Energy control method and system of light storage and charging micro-grid for zero-carbon park

Abstract

The application provides a zero-carbon park-oriented light storage and charging micro-grid energy control method and system, and belongs to the technical field of zero-carbon park energy control. The method is specifically as follows: based on real-time operation data of the micro-grid, source and load prediction is carried out in combination with corresponding environmental parameters, and dynamic carbon emission intensity is calculated; based on real-time photovoltaic output, real-time load of the park and source and load prediction results, the power balance state of the micro-grid is identified, and the energy distribution target is matched according to the power balance state; based on the power balance state, the dynamic carbon emission intensity and the preset carbon emission threshold, a dynamic low-carbon weight is generated; taking the photovoltaic direct supply rule as a constraint condition, the park energy distribution strategy is formulated based on the energy distribution target and in combination with the dynamic low-carbon weight; each energy object executes corresponding energy control according to the park energy distribution strategy. The application can adaptively adjust energy scheduling according to carbon emission intensity fluctuation, reduce the proportion of high-carbon power sources in the consumed electric energy of the park, and effectively reduce the carbon emission of the park.

Description

Technical Field

[0001] This invention relates to the field of energy control technology for zero-carbon industrial parks, and in particular to a method and system for energy control of photovoltaic-storage-charging microgrids for zero-carbon industrial parks. Background Technology

[0002] A zero-carbon park refers to a functional area within a certain region where annual net carbon dioxide emissions approach zero through clean energy substitution, energy efficiency improvement, and carbon management. A photovoltaic-storage-charging integrated microgrid, as a typical energy supply architecture for a zero-carbon park, typically includes distributed photovoltaic systems, energy storage systems, electric vehicle charging facilities, and various loads within the park.

[0003] Currently, photovoltaic-storage-charging microgrids in zero-carbon industrial parks generally manage energy based on instantaneous power balance. When the photovoltaic output exceeds the load, the excess power is redirected to the energy storage for charging or fed into the grid via the inverter. When the photovoltaic output is less than the load, the energy storage discharges or the grid is purchased to make up for the shortfall.

[0004] However, the carbon emission intensity of the power grid actually changes in real time with the power structure. If the power purchase and energy dispatch strategy is determined solely based on the instantaneous power difference, large-scale power purchases are likely to occur during periods of high carbon emissions from the power grid. This leads to an increase in the proportion of high-carbon power sources providing electricity in the energy consumed by the park, resulting in an overall high carbon emission level that is difficult to match the low-carbon operation requirements of a zero-carbon park. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of zero-carbon parks that manage energy based on instantaneous power balance, which cannot identify the time-varying characteristics of the grid's carbon emission intensity and result in high overall carbon emissions. This invention provides a photovoltaic-storage-charging microgrid energy control method and system for zero-carbon parks. It quantifies real-time changes in carbon emissions through dynamic carbon emission intensity and generates corresponding dynamic low-carbon weights to set energy allocation priorities. Combined with photovoltaic direct supply rules, it formulates energy allocation strategies to adaptively adjust energy dispatch according to carbon emission intensity fluctuations, thereby reducing the proportion of high-carbon power sources in the park's electricity consumption and reducing the park's carbon emissions.

[0006] The objective of this invention is achieved through the following technical solution: Energy control methods for photovoltaic-storage-charging microgrids in zero-carbon industrial parks include: Based on real-time operation data of the microgrid, source load prediction is performed in combination with corresponding environmental parameters, and dynamic carbon emission intensity is calculated. Based on real-time photovoltaic output, real-time load of the park, and source-load prediction results, the power balance status of the microgrid is identified, and energy allocation targets are matched according to the power balance status. Dynamic low-carbon weights are generated based on power balance state, dynamic carbon emission intensity, and preset carbon emission thresholds. Based on the direct photovoltaic power supply rules and energy allocation targets, and combined with dynamic low-carbon weights, the park's energy allocation strategy is formulated. Each energy source will implement corresponding energy control according to the park's energy allocation strategy.

[0007] Furthermore, the identification of the microgrid's power balance state based on real-time photovoltaic output, real-time load of the industrial park, and source-load prediction results includes: The real-time power difference is calculated based on the real-time photovoltaic output and the real-time load of the park, and candidate power balance states are identified by combining the preset state thresholds. Based on the source-load prediction results, the predicted power difference for each future time period is obtained, and the identified candidate power balance state is verified based on the predicted power difference for each future time period. The current power balance status of the microgrid is determined based on the verification results.

[0008] Furthermore, the step of verifying the identified candidate power balance states based on the predicted power differences in each future time period includes: Based on the predicted power difference at each future time, and combined with the preset state threshold, the power balance state corresponding to each predicted power difference is determined. Based on the distribution ratio and corresponding duration of power balance states for all future time periods, the steady-state supply and demand state for future time periods is determined. When the steady-state supply and demand state is consistent with the candidate power balance state, the judgment is passed and the candidate power balance state is taken as the current power balance state of the microgrid. If the steady-state supply and demand state is inconsistent with the candidate power balance state, the test is deemed to have failed, and the candidate power balance state is adjusted according to the steady-state supply and demand state.

[0009] Furthermore, the step of matching energy allocation targets based on power balance status includes: Identify the direction and degree of supply-demand deviation between photovoltaic power output and park load based on power balance status; Based on the direction and degree of supply-demand deviation, the energy allocation identity and corresponding power target range of each energy object are matched.

[0010] Furthermore, the generation of dynamic low-carbon weights based on power balance state, dynamic carbon emission intensity, and preset carbon emission thresholds includes: Identify the direction and degree of supply-demand deviation between photovoltaic power output and park load based on power balance status; The basic polarity of the weights of each energy object is set based on the direction of the supply and demand deviation, and the correction magnitude is calculated based on the relative deviation between the degree of supply and demand deviation and the dynamic carbon emission intensity and the preset carbon emission threshold. Multiply the basic polarity of the weight by the correction magnitude, and then add it to the preset basic weight to obtain the dynamic low-carbon weight.

