A demand side response transaction simulation method

Abstract

The present application relates to the technical field of power demand side transaction simulation, and discloses a demand side response transaction simulation method. The difference between the response end state of the demand side resource and the corresponding reference operation state is determined to obtain a to-be-restored state, and the resource operation constraint is combined to generate a recovery load sequence and a candidate recovery period; the response transaction section and the recovery transaction section are combined to form a transaction unit, the timing residual power supply capacity of the resource access node is preoccupied before writing the simulation result, and the resource operation constraint and the power supply network timing are checked; when the checking fails, the transaction unit is determined as invalid as a whole and the preoccupied capacity is released. The present application can avoid false transactions in which the response is feasible but the subsequent recovery load cannot be executed, and improve the reliability of the transaction simulation result.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power demand side transaction simulation, and particularly relates to a demand side response transaction simulation method. BACKGROUND

[0002] With the increasing participation of demand response, adjustable load and distributed resources in the electricity market, demand side resources can participate in electricity trading by reducing, shifting or postponing electricity consumption. To determine whether the demand side transaction declaration can be executed, the prior art usually forms a transaction simulation load according to the benchmark load, response period, response power and operation constraints of the demand side resources, and combines the operation limits such as the topology structure, node voltage and line capacity of the power supply network to simulate the clearing and safety check.

[0003] However, the operation impact of the demand side resources after executing the response instruction usually does not disappear immediately with the end of the response period. For example, when the resources reduce or suspend electricity consumption during the response process, the operation state such as temperature, energy storage level or production progress may gradually deviate from the benchmark operation state without participating in the response. In order to make the demand side resources continue to work according to the original operation plan, electricity consumption usually needs to be increased or rearranged after the response ends, thereby forming a recovery load. The size and occurrence period of the recovery load are related to the response end state, recovery ability and operation constraints of the resources, and thus cannot be determined only according to the reduced load in the response period.

[0004] For example, in the prior art, Chinese patent publication No. CN121618732A discloses a temperature-sensitive load orderly recovery method and system considering rebound effect, which calculates the aggregated heat gap according to the load thermal characteristic parameters, indoor temperature and outdoor temperature, obtains the future load prediction curve according to the total load, weather forecast data and the aggregated heat gap, and solves the recovery scheme under the consideration of load peak-valley difference, user comfort and power grid safety constraints to determine and correct the recovery load power. The scheme can identify the load recovery demand after demand response and orderly recover the temperature-sensitive load.

[0005] However, the above scheme mainly controls the load recovery process after the response ends, and treats the load response and load recovery as two scheduling stages. For demand side transaction simulation, if the transaction is determined to be established when the response period meets the requirements, and the recovery load is calculated or checked only thereafter, the transaction simulation accounts for the power supply capacity released in the response period, and the power supply capacity occupied by the recovery load in the future has not participated in the judgment of whether the transaction is established. When the node power supply capacity in the recovery period has been occupied by other loads or transactions, the situation that the response transaction has passed the simulation clearing but the recovery load generated by the response cannot be executed may occur.

[0006] Furthermore, simply overlaying restored loads onto subsequent load curves for power grid verification, while capable of detecting network overruns during the restoration period, lacks a transaction-level correlation between the restored load and the response transaction that triggered it. This makes it difficult to correct the validity of the corresponding response transaction based on the unexecutable result of the restored load. Therefore, existing demand-side transaction simulations may still separate the judgment of a continuous response and restoration process, resulting in transactions that are valid only during the response period but cannot be executed during the full operation being retained in the simulation clearing results. This affects the consistency between the transaction simulation results and the actual recovery capacity of demand-side resources and the timing carrying capacity of the power grid. Summary of the Invention

[0007] The purpose of this invention is to provide a demand-side response transaction simulation method to solve the problem that existing demand-side transaction simulations may separate the determination of a continuous response and recovery process, resulting in transactions that are only valid during the response period but cannot be executed during the complete operation being retained in the simulation clearing results, affecting the consistency between the transaction simulation results and the actual recovery capacity of demand-side resources and the timing carrying capacity of the power supply network.

