbon allowances based on the increased carbon emissions, ensuring it remains up-to-date. This dynamic determination method allows scheduling decisions to have real-time access to available carbon allowances. For example, a park initially allocates 10,000 tons of carbon allowances for the year. After the first quarter, 3,000 tons of carbon emissions have been generated, resulting in a remaining carbon allowance of 7,000 tons. After the second quarter, another 2,500 tons of carbon emissions are generated, updating the remaining carbon allowance to 4,500 tons.

[0086] This application's embodiments utilize real-time power generation and status feedback data to calculate generated carbon emissions, enabling the park's remaining carbon allowances to be dynamically updated based on actual energy consumption. Unlike static annual carbon allowance management, the dynamic update mechanism allows scheduling decisions to monitor carbon allowance consumption in real time, providing data support for the accurate construction of carbon allowance constraints. Real-time power generation reflects green electricity consumption, while status feedback data reflects actual energy consumption; the combination of both makes carbon emission calculation more closely reflect the park's actual operating status.

[0087] In some embodiments disclosed in this application, step S5 is followed by a carbon emission reduction data generation step: based on real-time power generation and status feedback data, and according to a preset carbon accounting methodology, the carbon emission reduction generated by the park within the completed scheduling cycle is calculated; the carbon emission reduction is associated with the corresponding scheduling cycle timestamp and equipment identifier to generate a carbon emission reduction data record.

[0088] It is understandable that the carbon emission reduction data generation step refers to the process of calculating the carbon emission reductions generated by the park during each scheduling cycle and forming data records based on the collected real-time power generation and status feedback data. This step is usually arranged after step S5, that is, executed immediately after obtaining the equipment execution feedback data. Carbon accounting methodology refers to the standardized calculation methods used to calculate carbon emission reductions. Carbon accounting methodology can be issued by government authorities or developed by industry organizations or international institutions, such as the voluntary greenhouse gas emission reduction methodology issued by the National Development and Reform Commission or the ISO 14064 series of standards issued by the International Organization for Standardization. Carbon accounting methodology typically specifies the method for determining the baseline scenario, the calculation method for project emissions, the accounting requirements for leakage, and the formula for calculating emission reductions. This application can select an applicable accounting methodology based on the carbon trading rules of the region where the park is located.

[0089] Carbon emission reduction refers to the reduction in carbon dioxide emissions in a park compared to a baseline scenario, achieved through measures such as operating distributed photovoltaic power and optimizing energy consumption. Carbon emission reduction is typically calculated using the baseline method, which is the baseline emissions minus the project emissions. The baseline emissions can be understood as the carbon emissions the park would have generated without implementing emission reduction measures, while the project emissions are the actual carbon emissions generated during the park's operation. Carbon emission reduction is measured in tons or kilograms. For example, if the baseline emissions are calculated to be 1000 kilograms and the actual project emissions are 850 kilograms during a certain scheduling cycle, then the carbon emission reduction for that cycle would be 150 kilograms.

[0090] A scheduling cycle timestamp is a time-series information used to identify the scheduling cycle to which carbon emission reduction data belongs. The timestamp can include the start and end times of the scheduling cycle. The introduction of timestamps gives carbon emission reduction data a clear time attribute, facilitating subsequent querying, statistics, and traceability. Equipment identifiers are identification information used to uniquely identify the equipment from which carbon emission reduction data originates. Equipment identifiers can be the equipment's serial number, asset number, or IoT device ID, etc. For example, the equipment identifier for a photovoltaic inverter is INV-2026-001, and the equipment identifier for a permanent magnet motor is MTR-PM-015. The introduction of equipment identifiers enables carbon emission reduction data to be traced back to specific emission reduction equipment, improving data reliability and auditability.