[0011] Furthermore, the energy allocation strategy for the industrial park, formulated based on the direct photovoltaic supply rule as a constraint, energy allocation targets, and dynamic low-carbon weights, includes: Based on the energy allocation objectives, obtain the energy allocation identity and corresponding power target range for each energy object; The direct photovoltaic supply rule serves as the constraint for the energy allocation order, and the dynamic low-carbon weight serves as the priority adjustment coefficient. The execution order of each energy object is set according to its corresponding energy allocation identity. Based on the power target range of each energy object and the revised execution order, the power allocation ratio is determined to obtain the park's energy allocation strategy.

[0012] Furthermore, after each energy object performs corresponding energy control according to the park's energy allocation strategy, the following also applies: The system collects the actual operating parameters of the microgrid in real time and corrects the dynamic low-carbon weight and preset carbon emission threshold based on the deviation between the actual operating parameters and the source-load prediction results.

[0013] Furthermore, the source-load prediction based on real-time microgrid operation data, combined with corresponding environmental parameters, includes: Based on the real-time operation data of the microgrid and combined with the corresponding environmental parameters, the predicted values ​​of photovoltaic power output and park load for future periods are obtained through time series forecasting. Based on the real-time operation data of the microgrid, the current operating condition of the microgrid is identified, and the corresponding operating condition deviation is matched according to the current operating condition of the microgrid. Based on the matching of environmental parameters with corresponding historical photovoltaic power output data, and combined with real-time operation data, the power output deviation value is obtained; The current performance degradation level of the power generation equipment is obtained based on the output deviation value, and the corresponding degradation deviation amount is matched according to the performance degradation level of the power generation equipment. The predicted load value of the park is corrected based on the operating condition deviation, and the predicted photovoltaic output value is corrected based on the attenuation deviation.

[0014] Furthermore, the calculation of dynamic carbon emission intensity includes: Based on the grid-connected power type of the microgrid, match the corresponding unit carbon emission benchmark value; The power consumption of each type of grid-connected electrical energy is obtained based on the real-time operation data of the microgrid, and the real-time dynamic carbon emission intensity is calculated by combining the corresponding unit carbon emission benchmark value.

[0015] An energy control system for a photovoltaic-storage-charging microgrid in a zero-carbon industrial park, used to execute any of the above-mentioned energy control methods, including: The data processing module is used to predict source load based on real-time operating data of the microgrid and corresponding environmental parameters, and to calculate dynamic carbon emission intensity. The status judgment module is used to identify the power balance status of the microgrid based on real-time photovoltaic output, real-time load of the park, and source-load prediction results, and to match energy allocation targets according to the power balance status. The energy allocation module is used to generate dynamic low-carbon weights based on power balance status, dynamic carbon emission intensity, and preset carbon emission thresholds. It also uses photovoltaic direct supply rules as constraints to formulate park energy allocation strategies based on energy allocation targets and dynamic low-carbon weights.

[0016] The beneficial effects of this invention are: By introducing dynamic carbon emission intensity calculation, power balance state identification, and dynamic low-carbon weight generation mechanism, the energy allocation priority of each energy object is adaptively adjusted according to the real-time fluctuation of the grid carbon emission intensity, thereby avoiding excessive reliance on grid power purchase during high carbon emission periods and effectively reducing the overall carbon emissions of the park.

[0017] Simultaneously, when formulating the park's energy allocation strategy, direct photovoltaic (PV) power supply rules are imposed to ensure that PV DC-side power is prioritized for direct supply to the park's load, reducing energy losses caused by repeated PV power inversion and improving the utilization efficiency of clean PV energy. Furthermore, through source-load prediction and dynamic identification of power balance, frequent charging and discharging switching of energy storage due to instantaneous PV fluctuations is avoided, extending the lifespan of energy storage equipment. This approach enables synergistic optimization of low-carbon efficiency and economy while ensuring power balance within the park, thereby improving the energy control efficiency of zero-carbon parks. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of a process of the present invention; Figure 2 This is a schematic diagram of the power balance state identification process according to an embodiment of the present invention. Detailed Implementation

[0019] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0020] Example: Energy control method for photovoltaic-storage-charging microgrids in zero-carbon industrial parks, such as... Figure 1 As shown, it includes: Based on real-time operation data of the microgrid, source load prediction is performed in combination with corresponding environmental parameters, and dynamic carbon emission intensity is calculated. Based on real-time photovoltaic output, real-time load of the park, and source-load prediction results, the power balance status of the microgrid is identified, and energy allocation targets are matched according to the power balance status. Dynamic low-carbon weights are generated based on power balance state, dynamic carbon emission intensity, and preset carbon emission thresholds. Based on the direct photovoltaic power supply rules and energy allocation targets, and combined with dynamic low-carbon weights, the park's energy allocation strategy is formulated. Each energy source will implement corresponding energy control according to the park's energy allocation strategy.

[0021] By collecting microgrid operation data in real time and integrating environmental parameters for source-load forecasting, and dynamically calculating grid carbon emission intensity, the system quantifies future supply-demand trends and real-time fluctuations in carbon emission intensity, using both carbon emission and power dimensions as the basis for energy allocation decisions. Based on this, real-time photovoltaic output, park load, and forecast results are used to identify the power balance state, and energy allocation targets are matched according to this state. This accurately quantifies the direction and extent of supply-demand deviations in the current and future periods, avoiding decision-making biases caused by relying solely on instantaneous power differences.

