A micro-grid transient energy distribution system based on a fusion terminal

Abstract

The application relates to the technical field of micro-grid control, in particular to a micro-grid instantaneous energy distribution system based on a fusion terminal, which comprises the following units: a data acquisition unit which receives grid-connected point metering data, bus electrical quantity, distributed power supply state, energy storage state, controllable load state and power feedback to form period section data; a data synchronization unit which calculates validity weight according to a sampling time label, clock deviation and data delay time and aligns to a fixed period end time or an event triggering time; a deficiency calculation unit which combines a pre-release reference power, a target power, an uncompleted amount and an on-the-way adjustment amount in an instruction record table to determine a residual net power deficiency; a power distribution unit which generates an adjustable power range and distributes a new adjustment amount according to equipment power range, climbing range and state of charge; and an instruction release unit which updates the instruction record table and corrects a next decision period target power; the application improves the synchronism and execution accuracy of micro-grid instantaneous energy distribution.

Description

Technical Field

[0001] This invention relates to the field of microgrid control technology, specifically to a microgrid instantaneous energy distribution system based on a fusion terminal. Background Technology

[0002] Existing microgrid energy allocation typically relies on centralized decision-making based on grid connection point metering, bus measurement, and feedback information from various distributed resources. In related technologies, when there are communication delays, inconsistent sampling windows, unclosed loops in command execution, or failures in some devices, the latest data often has to be used directly for calculation, or the results from the previous cycle can be simply used for correction. However, when the arrival times of multi-source data are inconsistent, the equipment status is constantly changing, and the coordinated regulation of energy storage, distributed power sources and controllable loads needs to be considered simultaneously, existing solutions are difficult to accurately identify the current actual adjustable capacity and are also difficult to avoid the mutual interference between historical command occupancy and current deficit calculation. Therefore, the microgrid instantaneous energy allocation method in related technologies is difficult to balance real-time performance, accuracy and allocation continuity. Summary of the Invention

[0003] The purpose of this invention is to provide a microgrid instantaneous energy allocation system based on a fusion terminal, avoiding interference between historical command occupancy and current deficit calculation, and more easily balancing the real-time performance, accuracy, and continuity of energy allocation; this objective can be achieved through the following technical solutions: The data acquisition unit is used to receive grid connection point metering data, bus electrical quantities and equipment status data of the microgrid within a preset decision cycle, and form periodic cross-sectional data according to the receiving cutoff time. The data synchronization unit is used to extract the clock deviation and measurement characteristics of the periodic cross-sectional data, determine the validity weight by combining the data delay time, and align the valid data to a unified time reference to form the instantaneous energy state. The deficit calculation unit is used to match the instantaneous energy state with the system command record table to assess the uncompleted amount, combine the equipment response capability and effectiveness weight to determine the in-transit adjustment amount, and calculate the remaining net power deficit after deducting the in-transit adjustment amount. The power allocation unit is used to determine the adjustable power range based on equipment operating constraints and effectiveness weights, and to allocate the new regulation amount to each device in the microgrid within the adjustable power range; The instruction issuing unit is used to issue power instructions containing new adjustment amounts, update the instruction status based on communication feedback, and calculate the execution deviation to correct the subsequent target power.

[0004] Optionally, the unified time reference includes the end time of a fixed cycle set by the data acquisition unit according to the decision cycle, and the event triggering time triggered by microgrid electrical parameters exceeding limits or islanding switching; The data acquisition unit truncates and divides the cross-sectional data of the current period and the next period based on the receiving cutoff time.

[0005] Optionally, the data synchronization unit obtains the sampling end time of the corresponding acquisition device's source end, adds the deviation of the source end clock relative to the system fusion terminal clock, and obtains the converted sampling end time. The difference between this converted sampling end time and the alignment time corresponding to the unified time reference is the data delay time. ; Preset reference delay time and maximum allowable delay time ,and ; when At that time, validity weight ; when At that time, validity weight ; when When this happens, the corresponding data will be marked as invalid and ; For the average power data of the metering window, the validity weight is adjusted according to the ratio of the metering window length to the preset standard window length.