[0008] To achieve the above objectives, the present invention provides a demand-side response transaction simulation method, comprising the following steps: Generate response instructions based on demand-side transaction declaration information; Simulate demand-side resources to execute the response instructions; The recovery status is determined based on the difference between the response completion status of demand-side resources and the corresponding baseline operating status; The recovery load sequence and candidate recovery periods are determined based on the recovery status and resource operation constraints. The response transaction segment corresponding to the response instruction and the recovery transaction segment corresponding to the recovery load sequence are combined into a transaction unit, and the transaction unit participates in simulated clearing. Before writing the transaction unit into the simulated clearing result, the remaining power supply capacity of the demand-side resource access node is pre-occupied according to the recovery load sequence during the candidate recovery period. The power supply network is simulated in a time sequence based on the transaction unit and the pre-allocated power supply capacity; If the restored load sequence has completed capacity pre-occupancy, the transaction unit meets the resource operation constraints, and the simulated power supply network operation status meets the power supply network operation limits, the transaction unit is written into the simulated clearing result; otherwise, the transaction unit is determined to be invalid. In response to the transaction unit being determined to be invalid, the pre-allocated power supply capacity is released.

[0009] As a preferred technical solution for demand-side response transaction simulation methods, the determination of the recovery load sequence and candidate recovery periods includes: Starting from the response end state, candidate recovery loads are applied hourly in each time period after the response ends, and the operating status of demand-side resources is updated according to the resource operation constraints. Candidate recovery loads and their time periods that reduce the difference between the updated operating state and the corresponding baseline operating state are retained until the difference is eliminated. The retained candidate recovery loads are then arranged into a recovery load sequence according to time sequence, and the corresponding time periods are used as the candidate recovery time periods.

[0010] As a preferred technical solution for the demand-side response transaction simulation method, the time-series remaining power supply capacity is the power supply capacity limit of the access node minus the baseline load of the corresponding time period and the capacity already occupied by other transaction units. The pre-occupancy includes: selecting time periods with remaining power supply capacity not less than the current restored load from the corresponding candidate restored time periods according to the time sequence of the restored load sequence, deducting the remaining power supply capacity of the selected time periods and recording the corresponding relationship, and determining that the restored load sequence has been fully pre-occupied only when all restored loads in the restored load sequence have been deducted.

[0011] As a preferred technical solution for demand-side response transaction simulation, in response to any recovery load failing to complete capacity pre-occupancy within the corresponding candidate recovery period, the time-series remaining power supply capacity deducted by the transaction unit is traced back and released, the winning status of the transaction unit in this round of simulated clearing is revoked, the time period corresponding to the response transaction segment and the time period corresponding to the released capacity are determined as the affected time period, and only other transaction units whose response transaction segment or recovery transaction segment overlaps with the affected time period are re-simulated and cleared.

[0012] As a preferred technical solution for the demand-side response transaction simulation method, the candidate recovery period consists of one or more consecutive or intermittent periods following the response transaction segment.

[0013] As a preferred technical solution for demand-side response transaction simulation methods, the recovery load in each time period shall not exceed the upper limit of the recoverable power of demand-side resources in the corresponding time period.

[0014] As a preferred technical solution for the demand-side response transaction simulation method, the response transaction segment and the recovery transaction segment carry the same transaction identifier.

[0015] As a preferred technical solution for demand-side response transaction simulation methods, a capacity change record is generated for each capacity pre-occupancy and release. The capacity change record includes at least the transaction identifier, access node, time period, and capacity.

[0016] As a preferred technical solution for demand-side response transaction simulation methods, the power supply network operation limits include at least one of node voltage limits, line capacity limits, and transformer capacity limits.

[0017] As a preferred technical solution for demand-side response transaction simulation methods, when multiple transaction units compete for the power supply capacity of the same access node in the same time period, the capacity is pre-allocated in sequence according to the simulated clearing order, and the time-series remaining power supply capacity is updated after each capacity pre-allocation.

[0018] Compared with existing technologies, the advantages of this invention are as follows: After demand-side resources execute response commands to reduce or transfer load, the power consumption decreases during the response period, but the resource operating state usually does not immediately return to the baseline operating state after the response ends. The remaining state differences will subsequently form recovery loads. Therefore, a demand-side response includes interrelated response and recovery processes in time; the former reduces or transfers load during the response period, while the latter reclaims power supply capacity for subsequent periods. If transaction simulations only verify the response process, they will include the power supply capacity released by the response but omit the recovery load, thus listing transactions that cannot be fully executed as feasible transactions. This invention uses the difference between the response end state and the corresponding baseline operating state as the starting point for recovery calculations, and combines resource operating constraints to determine the recovery load sequence and candidate recovery periods. State offsets such as temperature, energy, or operating progress are thus implemented as loads and time periods that can participate in simulation clearing and power network timing simulation. This recovery load is derived from the state differences caused by this response, and its size and executable time period are limited by the actual recovery capacity of demand-side resources, corresponding to the subsequent power demand accompanying this response.