[0022] Furthermore, dynamic low-carbon weights are generated based on power balance status, dynamic carbon emission intensity, and preset carbon emission thresholds. The real-time fluctuations in carbon emission intensity are quantified into a continuously adjustable priority parameter. Combined with the constraints of photovoltaic direct supply rules, an energy allocation strategy for the park is formulated according to the energy allocation target. This ensures that photovoltaic DC-side power is prioritized for direct supply to DC loads, reducing energy losses caused by repeated inversions. Moreover, the scheduling order and power ratio of energy storage, electric vehicles, and the power grid can be directly adjusted through dynamic low-carbon weights, enabling adaptive adjustment of energy allocation based on carbon emission intensity fluctuations.

[0023] After each energy source implements its corresponding energy control according to the assigned strategy, carbon emissions in the park can be reduced while ensuring power balance, thus achieving synergistic optimization of low-carbon and economic efficiency.

[0024] Because photovoltaic power output is intermittent and random, and park load fluctuates in real time, relying solely on instantaneous operating data can only reflect the current state and cannot predict future supply and demand changes, which can easily lead to lag and deviation in scheduling decisions. Therefore, based on the real-time operating data of the microgrid, source and load prediction is first performed in combination with corresponding environmental parameters to obtain data on changes in photovoltaic power output and park load in future periods. This allows for the quantification of future supply and demand trends, which serve as the basis for judging the power balance status. This ensures that the power balance status judgment results are consistent with the actual operating conditions and reduces decision-making deviations caused by short-term fluctuations.

[0025] When forecasting power generation and load, most methods rely on basic data and general time-series models for calculations. However, the actual power load in a park varies significantly depending on the microgrid's operating conditions. These forecasting methods cannot distinguish the power consumption characteristics corresponding to different operating conditions, resulting in a large discrepancy between the predicted load and the actual power demand. Furthermore, photovoltaic (PV) modules naturally age over time, and their actual power generation capacity gradually decreases under the same environmental conditions. These forecasting methods can only estimate the theoretical output based on environmental parameters, ignoring the power generation loss caused by equipment performance degradation. Consequently, the predicted PV output is likely to be higher than the actual output power.

[0026] Therefore, based on source-load forecasting, the current actual operating conditions are further determined and the operating condition deviation is matched. The predicted load value of the park is corrected according to the operating condition deviation to adapt to the real power consumption patterns under different operating scenarios. Then, the environmental parameters are matched with historical power generation data, and the actual output deviation is calculated by comparing with real-time operating data. The actual performance degradation of the power generation equipment is accurately determined. The photovoltaic output prediction value is then corrected according to the degradation deviation. This simultaneously reduces the prediction error caused by two types of actual operating interference factors: operating condition differences and power generation equipment performance degradation. This effectively makes up for the insufficient fit of ordinary source-load forecasting methods, improves the accuracy of photovoltaic output prediction value and park load prediction value, and makes the final prediction result fit the actual operating state of the microgrid, thereby improving the accuracy of subsequent power balance state determination.

[0027] The source-load prediction based on real-time microgrid operation data and corresponding environmental parameters includes: Based on the real-time operation data of the microgrid and combined with the corresponding environmental parameters, the predicted values ​​of photovoltaic power output and park load for future periods are obtained through time series forecasting. Based on the real-time operation data of the microgrid, the current operating condition of the microgrid is identified, and the corresponding operating condition deviation is matched according to the current operating condition of the microgrid. Based on the matching of environmental parameters with corresponding historical photovoltaic power output data, and combined with real-time operation data, the power output deviation value is obtained; The current performance degradation level of the power generation equipment is obtained based on the output deviation value, and the corresponding degradation deviation amount is matched according to the performance degradation level of the power generation equipment. The predicted load value of the park is corrected based on the operating condition deviation, and the predicted photovoltaic output value is corrected based on the attenuation deviation.

[0028] At the start of each scheduling cycle, real-time operating data of the microgrid is collected, including the real-time output power of the photovoltaic array, the state of charge and charging / discharging power limits of the energy storage system, the number of connected electric vehicles and the state of charge and charging demand of each vehicle, the DC load power of the park, the AC load power of the park, the grid interaction power and the real-time carbon emission intensity. At the same time, environmental parameters, including irradiance and ambient temperature, are also collected.

[0029] The collected real-time operational data and environmental parameters are input into a pre-trained time-series prediction model. This model is based on deep learning networks, such as Long Short-Term Memory networks or Gated Recurrent Unit networks, and has been trained using historical operational data and corresponding environmental parameters. After forward propagation, the time-series prediction model outputs predicted photovoltaic power output and park load for a specified future time period.

[0030] Based on the direction and magnitude of grid interaction power in real-time operational data, the system determines whether the current operation is under grid-connected power purchase, grid-connected power supply, or off-grid conditions. Based on the energy storage state of charge (SOC) values, it determines whether the current operation is in a full-charge, low-charge, or normal range. Based on the deviation of the sum of the DC and AC loads in the microgrid relative to historical statistics for the same period, it determines whether the current operation is under peak load, low load, or normal load conditions. The results from these three operating condition dimensions are combined to form a complete operating condition identifier for the current microgrid. This identifier is then used as an index to match the corresponding operating condition deviation from a pre-established operating condition deviation mapping table. The operating condition deviation mapping table is constructed by statistically analyzing the average or cumulative deviation between the actual load and the time-series predicted load within the corresponding historical period for each operating condition.

[0031] Simultaneously, using light intensity and ambient temperature as environmental parameters, historical photovoltaic (PV) output records under similar or identical environmental conditions are filtered out. Statistical characteristic values ​​of the PV output values ​​in these records are calculated, and these statistical characteristic values ​​are used as the expected standard output. The real-time PV output in the real-time operating data is compared with the expected standard output, and the difference is calculated as the output deviation value. The output deviation value is matched with preset attenuation level intervals, with each attenuation level interval corresponding to an attenuation deviation amount. The corresponding attenuation deviation amount is output based on the interval the output deviation value falls into.