[0006] Optionally, the deficit calculation unit determines the theoretical adjustment limit based on the equipment ramp-up rate and the remaining execution time of the instructions in the system instruction record table, and then reduces the theoretical adjustment limit by the validity weight before comparing it with the uncompleted amount to determine the actual adjustment amount in transit. The instruction issuing unit dynamically tracks the lifecycle status of an instruction from the point of being sent to completion or failure through the system instruction record table.

[0007] Optionally, when calculating the remaining net power deficit, the deficit calculation unit may selectively deduct or algebraically sum from the remaining net power deficit based on the consistency of the increase / decrease direction between the in-transit adjustment and the power deficit before deducting the in-transit adjustment. Expired instructions that have not been completed within the time limit will continue to occupy the adjustable power range of the corresponding device in the same adjustment direction, and it is prohibited to repeatedly issue instructions in the same direction with a target power difference within the preset tolerance to devices with unknown status before the instruction is clearly fed back.

[0008] Optionally, the remaining net power deficit The calculation formula is:

[0009] in, For load power, For network loss, For distributed power supply, For energy storage power, Reference power for grid connection point, These are in-transit adjustments in the same direction, during islanded network operation. .

[0010] Optionally, the power distribution unit can allocate power to the microgrid devices in a hierarchical manner according to their response speed from fastest to slowest. When the adjustable power range of the upper-level device is exhausted, the inter-level switch to the lower-level device is triggered, and the new adjustment amount is determined with the constraints of minimizing resource call amount, minimizing the change amount of instructions in adjacent cycles, and minimizing the unallocated deficit.

[0011] Optionally, when power feedback fails or the mean of data validity weights falls below a preset confidence threshold, the system triggers a degradation strategy: Stop adding new global power allocations, switch control energy storage devices and adjustable distributed power sources to local safety control mode, and maintain the existing state of controllable loads; After communication is restored, the system will resume allocating the newly added adjustment amounts to the corresponding devices based on the latest instruction feedback status and mode flags.

[0012] Compared with the prior art, the present invention has the following advantages: 1. This invention extracts the clock deviation and metering characteristics of cross-sectional data, dynamically determines the validity weight by combining data delay time and window length, and aligns the data to a unified time reference; it effectively solves the data distortion problem caused by inconsistent arrival of multi-source data and communication delay, accurately restores the real instantaneous energy state of the microgrid, and improves the real-time performance of decision-making; 2. This invention matches the instantaneous state with the instruction record table to evaluate the incomplete quantity, combines the equipment response capability and effectiveness weight to determine the in-transit adjustment quantity, and performs precise deduction or algebraic summation when calculating power deficit; it avoids the mutual interference between historical instruction occupation and current deficit calculation, prevents repeated instruction issuance to equipment with unknown status, and ensures the continuity of allocation; 3. This invention designs a degradation protection strategy for abnormal operating conditions; when power feedback fails or the average value of data validity weight is lower than the preset confidence threshold, the system automatically stops adding new global power allocations and controls the device to switch to local safety control mode; effectively avoiding the device from executing erroneous instructions in communication blind spots and improving the risk resistance capability in extreme environments. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of a microgrid instantaneous energy distribution system based on a fusion terminal, as described in an embodiment of the present invention. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0015] Please see Figure 1 A microgrid instantaneous energy distribution system based on a fusion terminal, comprising: The data acquisition unit is used to receive grid connection point metering data, bus electrical quantities and equipment status data of the microgrid within a preset decision cycle, and form periodic cross-sectional data according to the receiving cutoff time. The data synchronization unit is used to extract the clock deviation and measurement characteristics of the periodic cross-sectional data, determine the validity weight by combining the data delay time, and align the valid data to a unified time reference to form the instantaneous energy state. The deficit calculation unit is used to match the instantaneous energy state with the system command record table to assess the uncompleted amount, combine the equipment response capability and effectiveness weight to determine the in-transit adjustment amount, and calculate the remaining net power deficit after deducting the in-transit adjustment amount. The power allocation unit is used to determine the adjustable power range based on equipment operating constraints and effectiveness weights, and to allocate the new regulation amount to each device in the microgrid within the adjustable power range; The instruction issuing unit is used to issue power instructions containing new adjustment amounts, update the instruction status based on communication feedback, and calculate the execution deviation to correct the subsequent target power. The unified time reference includes the end time of a fixed cycle set by the data acquisition unit according to the decision cycle, as well as the event trigger time triggered by microgrid electrical parameters exceeding limits or islanding switching; The data acquisition unit truncates and divides the cross-sectional data of the current period and the next period based on the receiving cutoff time; The data synchronization unit obtains the sampling end time of the corresponding acquisition device's source end, adds the deviation of the source end clock relative to the system fusion terminal clock, and obtains the converted sampling end time. The difference between this converted sampling end time and the alignment time corresponding to the unified time reference is the data delay time. ; Preset reference delay time and maximum allowable delay time ,and ; when At that time, validity weight ; when At that time, validity weight ; when When this happens, the corresponding data will be marked as invalid and ; For the average power data of the metering window, the validity weight is adjusted according to the ratio of the metering window length to the preset standard window length.