[0091] Carbon emission reduction data records refer to complete data entries generated by associating carbon emission reductions with scheduling cycle timestamps and equipment identifiers. A carbon emission reduction data record may contain the following fields: record number, scheduling cycle start time, scheduling cycle end time, equipment identifier, carbon emission reduction amount, accounting methodology version, accounting personnel or system, etc. Carbon emission reduction data records can be stored in structured formats, such as relational database tables or JSON documents, or in unstructured formats.

[0092] This application embodiment generates carbon emission reduction data records at the end of each scheduling cycle, enabling the park's carbon emission reduction achievements to be stored and managed in a structured form. The association between carbon emission reductions and timestamps and equipment identifiers provides a data foundation for subsequent carbon asset confirmation and traceability. This periodic carbon emission reduction data generation method matches the rolling scheduling mechanism, ensuring that carbon asset management remains synchronized with scheduling operations.

[0093] In some embodiments disclosed in this application, after step S6, a carbon quota overrun warning and emergency control step is further included: based on the real-time carbon price signal and the real-time remaining carbon quota of the park, it is predicted whether there is a risk of carbon quota overrun at the end of the current scheduling cycle; if there is a risk of overrun, an emergency control instruction is generated and issued for execution. The emergency control instruction includes at least one of the following: a first instruction to control the energy storage system to discharge at the maximum allowable power, a second instruction to control all permanent magnet motor clusters to operate in the highest energy efficiency mode, and a third instruction to control the park microgrid to fully absorb the real-time power generation of distributed photovoltaics locally.

[0094] Understandably, the carbon quota exceedance early warning and emergency control step refers to the process of predicting whether there is a risk of exceeding the limit in the next cycle based on the currently available carbon quota information after the end of each scheduling cycle, and proactively triggering emergency control measures when such a risk exists. This step is usually arranged after step S6, that is, immediately after updating the real-time remaining carbon quota in the park, in order to respond promptly to possible carbon quota shortages.

[0095] Carbon quota exceedance risk refers to the uncertainty that, based on current forecasts, the actual carbon emissions of a park at the end of a future scheduling cycle may exceed the upper limit of available carbon quotas. The magnitude of the exceedance risk depends on the amount of remaining carbon quotas in real time, the predicted carbon emissions, and the gap between the two. It is understood that exceedance risk is only a possibility, not a certainty, but its existence prompts the park to take preventative measures. For example, if the park has 500 kg of remaining carbon quotas in real time, and the current scheduling plan predicts 550 kg of carbon emissions in the next cycle, there is a 50 kg exceedance risk. Forecasting refers to pre-estimating the carbon quota consumption at the end of the next scheduling cycle based on real-time carbon price signals, the park's remaining carbon quotas in real time, and the expected operating status of the next cycle. Forecasting can use simple linear extrapolation methods, assuming that the energy consumption pattern in the next cycle is similar to that of the current cycle; or it can use more complex forecasting models that consider factors such as load changes and fluctuations in photovoltaic output. The purpose of forecasting is to identify potential exceedance risks in advance, allowing reaction time for emergency regulation.

[0096] Emergency control directives refer to a set of mandatory control instructions generated to address the risk of carbon quota exceeding limits. Unlike conventional dispatch directives, emergency control directives prioritize meeting carbon quota constraints in their generation logic, with economic objectives taking a secondary role. Emergency control directives can contain multiple sub-directives, which can be executed individually or in combination, depending on the severity of the exceeding risk and the actual operational needs of the industrial park. Discharging an energy storage system at its maximum permissible power refers to controlling the energy storage system to release electrical energy into the park's microgrid at its rated power or the currently permissible maximum discharge power. The maximum permissible power is typically limited by the capacity of the energy storage converter, the battery's discharge rate limit, and the current state of charge. By discharging the energy storage system at full power, partial load demand can be met without increasing external power purchases, thereby reducing carbon emissions associated with purchasing electricity from the external grid. For example, if an energy storage system has a rated power of 500 kW and a current state of charge of 80%, it can discharge at a power of 500 kW.