[0032] The load forecast value of the park obtained by time series forecasting is corrected by addition or multiplication using the operating condition deviation. At the same time, the photovoltaic output forecast value obtained by time series forecasting is corrected by addition or multiplication using the attenuation deviation.

[0033] The projected photovoltaic output and park load values ​​obtained from source-load forecasting are used to determine the power balance status and total energy deficit or surplus of the microgrid. However, the power balance status can only characterize the power supply and demand relationship and cannot determine the carbon emission level corresponding to the grid's electricity consumption. Furthermore, the grid carbon emission intensity of the microgrid changes in real time with the power structure. Even if the accurate power supply and demand relationship is known, it is impossible to distinguish high carbon emission periods. It can only achieve accurate energy allocation, but it still cannot avoid the increase in the proportion of high-carbon power sources in the park's electricity consumption due to electricity purchases during high carbon emission periods, which in turn leads to a higher overall carbon emission of the park.

[0034] Therefore, while conducting source-load forecasting to obtain accurate power supply and demand relationships, we further calculate dynamic carbon emission intensity based on real-time operating data to quantify the real-time fluctuations of grid carbon emission intensity. Then, we use dynamic carbon emission intensity and power balance status together as the planning basis for energy control. Under the premise of achieving power balance, we reduce the dependence on grid power purchase during high carbon emission periods, reduce the proportion of high-carbon power sources in the park's power consumption, and thus reduce the carbon emissions of the zero-carbon park.

[0035] The calculation of dynamic carbon emission intensity includes: Based on the grid-connected power type of the microgrid, match the corresponding unit carbon emission benchmark value; The power consumption of each type of grid-connected electrical energy is obtained based on the real-time operation data of the microgrid, and the real-time dynamic carbon emission intensity is calculated by combining the corresponding unit carbon emission benchmark value.

[0036] First, determine the current grid-connected power type of the microgrid based on the power grid dispatch system. This refers to the type of electricity the microgrid obtains from the grid, including coal, gas, and renewable energy sources. Then, match the corresponding unit carbon emission benchmark value to each type of power. This value is a pre-calibrated value and can be set according to actual conditions.

[0037] At the microgrid grid connection interface, the total power purchased from the grid during the current period is monitored, and the power supply ratio of various types of electrical energy connected to the grid during this period is obtained from the grid dispatch system, that is, the share of each type of electrical energy in the total purchased power.

[0038] For each type of grid-connected electricity, the power contribution provided by that type of electricity in the total purchased power value is determined based on its supply ratio; this is the power consumption. Multiplying the power consumption by the corresponding unit carbon emission benchmark value yields the carbon emission intensity component generated by the corresponding grid-connected electricity type in the current time period.

[0039] Finally, the carbon emission intensity components of all grid-connected power types are summed, and the summation result is divided by the total power purchased to obtain the dynamic carbon emission intensity for the current period.

[0040] After obtaining the source load forecast results and dynamic carbon emission intensity, the power balance status is further identified by combining the source load forecast results to obtain the direction and degree of supply and demand deviation in the park, thereby determining the basis for energy allocation.

[0041] Among them, such as Figure 2 As shown, the process of identifying the power balance state of the microgrid based on real-time photovoltaic output, real-time load of the industrial park, and source-load prediction results includes: The real-time power difference is calculated based on the real-time photovoltaic output and the real-time load of the park, and candidate power balance states are identified by combining the preset state thresholds. Based on the source-load prediction results, the predicted power difference for each future time period is obtained, and the identified candidate power balance state is verified based on the predicted power difference for each future time period. The current power balance status of the microgrid is determined based on the verification results.

[0042] First, based on the real-time operation data of the microgrid, the real-time photovoltaic output and the real-time load of the park are obtained at the current moment, and the difference between the two is calculated as the real-time power difference. This real-time power difference is compared with a preset state threshold to determine the power balance state identified based on the instantaneous power state, and this is used as the candidate power balance state.

[0043] The real-time load of the park includes the real-time AC and DC load power of the zero-carbon park, the charging demand of electric vehicles, and the energy storage charging power.

[0044] The candidate power balance states include photovoltaic power surplus, photovoltaic power shortage, and photovoltaic power balance. When the real-time power difference is greater than the positive state threshold, it is determined that the current state is a candidate power balance state with photovoltaic power surplus. When the real-time power difference is less than the negative state threshold, it is determined that the current state is a candidate power balance state with photovoltaic power shortage. When the real-time power difference is between the positive and negative state thresholds, it is determined that the current state is a photovoltaic power balance state.

[0045] Considering that real-time power difference is affected by minute-level fluctuations in photovoltaic output, resulting in instantaneous alternation between positive and negative values, the candidate power balance state determined solely by real-time values ​​is unstable and prone to frequent jumps. In contrast, the previously obtained source-load forecast results reflect continuous supply and demand trends over future periods and are unaffected by instantaneous fluctuations. Therefore, the candidate power balance state is verified using the steady-state supply and demand state over future periods reflected in the source-load forecast results to eliminate misjudgments caused by instantaneous fluctuations, outputting a stable power balance state that reflects both the current actual supply and demand relationship and conforms to future evolution trends.

[0046] The step of verifying the identified candidate power balance states based on the predicted power differences in each future time period includes: Based on the predicted power difference at each future time, and combined with the preset state threshold, the power balance state corresponding to each predicted power difference is determined. Based on the distribution ratio and corresponding duration of power balance states for all future time periods, the steady-state supply and demand state for future time periods is determined. When the steady-state supply and demand state is consistent with the candidate power balance state, the judgment is passed and the candidate power balance state is taken as the current power balance state of the microgrid. If the steady-state supply and demand state is inconsistent with the candidate power balance state, the test is deemed to have failed, and the candidate power balance state is adjusted according to the steady-state supply and demand state.