[0016] The integrated terminal connects to the grid-connected metering device, bus-side measuring device, energy storage converter, distributed power controller, and controllable load controller. Feedback information from various devices arrives according to their respective sampling cycles and communication links. The system determines the receiving deadline within a fixed decision cycle and saves the data arriving before the deadline as the current cycle's cross-sectional data; data arriving after the deadline is not written back to the current cycle but enters the next cycle for processing. After the current cycle section data is formed, the data synchronization unit converts the sampling time label, metering window and clock deviation of each data to obtain the sampling end time corresponding to the unified time reference of the fusion terminal, and then calculates the data delay time accordingly. For data with a delay time no greater than the reference delay time, its full weight is retained; for data with a delay time between the reference delay time and the maximum allowable delay time, its weight is reduced linearly; data exceeding the maximum allowable delay time is directly marked as failed and no longer participates in the power allocation of the current cycle. For data such as average power at grid connection points that is based on metering windows, the weight is adjusted in conjunction with the window length. The aforementioned window length adjustment is based on the ratio of the standard window length to the actual metering window length: when the metering window length is not greater than the preset standard window length, the original validity weight remains unchanged. When the measurement window length exceeds the preset standard window length, the original validity weight is multiplied by the ratio of the standard window length to the actual measurement window length, ensuring that the corrected weight is no greater than the original validity weight; let the corrected validity weight be... The original validity weight was The actual metering window length is The default standard window length is Its quantification correction formula is:

[0017] The validity weight after delay and window correction ranges from zero to one. The larger the value, the more representative the data is of the current alignment state. After completing the weight calculation, the system aligns the weighted effective data to the end of a fixed period, or when the power, frequency, or bus voltage at the grid connection point exceeds the limit or when the grid is switched to an isolated grid, it regenerates the periodic cross-sectional data according to the event trigger time. For the power feedback of a single acquisition device, the system retains its converted power value and writes the effectiveness weight as the corresponding feedback credibility into the instantaneous energy state record. When there are multiple valid measurement sources for the same physical quantity, the system only performs weighted synthesis according to the proportion of each validity weight to the total valid weight when the dimensions and measurement objects are consistent; if there is no valid measurement source, the system marks the physical quantity as feedback failure and does not use historical values ​​to replace the current power feedback in the deficit calculation. The aligned results are written into the instantaneous energy state record of the current cycle for direct use by the subsequent deficit calculation unit; for data that are downweighted due to a delay time greater than the reference delay time or an actual metering window length greater than the preset standard window length, the system uses them together with other available measurements to form the power judgment for the current cycle. Once the instantaneous energy state is formed, the deficit calculation unit matches the power feedback in it with the baseline power before the instruction record table. It first identifies the adjustment amount that has been executed, and then determines the uncompleted amount based on the difference between the target power and the current power feedback. For instructions that are still being executed, the system estimates the amount of in-transit regulation that can be implemented within the current control time domain by combining the equipment ramp rate and the remaining execution time, and deducts the part in the same direction from the current remaining deficit; the remaining net power deficit does not include the regulation that has been clearly executed and is expected to be completed soon, and the subsequent power allocation unit makes new allocations based on the remaining net power deficit; The judgment of the aforementioned executed adjustment amount, incomplete amount and in-transit adjustment amount is based on the reference power before the issuance in the same instruction record as the comparison base point, and is checked one by one in combination with the power feedback that has been aligned in the current cycle, so as to ensure that the status of the same equipment is consistent in different cycles. The target power, the reference power before issuance, and the current power feedback all use the same power symbol system and all correspond to the active power on the same equipment side. If the target power and the current power feedback of a certain record belong to different measurement points, they are first converted to the equipment control measurement point corresponding to the instruction record before comparison. The power distribution unit determines the initial range based on the equipment power range, ramp range, and state of charge, then deducts the amount of unfulfilled instructions and reduces it according to the effectiveness weight to form an adjustable power range; the new adjustment amounts of energy storage devices, adjustable distributed power sources, and controllable loads are all calculated within this range; After the allocation result is written to the instruction issuing unit, the system issues a power instruction containing the new adjustment amount. At the same time, it updates the instruction record table based on the received response and power feedback, recording the instruction status, feedback status, and execution deviation. The execution deviation is used to correct the target power for the next decision cycle. When an instruction fails to be completed within the current cycle and there is insufficient remaining execution time, the system retains the uncompleted portion and continues to occupy the adjustable range of the corresponding device until clear feedback is received or the validity period expires. If the feedback delay time of a certain device is greater than the reference delay time, its effectiveness weight is reduced, and the allocation ratio of the corresponding device in the current period is correspondingly reduced. The system selects only one alignment reference within the same decision cycle; once an event is triggered and the cycle cross-sectional data is reconstructed accordingly, the cross-sectional result corresponding to the event trigger time replaces the cross-sectional result at the end of the original fixed cycle and participates in the subsequent calculations of this cycle.