[0019] Furthermore, the response transaction segment and the recovery transaction segment generated therefrom constitute a transaction unit, which participates in the simulated clearing as a single transaction. Before the transaction unit is written into the simulated clearing result, the system pre-allocates the remaining power supply capacity of the access node according to the candidate recovery period and the recovery load sequence. The pre-allocation only reserves the power supply space required for the future execution of the recovery load, and does not execute the recovery load in advance, thereby avoiding the duplicate occupation of this part of the capacity by other transactions. In this way, the simulated clearing takes into account both the load decrease during the response period and the load increase during the recovery phase, and makes a unified evaluation of the complete power consumption process of a demand-side transaction.

[0020] Furthermore, the valid status of a transaction unit depends on three verification results: the entire load restoration sequence has completed capacity pre-allocation, both the response transaction segment and the restoration transaction segment meet resource operation constraints, and the power supply network meets operation limits within the corresponding time period. Only when all three verifications pass is the transaction unit written into the simulated clearing result; if any one fails, both transaction segments are deemed invalid, and the pre-allocated power supply capacity is released. Thus, the executability of future load restoration directly constrains whether the current response transaction is valid, avoiding situations where the response transaction has already cleared, but the load restoration cannot be executed due to insufficient node capacity or network over-limits.

[0021] In summary, ensuring that response transactions and their restoration obligations remain consistently valid or invalid can reduce redundant allocation of power supply capacity and fictitious feasible transactions, making the simulated clearing results closer to the actual operating characteristics of demand-side resources and the time-series carrying capacity of the power supply network. Attached Figure Description

[0022] The accompanying drawings are provided to further understand the technical solutions of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the technical solutions of the present invention, and do not constitute a limitation on the technical solutions of the present invention.

[0023] Figure 1 The overall flowchart of the demand-side response transaction simulation method provided in the embodiments of the present invention is shown. Detailed Implementation

[0024] The present invention will be further described below with reference to embodiments. These embodiments are used to explain the technical solutions of the present invention and are not intended to limit the scope of protection of the present invention.

[0025] When simulating declared transactions on the demand-side trading platform, the declared response volume is typically written into the response period to form a simulated transaction load. Based on this load, resource constraint verification and power supply network flow calculations are performed. Cold storage refrigeration, building temperature control, user-side energy storage, and transferable production loads all have cross-time period operating states, and response commands will change their state trajectories. The load reduced during the response period will be transferred to subsequent periods in the form of temperature correction, energy storage replenishment, or production task replenishment. The additional electricity consumption formed subsequently shares the power supply capacity with the node baseline load and other traded loads.

[0026] Taking cold storage compressors and user-side energy storage as examples, after the cold storage compressor reduces its power during the response period, the temperature inside the cold storage rises along a new trajectory; after the user-side energy storage takes over the response by discharging, its state of charge decreases along a new trajectory. At the end of the response, although both types of resources may still be within the equipment's operating boundaries, their state has not yet returned to the original operating plan. If the transaction simulation stops with the end of the response, the subsequent recovery process will not be included in the output results, and the power supply capacity occupancy during the recovery period cannot be used as a clearing condition.

[0027] In this embodiment, the transaction simulation server is configured with a resource digital twin engine, a transaction unit processor, a node time-series capacity ledger, and a time-series network simulation interface. The resource digital twin engine deduces the resource status from the response command and continues to generate recovery load based on the response's end status. The transaction unit processor associates response transaction segments and recovery transaction segments generated by the same response. The node time-series capacity ledger records available capacity and its changes according to the access node and time period. The time-series network simulation interface calls the existing power flow calculation program of the trading platform. Price ranking, candidate transaction volume calculation, and transaction priority are still provided by the original simulated clearing program.