[0097] Operating the entire permanent magnet motor cluster in its highest efficiency mode means controlling each motor in the cluster to operate at its optimal energy efficiency. The highest efficiency mode typically corresponds to the motor's rated load point and can be achieved by adjusting inverter parameters or production processes. When the motors operate in the highest efficiency mode, the energy consumption per unit output is minimized, resulting in the lowest carbon emissions. Understandably, operating all motors in the highest efficiency mode may have some impact on production schedules, but it minimizes carbon emissions while ensuring output. The microgrid in the industrial park fully utilizes the real-time power generated by distributed photovoltaic (PV) power locally, prioritizing the consumption of all PV power generated by the PV system and preventing PV power from being fed back to the external grid. Full local consumption can be achieved by adjusting energy storage charging and discharging strategies and regulating motor loads. When PV power generation exceeds the park's real-time load, excess power can be used for energy storage charging; when energy storage is full, motor loads can be increased or backup equipment can be activated. Full consumption of PV power reduces the need to purchase electricity from the external grid, thereby reducing corresponding carbon emissions.

[0098] This application's embodiments predict carbon allowance consumption for the next cycle after the scheduling cycle ends, enabling the park to take countermeasures before the risk of exceeding carbon allowances arises. Emergency control commands are set separately for energy storage systems, permanent magnet motors, and photovoltaic (PV) integration, allowing the park to choose the most appropriate control method based on actual conditions. Maximum power discharge of the energy storage system reduces carbon emissions from purchasing electricity from the external grid; operating the permanent magnet motor in its highest energy efficiency mode reduces carbon emissions per unit of output; and full PV integration increases the proportion of green electricity and reduces carbon emissions from fossil fuel substitution. This early warning and emergency response mechanism provides a safety net for the park's carbon emission management.

[0099] Please refer to Figure 2 , Figure 2 This application provides a zero-carbon industrial park integrated source-grid-load-storage coordinated scheduling system. The system includes: a data acquisition module, communicatively connected to the inverters of distributed photovoltaic systems, the frequency converters of permanent magnet motor clusters, and the controllers of the energy storage system, for acquiring real-time power generation, real-time operating current, real-time operating voltage, and real-time state of charge, and receiving real-time electricity price signals and real-time carbon price signals; an optimization solution module, communicatively connected to the data acquisition module, for constructing and solving a multi-objective optimization model based on the data acquired by the data acquisition module, and generating integrated coordinated scheduling instructions; an instruction execution module, communicatively connected to the optimization solution module and the controllers of the energy storage system, the frequency converters of the permanent magnet motor clusters, and the grid-connected interface controller of the industrial park microgrid, for issuing and executing the instruction portion corresponding to the current time period in the integrated coordinated scheduling instructions, and receiving returned status feedback data; and a carbon quota update module, communicatively connected to the instruction execution module and the optimization solution module, for recalculating the remaining real-time carbon quota of the industrial park based on the status feedback data, and triggering the optimization solution module to re-solve the problem in each scheduling cycle.

[0100] In other words, the embodiments of this application disclose a novel system comprising a data acquisition module, an optimization solution module, an instruction execution module, and a carbon quota update module. The data acquisition module is communicatively connected to the inverters of the distributed photovoltaic system, the frequency converters of the permanent magnet motor cluster, and the controllers of the energy storage system, acquiring real-time power generation, operating current, voltage, and state of charge, while simultaneously receiving real-time electricity price signals from the grid dispatch system and real-time carbon price signals from the carbon trading platform. The optimization solution module constructs and solves a multi-objective optimization model based on the acquired data, generating an integrated collaborative dispatch instruction which is then sent to the instruction execution module. The instruction execution module issues the current time period instruction to the energy storage controller, the permanent magnet frequency converter, and the grid-connected interface controller of the park's microgrid for execution, and collects status feedback data. The carbon quota update module recalculates the remaining real-time carbon quota for the park based on the feedback data and triggers the optimization solution module to enter the next dispatch cycle, forming a closed-loop control.