[0018] In one embodiment, the deficit calculation unit determines the theoretical adjustment limit based on the equipment ramp-up rate and the remaining execution time of the instructions in the system instruction record table, and then reduces the theoretical adjustment limit by the validity weight and compares it with the uncompleted amount to determine the actual adjustment amount in transit. The instruction issuing unit dynamically tracks the lifecycle status of instructions from when they are to be sent to when they are completed or have failed through the system instruction record table; When calculating the remaining net power deficit, the deficit calculation unit selectively subtracts from or algebraically sums from the remaining net power deficit based on the consistency of the increase / decrease direction between the in-transit adjustment and the power deficit before deducting the in-transit adjustment. Expired instructions that have not been completed within the time limit will continue to occupy the adjustable power range of the corresponding device in the same adjustment direction, and it is prohibited to repeatedly issue instructions in the same direction with a target power difference within the preset tolerance to devices with unclear status before the instruction is clearly fed back. Remaining net power deficit The calculation formula is:

[0019] in, For load power, For network loss, For distributed power supply, For energy storage power, Reference power for grid connection point, These are in-transit adjustments in the same direction, during islanded network operation. .

[0020] When the output of distributed power sources declines or the load suddenly increases, the converged terminal first identifies which adjustments have been completed and which are still pending, in progress, or in an unclear state based on the instruction record table. Each record in the instruction record table maintains fixed fields, including instruction number, target power, baseline power before issuance, issuance time, planned effective time, validity period, and response and feedback information. The system updates the status flow based on these fields, with the status progressing from pending to received and pending execution, in execution, and completed. If no feedback is received within the confirmation period, the status changes to unknown. If the device explicitly rejects the request, the status changes to execution failure. When calculating the deficit, the system does not directly treat all uncompleted amounts as immediately allocable capacity, but first calculates the adjustment limit. The system incorporates the inherent dead time of the device's response, which includes mechanical inertia and motion hysteresis. When the time elapsed since the instruction was issued is less than or equal to At that time, the corresponding in-transit adjustment is forcibly latched to zero; When the time elapsed since the instruction was issued is greater than At that time, the upper limit of adjustment is obtained by multiplying the equipment ramp rate and the remaining execution time, and then reduced by the effectiveness weight. The smaller value is taken as the in-transit adjustment amount. The dimension of the equipment ramp rate is power divided by time, and then multiplied by the remaining execution time to obtain the power amount. The effectiveness weight is a dimensionless coefficient between zero and one, which still retains the power dimension after reduction. The system calculates the absolute values ​​of the uncompleted amount and the upper limit of adjustment according to the direction of increasing power and decreasing power respectively. Within the same direction, the smaller value of the two is taken as the amplitude of the in-transit adjustment amount, and then its sign is restored according to the direction. The portion of the unfinished work that cannot be completed within this control time domain will not be deducted, while the portion that can be completed within the time domain will be deducted from the remaining net power deficit. For determining the amount of incomplete work, the system uses the baseline power recorded at the previous moment before the issuance as the comparison base point, and compares the change in the current power feedback relative to the comparison base point with the adjustment change corresponding to the target power. When the two are in the same direction and the current change has not reached the target change, the difference is recorded as the amount of incomplete work. If the current change has reached or exceeded the target change, the instruction will not generate a new amount of incomplete work. The aforementioned target change amount represents the change in target power relative to the baseline power before issuance, and the current change amount represents the change in current power feedback relative to the baseline power before issuance; the aforementioned equipment ramp rate represents the power change rate that the corresponding equipment is allowed to reach per unit time, and the remaining execution time represents the remaining time from the current decision time until the end of the effective period of this instruction; The target power, load power, distributed power generation power, energy storage power, grid connection point reference power, and remaining net power deficit are all represented using the same power symbol system. The actual feedback of the equipment is expressed as power feedback, the equipment capacity boundary is expressed as adjustable power range, and the short-term energy demand is only used as the result of the power deficit combined with the decision cycle length, and is not used in conjunction with the above power quantities. Before calculating the remaining net power deficit, the system unifies all power quantities to the same alignment time and the same power symbol system; load power and network losses are taken as positive values ​​according to the power consumption side, and distributed generation power is taken as positive values ​​according to the power supply side; energy storage power is taken as positive values ​​when discharging and negative values ​​when charging; the grid connection point reference power is taken as positive values ​​when the grid sends power to the microgrid and negative values ​​when the microgrid sends power to the grid; Network losses are obtained from the power balance difference between the metered power at the grid connection point and the effective power, energy storage power, and load power on the bus side. Considering the matrix characteristics of multi-node devices in the microgrid, the specific calculation formula is as follows:

[0021] in, Measuring power at the grid connection point, For the first Effective power measurement of distributed power sources For the first One energy storage unit effectively measures power. For the first Effective power measurement of each load; This represents the total number of active nodes in the distributed power supply. This represents the total number of activated nodes in the energy storage device. This represents the total number of active nodes in the load. These are the index variables for the summation of the three types of equipment mentioned above; When the effective measurement in the current period is insufficient to obtain the network loss, the value of the most recent period with an effective network loss calculation result is adopted, and the effectiveness weight corresponding to that period is inherited; if the weight is zero, the network loss value is not used in the current period calculation, and the current period is marked as a state of reduced data credibility. The aforementioned network loss is limited to a non-negative value before being substituted into the calculation of the remaining net power deficit; if the result obtained by the power balance difference is negative, the result is recorded as zero and the difference is retained in the data credibility reduction status flag. The aforementioned data credibility reduction state is achieved by setting the credibility flag to 1 in the local configuration register. When the system is in the data credibility reduction state, the locking logic is triggered to maintain the power commands of each controllable device in the current decision cycle unchanged until the network loss calculation result is restored to validity and the credibility flag is reset to 0. For direction determination, the system classifies energy storage discharge, power output increase, and load shedding as power increase directions, and energy storage charging, power output decrease, and load input as power decrease directions. If the in-transit adjustment is consistent with the current deficit direction, it is directly deducted from the remaining net power deficit. If the two directions are opposite, they are merged in an algebraic manner to reflect the offsetting relationship between historical instructions and the current deficit. Specifically, the system defines a positive net power deficit as requiring an increase in supply or a reduction in load, and a negative net power deficit as requiring a reduction in supply or an increase in load. When substituting into the formula for calculation, if the direction of the adjustment of the in-transit adjustment is consistent with the direction of the current remaining net power deficit, then the absolute value of the in-transit adjustment should be substituted as a positive value into the formula. The amount should be deducted; if the two are in opposite directions, the absolute value of the in-transit adjustment should be substituted as a negative value. This allows for algebraic addition in the formula, maintaining consistency in the power dimensions and notation system on both sides of the formula; When an instruction is not completed and the remaining execution time is zero, the adjustment amount will no longer be used to deduct the current remaining net power deficit, but will continue to occupy the adjustable power range of the corresponding device in the same adjustment direction until the effective period expires or clear feedback is received. For instructions with unclear status, the system prohibits issuing new instructions in the same direction to the same device again before a clear result is obtained, provided that the difference in target power does not exceed the preset tolerance. The preset tolerance is used to characterize the maximum allowable difference in target power between two devices in the same direction when they can be regarded as the same adjustment intention. For instructions that fail to be executed, the corresponding target power will no longer be deducted from the current remaining net power deficit as an in-transit adjustment amount. The adjustable power range in the same direction occupied by the failed instruction will be released, and the unrealized portion will be merged back into the