[0028] Based on the above, please refer to Figure 1 As shown, this embodiment provides a demand-side response transaction simulation method. Before execution, resource identifiers, access nodes, baseline operating data, current operating status, and resource operating constraints are read from the demand-side resource management system; demand-side transaction declaration information is read from the transaction declaration system; and network topology, node power supply capacity, and operating limits are read from the power supply network model. All data are time-aligned according to the simulation step size adopted by the transaction platform, and the mapping relationship between resource identifiers and access nodes is loaded along with the resource model.

[0029] Regarding the first Each demand-side resource is configured with a state vector in the digital twin engine. Power input External disturbance vector and state transition model The state vector of temperature-controlled resources can include temperature and cold storage capacity; the state vector of energy storage resources can include state of charge; and the state vector of production load can include the amount of tasks to be completed and equipment start / stop status. Resource state updates between adjacent time periods use: (1), In the formula, Number the simulated time period; and Time periods The resource status at the start and end; Input the power consumption of the resource model for this period; This includes external quantities such as ambient temperature, production plans, or equipment availability. Existing mechanistic models, gray-box models, or state-space models calibrated using historical operational data from the resource management system can be used. Model training and parameter calibration utilize existing results from the resource side, and the transaction simulation server reads its model version and parameter snapshots to ensure that the baseline trajectory, response trajectory, and recovery trajectory use the same model caliber.

[0030] The transaction processing in this embodiment begins with a response instruction.

[0031] S101, generates response instructions based on demand-side transaction declaration information; Demand-side transaction declaration information must include at least the declared resources, response period, and declared response quantity, and may also include the declared price. The transaction unit processor, according to the resource control interface, converts the declared response quantity into target power or power adjustment amount, and writes the resource identifier, response start and end times, and the simulation number for this round. For example, a declaration for reduced-capacity cold storage can be converted into a compressor power limit, an energy storage declaration into discharge power, and a production load declaration into a shutdown period or reduced-load power.

[0032] The digital twin engine copies a snapshot of the resource state at the start of the response after the response command is generated, and uses this snapshot as the starting point for the response simulation.

[0033] S102, Simulate demand-side resources to execute the response command; During the response period, the power given by the response command is written into equation (1), and the response state trajectory is obtained hourly. At the same time, based on the baseline power in the original resource operation plan. Input the same state transition model to obtain the baseline state trajectory. The two trajectories use the same initial state, external disturbance, and time-scale granularity, with the power input being the response power and the reference power, respectively. The simulation is then performed until the end of the response time. Afterwards, the digital twin engine outputs the response completion status. and the baseline operating status at the same time. .

[0034] S103, determine the state to be restored based on the difference between the response end state of demand-side resources and the corresponding baseline operating state; For those with Resources with each status field are processed using normalized status deviation. Describe the two trajectories in the time period Distance: (2), In the formula, and These are the current projected trajectory and the baseline trajectory of demand-side resources, respectively. One status field; This is the normalization coefficient for the state field. State fields with clearly defined upper and lower limits of operation can take... ,in, This represents the upper limit of the m-th status field of demand-side resource i. This represents the lower bound of the m-th state field of demand-side resource i; ensuring that states with different dimensions, such as temperature, state of charge, and workload, can be classified under the same deviation. State fields lacking fixed upper and lower bounds can be normalized using per-unit baseline values ​​from the resource model. Resources with a single state field directly use the absolute difference of that field.

[0035] At the end of the response, the difference vector will be... , deviation amount The resource identifier and the time of response completion are written to the pending recovery status. Δ Provide the recovery direction for each status field. Used to determine whether candidate recovery loads are pushing resources closer to the baseline trajectory. When, the transaction segment is restored and written to an empty sequence; when When the value is greater than zero, the digital twin engine continues to extrapolate from the runtime phase after the response ends.

[0036] S104, determine the recovery load sequence and candidate recovery time period based on the recovery status and resource operation constraints; The determination of the recovery load sequence and candidate recovery time period includes: taking the response end state as the starting state, applying candidate recovery loads hourly according to the resource operation constraints in each time period after the response ends and updating the operation state of the demand-side resources; retaining the candidate recovery loads and their time periods that reduce the difference between the updated operation state and the corresponding baseline operation state until the difference is eliminated, and assembling the retained candidate recovery loads into the recovery load sequence according to time sequence, and using the corresponding time period as the candidate recovery time period.