[0047] Based on the source-load forecast results, the predicted power difference for each future time period is obtained, which is the predicted photovoltaic output value minus the predicted load value of the park at each time.

[0048] Each predicted power difference is compared with a preset state threshold. If it is greater than the positive state threshold, it is determined to be a state of predicted power excess. If it is less than the negative state threshold, it is determined to be a state of predicted power deficiency. If it is between the two, it is determined to be a state of predicted power balance. Thus, the predicted power balance state corresponding to each time point in the future period is obtained.

[0049] The distribution of all predicted power balance states within the future time period is statistically analyzed to determine the distribution ratio of each state, i.e., the proportion of the number of occurrences to the total number of time periods, and the duration of each state, i.e., the longest consecutive occurrence duration.

[0050] The steady-state supply and demand status for the future period is determined based on the distribution ratio and duration. If the distribution ratio of a predicted power balance state exceeds a preset ratio threshold and the duration exceeds a preset duration threshold, then the state is determined as the steady-state supply and demand status for the future period. If the distribution ratio of all states does not exceed the threshold, or if none of the states reach the preset duration threshold, then the steady-state supply and demand status is determined as a balance state.

[0051] Next, the candidate power balance state is compared with the steady-state supply and demand state in the future period. If they match, the verification is considered successful, indicating that the current instantaneous state is consistent with the future trend, and the candidate power balance state is used as the current power balance state output of the microgrid. If they do not match, the verification is considered unsuccessful, indicating that the current instantaneous state may be caused by fluctuations, and the future trend is more representative of the current true supply and demand trend. In this case, the steady-state supply and demand state in the future period is used as the current power balance state of the microgrid, that is, the steady-state supply and demand state replaces the candidate state.

[0052] This verification of steady-state supply and demand in future time periods can effectively avoid frequent changes in power balance caused by instantaneous fluctuations, thereby preventing unnecessary power command switching by energy storage, electric vehicles and other energy objects, and further improving the stability of zero-carbon park operation and the reliability of scheduling decisions.

[0053] After determining the power balance state, considering that it only describes the direction and degree of the supply and demand deviation in the zero-carbon park, in order to optimize the processing efficiency of subsequent scheduling decisions, the power balance state is further quantified into the specific power command direction and adjustable power range of each energy object through energy allocation target matching.

[0054] The step of matching energy allocation targets based on power balance status includes: Identify the direction and degree of supply-demand deviation between photovoltaic power output and park load based on power balance status; Based on the direction and degree of supply-demand deviation, the energy allocation identity and corresponding power target range of each energy object are matched.

[0055] Based on the identified power balance state, the direction and degree of the supply-demand deviation between photovoltaic output and park load are further determined. The power balance state itself already contains information on the direction of the supply-demand deviation, i.e., whether there is excess or deficiency. The degree of deviation can be quantified by the absolute value of the real-time power difference or the predicted deviation value after filtering. Specifically, the difference between the current real-time photovoltaic output and the real-time park load can be used as the basis for the degree of deviation, and then smoothed and corrected by combining the mean or peak value of the predicted power difference for future periods to obtain the final value of the supply-demand deviation.

[0056] The energy allocation identity of each energy object is then determined based on the direction of the supply-demand deviation. When the supply-demand deviation is in the form of power surplus, the energy storage system is assigned the charging identity, and the electric vehicle charging pile is assigned the charging identity to absorb excess power in advance. The grid interface is assigned the power feeding identity to feed surplus power into the grid. When the supply-demand deviation is in the form of power shortage, the energy storage system is assigned the discharging identity, the vehicles with V2G capability in the electric vehicle charging pile are assigned the discharging identity to feed power back to the park, and the grid interface is assigned the power purchasing identity to supplement power from the grid.

[0057] The power target range for each energy source is then determined based on the degree of supply-demand imbalance. For scenarios with excess power, the excess power is used as the total amount to be allocated, and power caps are set for energy storage system charging and electric vehicle charging, respectively, while no upper limit is set for grid feed. The power cap for energy storage system charging is determined by the maximum charging power and remaining capacity of the energy storage system, while the power cap for electric vehicle charging is determined by the maximum charging power demand of the connected vehicles.

[0058] For scenarios with insufficient power, the shortfall in power is considered as the total amount to be supplemented. Separate upper limits are set for the discharge power of the energy storage system and the V2G discharge power of the electric vehicle, while there is no upper limit for grid power purchases. Specifically, the upper limit for the discharge power of the energy storage system is determined by the maximum discharge power of the energy storage system and the current available power, while the upper limit for the V2G discharge power of the electric vehicle is determined by the maximum V2G discharge capacity of the connected vehicles.

[0059] Furthermore, the power target range of each energy object is used as the feasible domain boundary for subsequent energy allocation strategy formulation, so as to avoid the subsequent optimization allocation results from exceeding the physical limitations or safe operation boundaries of the equipment and ensure the reliability of the final allocation command.

[0060] Based on this, the fluctuation of carbon emission intensity, carbon emission threshold, and supply-demand deviation information are uniformly quantified into a continuously adjustable scheduling parameter, so that subsequent energy allocation can respond to real-time changes in carbon emission intensity and achieve synergistic optimization of the economy and low carbon emissions in energy allocation.

[0061] The generation of dynamic low-carbon weights based on power balance state, dynamic carbon emission intensity, and preset carbon emission thresholds includes: Identify the direction and degree of supply-demand deviation between photovoltaic power output and park load based on power balance status; The basic polarity of the weights of each energy object is set based on the direction of the supply and demand deviation, and the correction magnitude is calculated based on the relative deviation between the degree of supply and demand deviation and the dynamic carbon emission intensity and the preset carbon emission threshold. Multiply the basic polarity of the weight by the correction magnitude, and then add it to the preset basic weight to obtain the dynamic low-carbon weight.