remaining net power deficit of the current cycle or the next cycle for redistribution. When calculating the remaining net power deficit, the system uniformly converts the load power, network losses, distributed generation power, energy storage power, and grid connection point reference power to the current decision cycle, and then deducts the in-transit adjustment in the same direction to obtain the result. ; Among them, the grid connection reference power is taken as the target value of grid connection exchange power under the current operating mode: when running in grid connection mode, it is taken as one of the scheduling plan value, demand control value or grid connection exchange target value after the previous cycle is corrected; when running in islanded grid mode, the grid connection reference power is directly set to zero. The reference power at the grid connection point is determined by the scheduling plan value received in real time by the fusion terminal through the upper-level scheduling interface, or it is automatically generated by the fusion terminal according to the maximum demand value set by the local demand control strategy and stored in the local configuration register for the deficit calculation unit to read. The aforementioned grid connection point reference power and grid connection point metering data correspond to the same grid connection interface. The former is the target reference quantity used for allocation calculation, and the latter is the real-time measurement quantity of this interface. The two are registered in the record table and calculation process respectively. Short-term energy demand is then met by this remaining net power deficit. With the preset decision cycle length Multiplying them together gives the decision cycle length. The preferred value is a fixed value between 10ms and 100ms; when a finer time expression is required, it can also be approximated by the discrete integral of the remaining net power deficit within the decision period. The approximate discrete integral formula adopts the trapezoidal rule, smoothly incorporating the final state deficit data from the previous period. The formula is as follows:

[0022] in, For short-term energy demand, The sub-sampling microstep size within the decision-making cycle. The total number of sub-samples within a decision-making cycle. For the first The remaining net power deficit at each sampling time point The index variable is the sub-sampling time, and the boundary value is... Forced inheritance of the decision at the end of the previous fixed decision cycle Value; this short-term energy demand Write it into the scheduling calculation record for the power allocation unit to call in subsequent cycles; In cases where feedback fails, the system maintains a conservative approach to directional consistency: if the direction of the adjustment quantity in transit cannot be confirmed, it is not deducted as a completed quantity, but only its occupancy information for the adjustable boundary is retained; if subsequent feedback confirms that the adjustment quantity has been completed, the corresponding occupancy is released and written back to the instruction record table.

[0023] In one embodiment, the power distribution unit allocates power to the microgrid devices in a hierarchical manner according to their response speed from fastest to slowest. When the adjustable power range of the upper-level device is exhausted, the inter-layer switch to the lower-level device is triggered, and the new adjustment amount is determined with the constraints of minimizing resource call amount, minimizing the change amount of instructions in adjacent cycles, and minimizing the unallocated deficit during allocation. When power feedback fails or the mean of data validity weights falls below a preset confidence threshold, the system triggers a degradation strategy: Stop adding new global power allocations, switch control energy storage devices and adjustable distributed power sources to local safety control mode, and maintain the existing state of controllable loads; After communication is restored, the system will resume allocating the newly added adjustment amounts to the corresponding devices based on the latest instruction feedback status and mode flags.