[0037] Apply candidate recovery load during time period t At that time, the power input of the digital twin model is taken Start by trying to determine the maximum recoverable power of the resource during that time period. Then use equation (1) to calculate the updated state and use equation (2) to calculate The candidate recovery load satisfies... And when the updated power, status, ramp rate, and start / stop status meet the resource operation constraints, The time period and its duration are written into the candidate record. For resources such as temperature-controlled loads and energy storage resources, whose states monotonically change with the recovery power, it is possible to... A binary search is used within the interval, with the search termination accuracy set to the minimum power resolution allowed by the equipment control interface. Candidate loads in the final time period converge to the power value that brings the state to the baseline trajectory without exceeding limits. For production loads with start-stop combinations or process stage switching, the discrete executable instructions given by the resource control interface can be traversed, retaining... The minimum load corresponding to instructions that satisfy operational constraints.

[0038] Candidate records use a data structure of <Sequence Number, Recovery Power, Set of Executable Time Periods, Pre-Execution State, Post-Execution State>. The post-execution state of the previous candidate recovery load serves as the starting point for the derivation of the next, until... Reduced to zero. The recovery power arranged by sequence number constitutes the recovery load sequence, and the set of executable time periods is merged to form candidate recovery time periods.

[0039] The candidate recovery period consists of one or more consecutive or intermittent periods following the response transaction segment.

[0040] The digital twin engine filters candidate recovery periods based on resource availability plans. Continuous recovery periods correspond to periods where the cold storage compressor remains available after response; intermittent recovery periods correspond to periods where production equipment is restricted by shift changes, process waiting times, and maintenance plans; and restricted recovery periods are excluded for energy storage resources based on charging / discharging plans, grid-connected power, and state-of-charge boundaries. If multiple executable recovery periods exist for the same recovery load, all are written into this candidate record, while preserving the order of recovery load sequences.

[0041] The recovery load for each time period shall not exceed the upper limit of the recoverable power of the demand-side resources for the corresponding time period.

[0042] Maximum recoverable power It is determined by the equipment's rated input power, reference power, current-state permissible input power, and ramp limits. The rated input power is used as the basis for this determination. and reference power For example, the upper limit of additional power given by the device power boundary can be written as: When the resource status model or ramp-up constraint gives a smaller value, the minimum value among all constraints is taken. This is after all candidate recovery periods have been simulated. When a recovery is incomplete flag is written in the recovery transaction segment, it will be used as the basis for determining whether a transaction unit does not meet resource operation constraints.

[0043] S105, the response transaction segment corresponding to the response instruction and the recovery transaction segment corresponding to the recovery load sequence are combined into a transaction unit, and the transaction unit participates in simulated clearing; Trading Unit The following record structure can be used: .in, For transaction identification; For access nodes to demand-side resources; Save the response time period, response power, and response status trajectory; Save the recovery load sequence, candidate recovery periods, and recovery status trajectory; A bitmap showing the time period involved in the transaction; The status field indicates whether the transaction is selected, valid, or invalid. The status field of a transaction unit uniformly controls both transaction segments, and status updates are performed in a single write operation to prevent the separate submission of one transaction segment.

[0044] The response transaction segment and the recovery transaction segment carry the same transaction identifier.

[0045] A transaction identifier is generated upon receiving a transaction application and is used throughout the response simulation, recovery sequence generation, capacity pre-allocation, and rollback release. Deduction records and the mapping records between recovery load and time periods in the capacity ledger are all written to this transaction identifier, enabling transaction units to be submitted or withdrawn as a group based on the identifier. The original simulated clearing procedure could first clear the transaction based on the quoted price and the application quantity. Set to the selected status, and update to the valid status after subsequent capacity and operation verifications are passed.

[0046] S106, Before writing the transaction unit into the simulated clearing result, the remaining power supply capacity of the demand-side resource access node is pre-occupied according to the recovery load sequence during the candidate recovery period. The node time-series capacity ledger uses a two-dimensional capacity matrix. The rows of the matrix correspond to the access nodes. The matrix columns correspond to the simulation time periods. All power types are standardized to the power type and unit used in the power supply network model before being written into the matrix. Matrix elements are initialized and updated using the following formula: (3), In the formula, For nodes During the period Power supply capacity limits; This is the baseline load for that node; For the set of prior transaction units; For the first trading unit The capacity already occupied. When the matrix uses sparse time-series storage, it only needs to save the node time period entries where the baseline value changes, there are transaction occupancy, or the running limit changes.