[0062] First, based on the power balance state, identify the corresponding direction and degree of supply-demand deviation. The identification method is the same as the identification method when matching energy allocation targets.

[0063] The basic polarity of the weights for each energy object is set according to the direction of the supply-demand deviation. When the supply-demand deviation is in the direction of power shortage, the basic polarity of the weights is set to positive (+1), indicating that the priority of low-carbon measures needs to be strengthened, that is, the priority of low-carbon measures, such as V2G discharge and energy storage discharge, should be increased in subsequent dispatch. When the supply-demand deviation is in the direction of power surplus, the basic polarity of the weights is set to negative (-1), indicating that the priority of low-carbon measures needs to be weakened, that is, the priority of low-carbon measures should be appropriately reduced, and priority should be given to absorbing surplus electricity or economic dispatch.

[0064] Simultaneously, the dynamic carbon emission intensity is subtracted from the preset carbon emission threshold, and the difference is divided by the preset carbon emission threshold to obtain a relative deviation value that can be positive or negative. This relative deviation reflects the direction and degree of deviation of the current carbon emission intensity from the benchmark value.

[0065] The correction magnitude is obtained by weighting and summing the supply-demand deviation value and the relative deviation value. The weighting coefficients used in the weighted summation of the supply-demand deviation value and the relative deviation value can be preset according to actual operational needs.

[0066] Multiplying the base polarity of the weight by the correction magnitude yields an adjustment amount, which is then added to the preset base weight to obtain the initial dynamic low-carbon weight. The preset base weight can be set according to the zero-carbon target level of the zero-carbon park; the higher the zero-carbon target level, the higher the preset base weight. The initial dynamic low-carbon weight is then constrained, limiting its value to a closed interval between 0 and 1 to output the final dynamic low-carbon weight.

[0067] The obtained dynamic low-carbon weight increases with the increase of carbon emission intensity and power deficit, and decreases with the decrease of carbon emission intensity or power surplus. Through the dynamic low-carbon weight, the energy allocation and scheduling priority parameters of each energy object can be dynamically adjusted.

[0068] Based on quantifying the overall power supply and demand of the microgrid through energy allocation targets and quantifying the scheduling priority of different energy entities through dynamic low-carbon weights, energy allocation and scheduling planning is carried out to determine the specific operating modes and power allocation strategies of various energy entities.

[0069] Meanwhile, considering that photovoltaic power will generate energy loss during inverter conversion and has low utilization efficiency, the photovoltaic direct supply rule is used as a constraint when carrying out energy allocation planning to limit the priority supply path of photovoltaic power, reduce power conversion loss, and ensure the reliability of energy allocation planning.

[0070] The aforementioned energy allocation strategy for the industrial park, based on direct photovoltaic power supply rules as constraints, energy allocation targets, and dynamic low-carbon weights, includes: Based on the energy allocation objectives, obtain the energy allocation identity and corresponding power target range for each energy object; The direct photovoltaic supply rule serves as the constraint for the energy allocation order, and the dynamic low-carbon weight serves as the priority adjustment coefficient. The execution order of each energy object is set according to its corresponding energy allocation identity. Based on the power target range of each energy object and the revised execution order, the power allocation ratio is determined to obtain the park's energy allocation strategy.

[0071] Based on the energy allocation target, the energy allocation identity of each energy object, determined according to the power balance state, and the range of power targets that can be executed under that identity are extracted.

[0072] The photovoltaic direct supply rule specifically stipulates that photovoltaic DC-side power is prioritized for direct supply to the park's DC load, and any remaining power after satisfying the DC load is redistributed to other energy sources. Based on this rule, each energy source is divided into two levels. The first level is the park's DC load, which always has priority in energy allocation and does not participate in subsequent order adjustments. The second level includes energy storage systems, electric vehicle charging piles, and grid interfaces. A dynamic low-carbon weight is used as the priority adjustment coefficient, and the execution order of energy control for each energy source within the second level is modified according to their corresponding energy allocation status.

[0073] Among them, the dynamic low-carbon weight is a dimensionless coefficient with a value between 0 and 1. The larger the value, the higher the priority of low carbon.

[0074] When the power is insufficient and each energy source is assigned either as a discharger or a purchaser, if the dynamic carbon weight is greater than the preset high threshold, the execution order is: energy storage discharge, electric vehicle V2G discharge, and grid purchase. If the dynamic low carbon weight is less than the preset low threshold, the execution order is: grid purchase, energy storage discharge, and electric vehicle V2G discharge. If the dynamic low carbon weight is between the high and low thresholds, the execution order is: energy storage discharge, grid purchase, and electric vehicle V2G discharge.

[0075] When there is a power surplus and each energy object is assigned the status of charging or power feeding, the execution order of energy storage charging, electric vehicle V2G charging and grid power feeding will be directly executed.

[0076] Next, the total power to be allocated is determined based on the degree of supply-demand deviation. Following the set execution order, each energy object is traversed from front to back. For the current object, the upper limit of its power target range is used as the proposed power value. If the proposed power value is less than or equal to the current remaining power to be allocated, the entire proposed power value is allocated to the current object, and the remaining power to be allocated is updated by subtracting this value. If the proposed power value is greater than the current remaining power to be allocated, only the remaining power to be allocated is allocated to the current object, and no further allocations are made to subsequent objects.

[0077] After the allocation is completed, each energy object receives a specific power instruction value. These power instruction values ​​are arranged in the corresponding execution order to form the final energy allocation strategy for the park.

[0078] The execution order in the energy allocation strategy for the industrial park is directly determined by dynamic low-carbon weights. When carbon emission intensity is high, grid power purchases are automatically placed last, while when carbon emission intensity is low, power purchases are moved forward. This allows the park to proactively reduce its dependence on high-carbon power sources during periods of high carbon emissions, thereby reducing overall carbon emissions. Simultaneously, direct photovoltaic power supply rules are used as the highest priority constraint to ensure that photovoltaic DC power is preferentially supplied to the park's DC loads, avoiding execution failures or additional inverter losses due to conflicts between dispatch instructions and physical wiring sequences.