[0101] Understandably, the data acquisition module is a hardware or software unit in the system responsible for interacting with field devices and acquiring operational data. This module typically consists of multiple edge computing gateways deployed at the sites of devices such as inverters in distributed photovoltaic systems, frequency converters in permanent magnet motor clusters, and controllers in energy storage systems. The data acquisition module establishes connections with these devices via industrial communication protocols, such as Modbus TCP, Profinet, or IEC 61850, to read in real-time the real-time power output of the photovoltaic inverter, the real-time operating current and voltage of the permanent magnet motor recorded by the frequency converter, and the real-time state of charge reported by the energy storage controller. Simultaneously, the data acquisition module also interfaces with external systems through application programming interfaces (APIs) to obtain real-time electricity price signals from the grid dispatching system and real-time carbon price signals from the carbon trading platform. Understandably, the data collected by the data acquisition module forms the basis for subsequent optimization decisions, and its accuracy and real-time performance directly affect the dispatching effect. For example, the data acquisition module can collect data from each device every 15 minutes, add a unified timestamp, and then upload it.

[0102] The optimization and solution module is the core algorithm computation unit in the system, typically deployed on a cloud server or local server. This module communicates with the data acquisition module via message queue telemetry transmission protocols or hypertext transfer protocols, receiving various real-time data and external signals uploaded by the data acquisition module. Internally, the optimization and solution module deploys a multi-objective optimization model and a particle swarm optimization algorithm. It can construct a mathematical model based on the received data, incorporating three optimization objectives: energy cost, carbon emissions, and green electricity consumption rate, while also including the park's remaining carbon quota as a constraint. Through iterative solving, the optimization and solution module generates a future charging and discharging power timing plan for the energy storage system, start-stop control commands and load adjustment commands for the permanent magnet motor cluster, and interactive power setpoints between the park's microgrid and the external power grid. These commands are then integrated into a unified collaborative scheduling command. It is understandable that the computational power of the optimization and solution module determines the optimization level and response speed of the scheduling scheme.

[0103] The instruction execution module is the unit in the system responsible for issuing scheduling instructions to devices and receiving feedback. This module communicates with the optimization solution module and each device controller, obtaining integrated collaborative scheduling instructions from the optimization solution module, extracting the instruction portion corresponding to the current time period, and then sending the instructions to the energy storage system controller, the inverter of the permanent magnet motor cluster, and the grid-connected interface controller of the microgrid via an edge computing gateway or direct communication with the controller. The instruction execution module is also responsible for collecting status feedback data returned by each controller after executing instructions, including the actual charging and discharging power of the energy storage system, the actual operating power of the permanent magnet motor, and the actual interaction power of the microgrid, and transmitting this data back to the optimization solution module and the carbon quota update module. For example, when the energy storage controller receives a 500 kW discharge instruction, the instruction execution module continuously monitors its actual discharge power and returns the actual value at the end of the cycle.

[0104] The carbon quota update module is a dedicated unit in the system responsible for the dynamic management of carbon quotas. This module communicates with both the instruction execution module and the optimization solution module, receiving status feedback data from the instruction execution module. Based on the actual interactive power and grid carbon emission factor in the status feedback data, the carbon quota update module calculates the actual carbon emissions generated by the park within the completed scheduling cycle and subtracts this value from the park's initially allocated total carbon quota to obtain the park's real-time remaining carbon quota. The updated carbon quota data is stored and synchronized to the optimization solution module. Simultaneously, the carbon quota update module triggers the optimization solution module to restart the solution process at the end of each scheduling cycle. In essence, the carbon quota update module enables carbon quotas to dynamically change with actual energy consumption, providing accurate constraint boundaries for scheduling decisions in the next cycle.