[0047] The time-series remaining power supply capacity is the power supply capacity limit of the access node minus the base load and the capacity already occupied by other transaction units in the corresponding time period; the pre-occupancy includes: according to the time sequence of the recovery load sequence, sequentially selecting time periods from the corresponding candidate recovery time periods with a time-series remaining power supply capacity not less than the current recovery load, deducting the time-series remaining power supply capacity of the selected time periods and recording the corresponding relationship, and determining that the recovery load sequence has been fully pre-occupied only when all recovery loads in the recovery load sequence have been deducted.

[0048] When performing capacity pre-allocation, starting from the first item in the recovery load sequence, the executable time periods of that item are read sequentially by time. .when Select the time period and execute. Simultaneously, the selected time period is used as the lower bound for the next load recovery time. This process continues until the end of the sequence. If no available time period is found for any item, the pre-occupancy fails, and the already executed matrix deduction is added to the backtracking queue.

[0049] When multiple trading units compete for the power supply capacity of the same access node in the same time period, the capacity is pre-allocated in sequence according to the simulated clearing order, and the remaining power supply capacity for the corresponding time period is updated after each capacity pre-allocation.

[0050] The simulated clearing order serves as the write order for the capacity matrix. After the previous transaction unit completes the deduction, the next transaction unit reads the updated value. The order remains consistent with the original clearing rules, such as pricing and declaration time. The capacity pre-occupancy procedure adds node time slot occupancy status to the existing order.

[0051] A capacity change record is generated for each capacity pre-allocation and release. The capacity change record includes at least the transaction identifier, access node, time period, and capacity.

[0052] Capacity change records can also include the operation type, capacity before the change, capacity after the change, and record sequence number, forming an append-only event log indexed by transaction identifier. Negative capacity increments for operation records are pre-allocated, while positive capacity increments are released. In the event of a rollback, records are restored in reverse order of their sequence numbers. It can handle capacity deduction for the same transaction across multiple intermittent periods.

[0053] After the load restoration and capacity mapping are completed, the digital twin engine re-runs the restoration trajectory according to the selected actual time period to verify the continuity of the state after the time period migration. Transaction units that pass the verification are then entered into the power supply network timing simulation.

[0054] S107, Perform time-series operation simulation of the power supply network based on the transaction unit and the pre-allocated power supply capacity; First, the load changes caused by the response transaction segment and the recovery load mapped to the actual time period are written into the node load sequence. Then, this sequence is superimposed with the node baseline load and the loads of other transaction units in the selected state to form the transaction simulation load. The timing network simulation interface calls the power flow calculation program hourly and outputs the node voltages. Line load rate and transformer load rate The topology, line parameters, and transformer parameters used in the simulation are read from the power supply network model.

[0055] The power supply network operation limits include at least one of node voltage limits, line capacity limits, and transformer capacity limits.

[0056] For each time period involving response or recovery, compare separately. With node voltage upper and lower limits, With line capacity limits and With transformer capacity limits. The node time-series capacity ledger completes preliminary screening at the node power level, and the time-series power flow results supplement voltage distribution, line power flow, and transformer load verification. The resource digital twin engine synchronously outputs the restored state, confirming... It has dropped to zero.

[0057] S108, if the restored load sequence has completed capacity pre-occupancy, the transaction unit meets the resource operation constraints, and the simulated power supply network operation status meets the power supply network operation limits, the transaction unit is written into the simulated clearing result; otherwise, the transaction unit is determined to be invalid. When a trading unit writes the simulated clearing result, it also saves the response trading segment, the recovery trading segment, the capacity mapping relationship, the digital twin model version, and a summary of the time-series network simulation results. This data can reproduce the state trajectory of a transaction from declaration and response to completion of recovery. If any verification condition fails, the trading unit state... If the update is deemed invalid, the response and recovery transaction segments will cease to be committed along with this status.

[0058] S109, in response to the transaction unit being determined to be invalid, the pre-allocated power supply capacity is released.

[0059] Capacity ledger based on Read the capacity change record of this transaction unit and execute. The mapping between the restored load and the selected time period is cancelled. If a trading unit has already expired before the capacity deduction, its selected status is directly revoked.