[0024] Once the remaining net power deficit has been calculated from the instruction log table, the power allocation unit does not distribute it evenly to all devices at the same time. Instead, it calls up devices in layers from fastest to slowest response speed. Energy storage devices and fast power control distributed power sources participate in the allocation first, conventional adjustable distributed power sources intervene, and controllable loads participate after the first two layers are exhausted. The allocation result of each layer is written to the allocation result cache of the current period and then matched with the adjustable range of the next layer; only when the adjustable range of the previous layer cannot cover the remaining deficit, does it proceed to the next layer; Within each layer, the system first determines the maximum deficit that can be covered without exceeding the adjustable power range of each device, and then uses the smallest unallocated deficit as the first determining rule; among the candidate results with the same unallocated deficit, the smallest total resource call is used as the second determining rule. Among the candidate results where the two aforementioned conditions are the same, the third determining rule is to use the sum of the changes in instructions between adjacent cycles of each device that is the smallest; if there are still identical results, the devices are allocated in ascending order of their fixed device serial numbers in the converged terminal. Each layer's new adjustment amount is uniquely allocated according to the aforementioned rules. When the previous layer is insufficient to cover the deficit, the unallocated portion is used as the input for the next layer. The aforementioned inter-layer switching triggering mechanism is as follows: before the allocation calculation, if the evaluation finds that the sum of the adjustable power ranges of all devices in the current adjustment direction is less than the deficit to be allocated in the current layer, then the inter-layer switching is directly triggered; or, after the optimization solution of the current layer is completed, if the calculated unallocated deficit is still greater than zero, then the unallocated deficit is used as the input for the next layer and the switching to the next layer's devices is triggered. When power feedback fails or the historical command status is unclear, the system does not continue to expand the new allocation in the conventional way, but first maintains the corresponding device in the current safe state; only when the current safe state cannot meet the device's own operating constraints will it return to the pre-calibrated rated safe power of the device; only when the energy storage device returning to the rated safe power may still cause the state of charge to exceed the preset safety boundary will its charging and discharging power be reduced. If the average validity weight of all devices participating in the current cycle allocation is lower than the preset confidence threshold, then new global power allocation will be stopped, energy storage devices will switch to droop control, adjustable distributed power sources will switch to power limiting control, and controllable loads will remain in the confirmed state and no new cross-device allocation results will be accepted. In this embodiment, the preset confidence threshold is set to 0.6. In practical applications, the preset confidence threshold can be dynamically adjusted between 0.5 and 0.8 according to the communication channel quality. The mean value of the validity weight ranges from zero to one. If it is lower than this threshold, it means that the current available feedback is insufficient to support the addition of global power allocation. After communication is restored, the system locates the corresponding entry in the instruction record table based on the most recently executed instruction number, and then combines the current power correction status field and the operating mode update direction mark to gradually restore the allocation of new adjustment amounts to the corresponding devices; after completing the entry location, status comparison and direction update during the restoration process, the new allocation logic is reactivated. The current power correction status field is used to identify the correction result of the current power feedback relative to the reference power before the table entry was issued. It is divided into at least three categories: corrected, to be corrected, and cannot be corrected. Equipment in the uncorrectable state will only have its status checked during the recovery period and will not participate in the allocation of new adjustment amounts. When the adjustable power range of the upper-level device is exhausted, the inter-level switching is triggered, and the lower-level device participates in the allocation according to the same constraints. The minimum amount of resource calls, the minimum amount of changes in instructions between adjacent cycles, and the minimum amount of unallocated deficit work together as constraint objectives to make the allocation result converge to a range that can cover the deficit and avoid introducing repeated adjustments. For any unallocated deficits, the system retains them in the target power correction item of the next cycle. If feedback from a device is missing for several consecutive decision cycles, its effectiveness weight continues to decrease, and the corresponding adjustable power range shrinks accordingly. After communication is restored, the system restores the data alignment of the device's instruction record table and restores the adjustment authority in the corresponding direction according to the latest operating mode.

[0025] The above-described embodiments do not constitute a limitation on the scope of protection of this technical solution; any modifications, equivalent substitutions, and improvements made within the spirit and principles of the above embodiments should be included within the scope of protection of this technical solution.