[0105] This application's embodiments modularize the functions corresponding to each step of the above method embodiments, enabling the method to be deployed and run in actual hardware systems. The data acquisition module is responsible for interacting with field equipment to obtain real-time operating data and external signals; the optimization and solution module undertakes the core algorithm calculations and generates scheduling instructions; the instruction execution module is responsible for issuing instructions and collecting feedback; and the carbon quota update module realizes the dynamic update of carbon quotas and the triggering of scheduling cycles. The division of labor and cooperation among these modules enables the integrated source-grid-load-storage coordinated scheduling method to be implemented in actual industrial parks, forming a complete closed loop from data acquisition to instruction execution and quota update.

[0106] Figure 3 This is a flowchart illustrating the closed-loop process of carbon asset securitization and carbon benefit realization in this application, demonstrating how the carbon emission reduction data generated by the scheduling method of this application is further transformed into carbon assets and achieves a value closed loop. Figure 3As shown, this process begins with real-time calculation of carbon emission reductions. The carbon emission reduction data records generated after step S5 of this application provide the basic data for carbon emission reduction calculation. After being aggregated, this data enters the carbon asset confirmation and certification stage. Blockchain technology is used to store the verified carbon emission reductions, forming an immutable digital certificate of carbon assets. The confirmed carbon assets are packaged into a carbon asset package, and the energy service company acquires the future revenue rights through the establishment of a special purpose vehicle. After credit enhancement, carbon asset-backed securities are issued to complete market-based financing, and the raised funds are deposited into a tripartite co-managed account.

[0107] The raised funds will be allocated according to a predetermined ratio. One part will be used to support the park's construction, including the expansion of distributed photovoltaic capacity, the addition of energy storage systems, and the upgrading of permanent magnet motors. These investments will further enhance the park's emission reduction capabilities. The other part will be injected into the carbon incentive pool as a reserve for employee points redemption. Carbon incentive points earned by employees through green behaviors can be redeemed through a two-way redemption channel. Employees can choose to redeem their points for a share of the proceeds from carbon asset securitization products to obtain financial returns, or they can choose to use their points to offset their company's carbon quota shortfall. After verification by the carbon trading center, the quota will be cancelled, and the individual's points will be deducted accordingly.

[0108] The carbon emission reduction data generated by the scheduling method provides the underlying assets for carbon asset securitization. Funds raised through securitization are used to support park construction, upgrading equipment such as distributed photovoltaic systems, energy storage systems, and permanent magnet motors. These upgrades further enhance the operational effectiveness of the scheduling method. Simultaneously, the establishment of a carbon incentive pool allows employees to receive tangible rewards for their green behaviors. Employees accumulate points through green cycling, charging new energy vehicles, etc., and the points redemption mechanism enhances their enthusiasm for participating in carbon emission reduction within the park. By allowing park enterprises to offset carbon quotas with employee points, they can obtain additional buffer space when carbon quotas are tight, fully leveraging the flexibility of the carbon quota constraints in the scheduling method. Through this closed-loop design, the scheduling method of this application, together with carbon asset management and carbon incentives, forms a complete technological chain from energy optimization to carbon asset monetization and then to inclusive incentives.

[0109] Please refer to Figure 4 , Figure 4 This is a schematic block diagram of an electronic device provided according to an embodiment of this application. Figure 4The electronic device 300 in this embodiment may include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memories 304 store computer programs, including program instructions. The processors 301 execute the program instructions stored in the memories 304. Specifically, the processors 301 are configured to invoke the program instructions to execute the aforementioned zero-carbon park source-grid-load-storage integrated coordinated scheduling method.

[0110] It should be understood that, in the embodiments of this application, the processor 301 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0111] Input device 302 may include a touchpad, a fingerprint sensor (for collecting the user's fingerprint information and fingerprint orientation information), a microphone, etc., and output device 303 may include a display (LCD, etc.), a speaker, etc.

[0112] The memory 304 may include read-only memory and random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include non-volatile random access memory. For example, the memory 304 may also store device type information.

[0113] In specific implementations, the processor 301, input device 302, and output device 303 described in the embodiments of this application can execute the implementation methods described in any embodiment of the zero-carbon park source-grid-load-storage integrated collaborative scheduling method provided in the embodiments of this application, or they can execute the implementation methods of the electronic devices described in the embodiments of this application, which will not be repeated here.