[0060] If any recovery load fails to complete capacity pre-occupancy within the corresponding candidate recovery period, the time-series remaining power supply capacity deducted for that transaction unit is backtracked and released, the winning status of that transaction unit in this round of simulated clearing is revoked, the time period corresponding to the response transaction segment and the time period corresponding to the released capacity are determined as the affected time period, and only other transaction units whose response transaction segment or recovery transaction segment overlaps with the affected time period are re-simulated clearing.

[0061] To improve the efficiency of local recalculation, each time period within the simulation period can be mapped to one bit to form a trading unit time period bitmap. Bitmap of affected time periods .satisfy Other transaction units enter the re-clearing queue, while the remaining transaction units retain their original capacity mapping and valid status. The corresponding time period for the response transaction segment is added. Afterwards, the node load restored by the undo response can be updated; the released capacity corresponding to the time period can be added. The released recovery capacity can then be reallocated. This bitmap filtering can be performed via processor bitwise operations and is suitable for rolling simulation scenarios with a large number of transaction units.

[0062] The following uses a set of exemplary operating parameters to illustrate the data flow of the cold storage refrigeration load. The simulation step size is set to 15 minutes, the base power of the cold storage compressor is 120kW, and the transaction declaration requires the power to be reduced to 70kW for two consecutive periods. The digital twin model reads the temperature inside the cold storage, the outdoor temperature, and the compressor operating status at the start of the response; the base temperature at the end of the response is -18.0℃, and the response temperature is -17.4℃. If the temperature operating boundary configured by the resource model is [-22℃, -16℃], then the normalization coefficient in equation (2) is taken as 1 / 6, corresponding to the normalized state deviation. .

[0063] In the fifth simulation period after the response ends, the digital twin engine begins searching for recovery loads. Based on the compressor's rated power, baseline power, and temperature boundaries, additional recovery power [40kW, 35kW, 25kW] is obtained in a time-series order. The total additional electricity consumption for the three recovery periods is 25kWh, which is on the same order of magnitude as the electricity reduction during the response phase; the specific value is still determined by the temperature state trajectory. Item 1 can be executed in the fifth simulation period, item 2 in the sixth or seventh simulation period, and item 3 in the eighth simulation period. The capacity limits and capacity ledger data for the cold storage access nodes are shown in the table below.

[0064]

[0065] Based on the capacity data in the table, the restored load sequence is mapped sequentially to simulation periods 5, 7, and 8. The digital twin engine re-imagines the cold storage temperature according to this mapping, and the restored state returns to the corresponding baseline temperature trajectory. Subsequently, the node voltage, line load rate, and transformer load rate for each restoration period are obtained through time-series power flow calculations. In this example, the lowest node voltage is 0.97 pu, the highest line load rate is 86%, and the highest transformer load rate is 82%; the voltage range configured in the model is 0.95 pu to 1.05 pu. All results meet the power supply network operating limits, and the response transaction segment and the restoration transaction segment are written into the simulation clearing results.

[0066] If the remaining power supply capacity in the 7th simulation period drops to 30kW, the second 35kW restoration load cannot be pre-occupied in any of the candidate periods. The capacity ledger is then filled with the 40kW deducted in the 5th simulation period based on the transaction identifier. Subsequently, the cold storage response transaction is cancelled, and the corresponding period of the response transaction segment and the 5th simulation period are written into the affected period bitmap. Only other transaction units that intersect with this bitmap will re-execute the simulation clearing.

[0067] User-side energy storage can set the state of charge (SOC) and charge / discharge state as state vectors. After the energy storage device completes its discharge response, the digital twin engine compares the actual SOC trajectory with the baseline SOC trajectory, and then generates a recovery charging sequence based on charging efficiency, grid-connected power limit, and the original charge / discharge plan. High-load periods can be excluded from the candidate recovery periods, and the remaining charging periods are reserved for capacity. After the complete charging sequence passes capacity and network verification, the discharge response obtains a valid state; if any charging load cannot be scheduled, the response transaction and charging recovery record are cancelled using the same transaction identifier.

[0068] When the response end state is consistent with the corresponding baseline operating state, the recovery transaction segment saves an empty sequence, the recovery capacity is not deducted during the capacity pre-occupancy stage, and the response transaction segment is still subject to resource operation constraints and power supply network operation limit verification. When the recovery transaction segment includes recovery load, the transaction unit can only be written into the simulation clearing result after the complete sequence has completed capacity pre-occupancy, resource state regression, and power supply network verification.