Claims

1. A microgrid instantaneous energy distribution system based on a fusion terminal, characterized in that, include: The data acquisition unit is used to receive grid connection point metering data, bus electrical quantities and equipment status data of the microgrid within a preset decision cycle, and form periodic cross-sectional data according to the receiving cutoff time. The data synchronization unit is used to extract the clock deviation and measurement characteristics of the periodic cross-sectional data, determine the validity weight by combining the data delay time, and align the valid data to a unified time reference to form an instantaneous energy state. The deficit calculation unit is used to match the instantaneous energy state with the system command record table to evaluate the uncompleted amount, combine the equipment response capability with the effectiveness weight to determine the in-transit adjustment amount, and calculate the remaining net power deficit after deducting the in-transit adjustment amount. The power allocation unit is used to determine the adjustable power range based on the equipment operation constraints and the effectiveness weight, and to allocate the new adjustment amount to each device in the microgrid within the adjustable power range; The instruction issuing unit is used to issue a power instruction containing the newly added adjustment amount, update the instruction status based on communication feedback, and calculate the execution deviation to correct the subsequent target power.

2. The microgrid instantaneous energy distribution system based on a fused terminal according to claim 1, characterized in that, The unified time reference includes the end time of a fixed cycle set by the data acquisition unit according to the decision cycle, and the event triggering time triggered by microgrid electrical parameters exceeding limits or islanding switching; The data acquisition unit truncates and divides the cross-sectional data of the current period and the next period based on the receiving cutoff time.

3. The microgrid instantaneous energy distribution system based on a fused terminal according to claim 1, characterized in that, The data synchronization unit obtains the source end sampling end time of the corresponding acquisition device, adds the deviation of the source clock relative to the system fusion terminal clock, and obtains the converted sampling end time. The difference between this converted sampling end time and the alignment time corresponding to the unified time reference is the data delay time. ; Preset reference delay time and maximum allowable delay time ,and ; when At that time, validity weight ; when At that time, validity weight ; when When this happens, the corresponding data will be marked as invalid and ; For the average power data of the metering window, the validity weight is adjusted according to the ratio of the metering window length to the preset standard window length.

4. A microgrid instantaneous energy distribution system based on a fused terminal according to claim 1, characterized in that, The deficit calculation unit determines the theoretical adjustment limit based on the equipment ramp-up rate and the remaining execution time of the instructions in the system instruction record table. After reducing the theoretical adjustment limit by the effectiveness weight, it is compared with the uncompleted amount to determine the actual adjustment amount in transit. The instruction issuing unit dynamically tracks the lifecycle status of an instruction from pending delivery to completion or failure through the system instruction record table.

5. A microgrid instantaneous energy distribution system based on a fused terminal according to claim 1, characterized in that, When calculating the remaining net power deficit, the deficit calculation unit selectively subtracts or algebraically adds from the remaining net power deficit based on the consistency of the increase / decrease direction between the in-transit adjustment amount and the power deficit before deducting the in-transit adjustment amount. Expired instructions that have not been completed within the time limit will continue to occupy the adjustable power range of the corresponding device in the same adjustment direction, and it is prohibited to repeatedly issue instructions in the same direction with a target power difference within the preset tolerance to devices with unknown status before the instruction is clearly fed back.

6. A microgrid instantaneous energy distribution system based on a fused terminal according to claim 1, characterized in that, The remaining net power deficit The calculation formula is:

7. Among them, For load power, For network loss, For distributed power supply, For energy storage power, Reference power for grid connection point, These are in-transit adjustments in the same direction, during islanded network operation. .

8. A microgrid instantaneous energy distribution system based on a fused terminal according to claim 1, characterized in that, The power allocation unit allocates power to each device in the microgrid in a hierarchical manner, from fastest to slowest response speed. When the adjustable power range of the upper-layer device is exhausted, the inter-layer switch to the lower-layer device is triggered, and the new adjustment amount is determined with the constraints of minimizing resource call amount, minimizing the change amount of instructions in adjacent cycles, and minimizing the unallocated deficit.

9. A microgrid instantaneous energy distribution system based on a fused terminal according to claim 1, characterized in that, When power feedback fails or the mean of data validity weights falls below a preset confidence threshold, the system triggers a degradation strategy: Stop adding new global power allocations, switch control energy storage devices and adjustable distributed power sources to local safety control mode, and maintain the existing state of controllable loads; After communication is restored, the system will resume allocating the newly added adjustment amounts to the corresponding devices based on the latest instruction feedback status and mode flags.

Images