[0079] After obtaining the park's energy allocation strategy, the power command value of each energy object is converted into a control signal that can be identified by each energy object, and then sent to the energy storage system, electric vehicle charging pile and grid interface through the communication network.

[0080] After receiving a power command value, the energy storage system determines the charging / discharging direction based on the sign of the command value: a positive value indicates charging, and a negative value indicates discharging. The bidirectional DC / DC converter of the energy storage system adjusts the charging or discharging power according to the command value, while simultaneously monitoring changes in the energy storage's state of charge to ensure that the charging and discharging process does not exceed the safe operating boundaries.

[0081] After receiving a power command value, the electric vehicle charging station determines its operating mode based on the sign of the command value: a positive value indicates charging the vehicle, while a negative value indicates the vehicle discharging into the charging station (V2G mode). The charging station converts the command value into a charging / discharging power request from the vehicle and communicates with the vehicle through the charging interface to control the vehicle to perform the corresponding charging or discharging operation. For vehicles without V2G capability, only positive commands are accepted; negative commands are ignored or set to zero.

[0082] After receiving the power command value, the grid interface determines the power flow direction based on the sign of the command value: a positive value indicates power purchase from the grid, and a negative value indicates power feedback to the grid. The bidirectional inverter at the grid interface adjusts the switching power according to the command value, while simultaneously monitoring the grid voltage and frequency to ensure that the switching power remains within the safe and permissible range for grid connection.

[0083] After each energy object performs corresponding energy control according to the park's energy allocation strategy, the following also applies: The system collects the actual operating parameters of the microgrid in real time and corrects the dynamic low-carbon weight and preset carbon emission threshold based on the deviation between the actual operating parameters and the source-load prediction results.

[0084] During the execution process, the actual fluctuation value of photovoltaic output and the actual load value of the park are collected in real time.

[0085] The photovoltaic output average value collected during the current scheduling cycle is compared with the photovoltaic output forecast value to obtain the photovoltaic output deviation. The actual average load value of the park is compared with the load forecast value to obtain the load deviation.

[0086] When the photovoltaic output deviation is negative and the load deviation is positive, it indicates that the actual deficit is greater than expected, requiring an increase in low-carbon priority. Therefore, the dynamic low-carbon weight is adjusted upwards by one step, further advancing V2G discharge or energy storage discharge in the sequence, prioritizing the use of low-carbon resources to fill the additional deficit, thereby maintaining the low-carbon emission target. Conversely, when the photovoltaic output deviation is positive and the load deviation is negative, it indicates that the actual deficit is less than expected, and the dynamic low-carbon weight can be adjusted downwards by one step. The step size is determined based on the deviation magnitude; the larger the deviation, the larger the step size.

[0087] Meanwhile, based on the actual operating parameters of the microgrid, the actual carbon emissions of the park are calculated. When the actual carbon emissions of the park exceed the target value for several consecutive scheduling cycles, the preset carbon emission threshold is adjusted downward, making it easier to trigger an increase in the dynamic low-carbon weight in subsequent cycles. Furthermore, when adjusting the preset carbon emission threshold, since grid feeding does not affect the change in grid carbon emissions, deviations in the feeding portion do not affect the correction of the preset carbon emission threshold.

[0088] The dynamic low-carbon weight and preset carbon emission threshold are adjusted online in real time based on actual operating results, so that the energy allocation strategy can dynamically adapt to long-term changes such as changes in park load, fluctuations in photovoltaic output, and equipment performance degradation, ensuring the reliability of low-carbon energy control in the zero-carbon park.

[0089] Another aspect of this embodiment also provides an energy control system for a photovoltaic-storage-charging microgrid in a zero-carbon park. This system is communicatively connected to the photovoltaic-storage-charging microgrid located in the zero-carbon park. It can collect corresponding operating data and issue corresponding control commands to achieve energy control of each energy object within the photovoltaic-storage-charging microgrid.

[0090] The energy control system includes: The data processing module is used to predict source load based on real-time operating data of the microgrid and corresponding environmental parameters, and to calculate dynamic carbon emission intensity. The status judgment module is used to identify the power balance status of the microgrid based on real-time photovoltaic output, real-time load of the park, and source-load prediction results, and to match energy allocation targets according to the power balance status. The energy allocation module is used to generate dynamic low-carbon weights based on power balance status, dynamic carbon emission intensity, and preset carbon emission thresholds. It also uses photovoltaic direct supply rules as constraints to formulate park energy allocation strategies based on energy allocation targets and dynamic low-carbon weights.

[0091] The data processing module is connected to the status determination module, and the energy distribution module is connected to both the data processing module and the status determination module. The data processing module, status determination module, and energy distribution module are all devices such as computers and microprocessors with corresponding data processing capabilities.

[0092] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any way. Other variations and modifications are possible without departing from the technical solutions described in the claims.

Claims

1. An energy control method for photovoltaic-storage-charging microgrids in zero-carbon industrial parks, characterized in that: include: Based on real-time operation data of the microgrid, source load prediction is performed in combination with corresponding environmental parameters, and dynamic carbon emission intensity is calculated. Based on real-time photovoltaic output, real-time load of the park, and source-load prediction results, the power balance status of the microgrid is identified, and energy allocation targets are matched according to the power balance status. Dynamic low-carbon weights are generated based on power balance state, dynamic carbon emission intensity, and preset carbon emission thresholds. Based on the direct photovoltaic power supply rules and energy allocation targets, and combined with dynamic low-carbon weights, the park's energy allocation strategy is formulated. Each energy source will implement corresponding energy control according to the park's energy allocation strategy.