[0114] In another embodiment of this application, an electronic device is provided. The electronic device stores a computer program, which includes program instructions. When executed by a processor, the program instructions implement all or part of the processes in the zero-carbon park source-grid-load-storage integrated coordinated scheduling method described above. Alternatively, the computer program can instruct related hardware to complete the process. The computer program can be stored in an electronic device, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. Computer-readable media can include any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0115] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the foregoing embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device. Furthermore, the computer-readable storage medium can include both internal and external storage units of the electronic device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0116] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design 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 implementations should not be considered beyond the scope of this application.

[0117] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the electronic devices and units described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0118] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces or units, or it may be an electrical, mechanical, or other form of connection.

[0119] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of this application, depending on actual needs.

[0120] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0121] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered 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.

Claims

1. A method for integrated coordinated scheduling of source, grid, load, and storage in a zero-carbon industrial park, characterized in that, Includes the following steps: S1: Collect the real-time power generation of distributed photovoltaic on the source side, the real-time operating current and voltage of the permanent magnet motor cluster on the load side, the real-time state of charge of the energy storage system on the storage side, and obtain the real-time electricity price signal of the external power grid and the real-time carbon price signal of the carbon trading market. S2: Based on the real-time carbon price signal, the real-time remaining carbon quota of the park, the real-time electricity price signal, and the real-time power generation, real-time operating current, real-time operating voltage and real-time state of charge, construct a multi-objective optimization model with the park's comprehensive energy cost, park carbon emissions and green electricity consumption rate as optimization objectives. S3: Solve the multi-objective optimization model to generate a charging and discharging power timing plan for the energy storage system in the future time period, start-stop control commands and load adjustment commands for the permanent magnet motor cluster, and interactive power setting values ​​between the park microgrid and the external power grid, as integrated collaborative scheduling commands. S4: Send the instruction portion corresponding to the current time period in the integrated collaborative scheduling instruction to the controller of the energy storage system, the frequency converter of the permanent magnet motor cluster, and the grid connection interface controller of the park microgrid for execution; S5: Collect the actual charging and discharging power returned by the controller of the energy storage system, the actual operating power returned by the frequency converter of the permanent magnet motor cluster, and the actual interactive power returned by the grid-connected interface controller as status feedback data; S6: Recalculate the remaining carbon allowance for the park in real time based on the status feedback data, and return to step S1 when the next scheduling cycle arrives.

2. The method according to claim 1, characterized in that, The multi-objective optimization model constructed in step S2 includes a carbon quota constraint condition, which is used to limit the total predicted carbon emissions of the park in the future period to not exceed the sum of the park's real-time remaining carbon quota and the quota expected to be offset through carbon credits.

3. The method according to claim 1, characterized in that, Step S3 generates start / stop control commands and load adjustment commands for the permanent magnet motor cluster, including: Based on the real-time operating current and real-time operating voltage, the energy efficiency of each permanent magnet motor in the permanent magnet motor cluster is evaluated in real time to obtain the real-time energy efficiency value of each permanent magnet motor. The real-time energy efficiency value is compared with a preset energy efficiency threshold to identify inefficient motors whose real-time energy efficiency value is lower than the energy efficiency threshold. Based on the identification results of the inefficiently operating motors, the real-time electricity price signal, and the real-time carbon price signal, combined with the real-time remaining carbon quota of the park, a shutdown command or load reduction command is generated for the inefficiently operating motors.

4. The method according to claim 3, characterized in that, When generating the shutdown command or load reduction command based on the identification result of the inefficient operating motor, the shutdown or load reduction operation is performed on the motor with the lowest real-time energy efficiency value among the inefficient operating motors first, until the predicted total carbon emissions meet the carbon quota constraint conditions.