[0069] The above embodiments are used to illustrate the technical solutions of the present invention. For those skilled in the art, any equivalent modifications or substitutions made to the relevant technical features without departing from the principles of the present invention should fall within the protection scope of the present invention.

Claims

1. A demand-side response transaction simulation method, characterized in that, include: Generate response instructions based on demand-side transaction declaration information; Simulate demand-side resources to execute the response instructions; The state to be restored is determined based on the difference between the response end state of the demand-side resources and the corresponding baseline operating state; the recovery load sequence and candidate recovery time period are determined based on the recovery state and resource operating constraints; the response transaction segment corresponding to the response instruction and the recovery transaction segment corresponding to the recovery load sequence are combined into a transaction unit, and the transaction unit participates in the simulated clearing; before the transaction unit is written into the simulated clearing result, the remaining power supply capacity of the demand-side resource access node is pre-occupied according to the candidate recovery time period and the recovery load sequence; The power supply network is simulated in a time sequence based on the transaction unit and the pre-allocated power supply capacity. If the recovery load sequence has completed the capacity pre-allocation, the transaction unit meets the resource operation constraints, and the simulated power supply network operation status meets the power supply network operation limits, the transaction unit is written into the simulation clearing result; otherwise, the transaction unit is determined to be invalid. In response to the transaction unit being determined to be invalid, the pre-allocated power supply capacity is released.

2. The demand-side response transaction simulation method according to claim 1, characterized in that, The determination of the recovery load sequence and candidate recovery time period includes: taking the response end state as the starting state, applying candidate recovery loads hourly according to the resource operation constraints in each time period after the response ends and updating the operation state of the demand-side resources; retaining the candidate recovery loads and their time periods that reduce the difference between the updated operation state and the corresponding baseline operation state until the difference is eliminated, and assembling the retained candidate recovery loads into the recovery load sequence according to the time sequence, and taking the corresponding time period as the candidate recovery time period.

3. The demand-side response transaction simulation method according to claim 2, characterized in that, The remaining power supply capacity in the time series is the power supply capacity limit of the access node minus the base load and the capacity already occupied by other transaction units in the corresponding time period; The pre-occupancy includes: selecting time periods with remaining power supply capacity not less than the current restored load from the corresponding candidate restored time periods according to the time sequence of the restored load sequence, deducting the remaining power supply capacity of the selected time periods and recording the corresponding relationship, and determining that the restored load sequence has been fully pre-occupied only when all restored loads in the restored load sequence have been deducted.

4. The demand-side response transaction simulation method according to claim 3, characterized in that, If any recovery load fails to complete capacity pre-occupancy within the corresponding candidate recovery period, the time-series remaining power supply capacity deducted for that transaction unit is backtracked and released, the winning status of that transaction unit in this round of simulated clearing is revoked, the time period corresponding to the response transaction segment and the time period corresponding to the released capacity are determined as the affected time period, and only other transaction units whose response transaction segment or recovery transaction segment overlaps with the affected time period are re-simulated clearing.

5. The demand-side response transaction simulation method according to claim 2, characterized in that, The candidate recovery period consists of one or more consecutive or intermittent periods following the response transaction segment.

6. The demand-side response transaction simulation method according to claim 5, characterized in that, The recovery load for each time period shall not exceed the upper limit of the recoverable power of the demand-side resources for the corresponding time period.

7. The demand-side response transaction simulation method according to claim 1, characterized in that, The response transaction segment and the recovery transaction segment carry the same transaction identifier.

8. The demand-side response transaction simulation method according to claim 7, characterized in that, A capacity change record is generated for each capacity pre-allocation and release. The capacity change record includes at least the transaction identifier, access node, time period, and capacity.

9. The demand-side response transaction simulation method according to claim 1, characterized in that, The power supply network operation limits include at least one of node voltage limits, line capacity limits, and transformer capacity limits.

10. The demand-side response transaction simulation method according to claim 3, characterized in that, When multiple trading units compete for the power supply capacity of the same access node in the same time period, the capacity is pre-allocated in sequence according to the simulated clearing order, and the remaining power supply capacity for the corresponding time period is updated after each capacity pre-allocation.

Images