2. The energy control method for a photovoltaic-storage-charging microgrid for zero-carbon industrial parks according to claim 1, characterized in that, The process of identifying the power balance state of the microgrid based on real-time photovoltaic output, real-time load of the industrial park, and source-load prediction results includes: The real-time power difference is calculated based on the real-time photovoltaic output and the real-time load of the park, and candidate power balance states are identified by combining the preset state thresholds. Based on the source-load prediction results, the predicted power difference for each future time period is obtained, and the identified candidate power balance state is verified based on the predicted power difference for each future time period. The current power balance status of the microgrid is determined based on the verification results.

3. The energy control method for a photovoltaic-storage-charging microgrid for zero-carbon industrial parks according to claim 2, characterized in that, The step of verifying the identified candidate power balance states based on the predicted power differences in each future time period includes: Based on the predicted power difference at each future time, and combined with the preset state threshold, the power balance state corresponding to each predicted power difference is determined. Based on the distribution ratio and corresponding duration of power balance states for all future time periods, the steady-state supply and demand state for future time periods is determined. When the steady-state supply and demand state is consistent with the candidate power balance state, the judgment is passed and the candidate power balance state is taken as the current power balance state of the microgrid. If the steady-state supply and demand state is inconsistent with the candidate power balance state, the test is deemed to have failed, and the candidate power balance state is adjusted according to the steady-state supply and demand state.

4. The energy control method for a photovoltaic-storage-charging microgrid for zero-carbon industrial parks according to claim 1, characterized in that, The method of matching energy allocation targets based on power balance status includes: Identify the direction and degree of supply-demand deviation between photovoltaic power output and park load based on power balance status; Based on the direction and degree of supply-demand deviation, the energy allocation identity and corresponding power target range of each energy object are matched.

5. The energy control method for a photovoltaic-storage-charging microgrid for zero-carbon industrial parks according to claim 1, characterized in that, The generation of dynamic low-carbon weights based on power balance state, dynamic carbon emission intensity, and preset carbon emission thresholds includes: Identify the direction and degree of supply-demand deviation between photovoltaic power output and park load based on power balance status; The basic polarity of the weights of each energy object is set based on the direction of the supply and demand deviation, and the correction magnitude is calculated based on the relative deviation between the degree of supply and demand deviation and the dynamic carbon emission intensity and the preset carbon emission threshold. Multiply the basic polarity of the weight by the correction magnitude, and then add it to the preset basic weight to obtain the dynamic low-carbon weight.

6. The energy control method for a photovoltaic-storage-charging microgrid for zero-carbon industrial parks according to claim 1, characterized in that, The aforementioned energy allocation strategy for the industrial park, based on direct photovoltaic power supply rules and energy allocation targets, and combined with dynamic low-carbon weights, includes: Based on the energy allocation objectives, obtain the energy allocation identity and corresponding power target range for each energy object; The direct photovoltaic supply rule serves as the constraint for the energy allocation order, and the dynamic low-carbon weight serves as the priority adjustment coefficient. The execution order of each energy object is set according to its corresponding energy allocation identity. Based on the power target range of each energy object and the revised execution order, the power allocation ratio is determined to obtain the park's energy allocation strategy.

7. The energy control method for a photovoltaic-storage-charging microgrid for zero-carbon industrial parks according to claim 1, characterized in that, After each energy object performs corresponding energy control according to the park's energy allocation strategy, the following also applies: The system collects the actual operating parameters of the microgrid in real time and corrects the dynamic low-carbon weight and preset carbon emission threshold based on the deviation between the actual operating parameters and the source-load prediction results.

8. The energy control method for a photovoltaic-storage-charging microgrid for zero-carbon industrial parks according to claim 1, characterized in that, The source-load prediction based on real-time microgrid operation data, combined with corresponding environmental parameters, includes: Based on the real-time operation data of the microgrid and combined with the corresponding environmental parameters, the predicted values ​​of photovoltaic power output and park load for future periods are obtained through time series forecasting. Based on the real-time operation data of the microgrid, the current operating condition of the microgrid is identified, and the corresponding operating condition deviation is matched according to the current operating condition of the microgrid. Based on the matching of environmental parameters with corresponding historical photovoltaic power output data, and combined with real-time operation data, the power output deviation value is obtained; The current performance degradation level of the power generation equipment is obtained based on the output deviation value, and the corresponding degradation deviation amount is matched according to the performance degradation level of the power generation equipment. The predicted load value of the park is corrected based on the operating condition deviation, and the predicted photovoltaic output value is corrected based on the attenuation deviation.

9. The energy control method for a photovoltaic-storage-charging microgrid for zero-carbon industrial parks according to claim 1, characterized in that, The calculation of dynamic carbon emission intensity includes: Based on the grid-connected power type of the microgrid, match the corresponding unit carbon emission benchmark value; The power consumption of each type of grid-connected electrical energy is obtained based on the real-time operation data of the microgrid, and the real-time dynamic carbon emission intensity is calculated by combining the corresponding unit carbon emission benchmark value.

10. An energy control system for a photovoltaic-storage-charging microgrid in a zero-carbon industrial park, used to execute the energy control method according to any one of claims 1 to 9, characterized in that, include: The data processing module is used to predict source load based on real-time operating data of the microgrid and corresponding environmental parameters, and to calculate dynamic carbon emission intensity. The status judgment module is used to identify the power balance status of the microgrid based on real-time photovoltaic output, real-time load of the park, and source-load prediction results, and to match energy allocation targets according to the power balance status. The energy allocation module is used to generate dynamic low-carbon weights based on power balance status, dynamic carbon emission intensity, and preset carbon emission thresholds. It also uses photovoltaic direct supply rules as constraints to formulate park energy allocation strategies based on energy allocation targets and dynamic low-carbon weights.

Images