5. The method according to claim 1, characterized in that, Step S3 uses the particle swarm optimization algorithm to solve the multi-objective optimization model, including: The charging and discharging power of the energy storage system at each time point in the future period, the total operating power of the permanent magnet motor cluster at each time point in the future period, and the interaction power between the park microgrid and the external power grid at each time point in the future period are encoded as particle position vectors. Using the optimization objective of the multi-objective optimization model as the evaluation criterion for particle fitness, the global optimal particle position that achieves optimal fitness is searched by iteratively updating the particle's velocity and position. The globally optimal particle position is decoded to obtain the charging and discharging power timing plan, the start-stop control command and load adjustment command, and the interactive power setting value.

6. The method according to claim 1, characterized in that, The real-time operating current and real-time operating voltage collected in step S1 also include the dispatchable capacity signal uploaded by new energy vehicles connected to the park microgrid via V2G terminals; The integrated collaborative scheduling instruction generated in step S3 also includes a charging and discharging scheduling instruction for the new energy vehicle, which is used to control the new energy vehicle to charge from the park microgrid when the real-time electricity price signal is at a valley value, and to discharge to the park microgrid when the real-time electricity price signal is at a peak value.

7. The method according to claim 1, characterized in that, The method further includes a virtual power plant aggregation and control step, which is performed after step S3 and before step S4: The distributed photovoltaic system, the energy storage system, and the permanent magnet motor cluster are aggregated into a virtual power plant. According to the integrated coordinated dispatch instructions, under the premise of meeting the load demand within the park, the regulation capacity reserved for participating in grid ancillary services is used to respond to the frequency regulation instructions or peak regulation instructions issued by the grid dispatch system. The revenue generated from ancillary services in response to the frequency regulation or peak shaving commands will be included in the overall energy cost of the park.

8. The method according to claim 2, characterized in that, Step S6 is followed by a step of adaptive adjustment of weight coefficients: Based on the deviation between the status feedback data and the instruction portion corresponding to the current time period in the integrated collaborative scheduling instruction, the actual energy cost, actual carbon emissions, and actual green electricity consumption rate for the current scheduling cycle are calculated. Based on the degree of closeness between the actual energy cost, the actual carbon emissions, and the actual green electricity consumption rate and the optimization target, the weight coefficients corresponding to each optimization target in the multi-objective optimization model are corrected online. The corrected weighting coefficients will be used to solve the model in the next scheduling cycle.

9. The method according to claim 1, characterized in that, The remaining carbon allowance in the park in step S2 is dynamically determined by subtracting the carbon emissions already generated up to the current time from the total carbon allowance initially allocated to the park based on the real-time power generation and the status feedback data.

10. A zero-carbon industrial park source-grid-load-storage integrated collaborative scheduling system, applied to the method of any one of claims 1 to 9, characterized in that, include: The data acquisition module is communicatively connected to the inverter of the distributed photovoltaic system, the frequency converter of the permanent magnet motor cluster, and the controller of the energy storage system. It is used to acquire the real-time power generation, the real-time operating current, the real-time operating voltage, and the real-time state of charge, and to receive the real-time electricity price signal and the real-time carbon price signal. The optimization and solution module is communicatively connected to the data acquisition module and is used to construct and solve the multi-objective optimization model based on the data acquired by the data acquisition module, and generate the integrated collaborative scheduling instruction. The instruction execution module is communicatively connected to the optimization solution module, the controller of the energy storage system, the frequency converter of the permanent magnet motor cluster, and the grid-connected interface controller of the park microgrid. It is used to issue and execute the instruction portion corresponding to the current time period in the integrated collaborative scheduling instruction and to receive the returned status feedback data. The carbon quota update module is communicatively connected to both the instruction execution module and the optimization solution module. It is used to recalculate the real-time remaining carbon quota of the park based on the status feedback data, and to trigger the optimization solution module to re-solve the problem in each scheduling cycle.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the zero-carbon park source-grid-load-storage integrated collaborative scheduling method as described in any one of claims 1 to 9.

Images