Simulation engine processing system and method based on power distribution digitization platform

Abstract

The application discloses a simulation engine processing system and method based on a power distribution digital platform, and relates to the technical field of power system simulation.The application receives real-time data and identifies power grid events through an event monitoring and analysis module; a model assembly module calls simulation model units and configures parameters to generate a core area simulation instance; a boundary equivalent module performs equivalent simplification processing on adjacent boundary areas to generate an equivalent model; a task execution module assembles the two into a complete calculation model and executes simulation; and a result processing module generates an optimized suggestion scheme and a verification report for the result.The application realizes a closed loop from automatic identification of power grid events to intelligent optimization decision-making, and greatly improves the automation, intelligent level and forward-looking risk early warning capability of power distribution network simulation analysis.

Description

Technical Field

[0001] This invention relates to the field of power system simulation technology, and more specifically, to a simulation engine processing system and method based on a digital power distribution platform. Background Technology

[0002] The distribution network is a crucial hub connecting the main grid and a vast number of power users, and its operational level directly determines the quality of power supply and service level. Under the background of new power system construction, a large number of diverse loads, such as distributed power sources, energy storage, and electric vehicle charging facilities, are being connected to the distribution network, resulting in highly volatile and complex operational characteristics. Traditional distribution network simulation analysis models are mostly built offline, failing to accurately describe the dynamic interactions among multiple elements such as "source-grid-load-storage." Existing simulation processes heavily rely on human experience, involving numerous and time-consuming steps from problem identification, model building, calculation execution to result analysis, making it difficult for frontline work teams to master and utilize them. This hinders the rapid analysis of emerging problems in the distribution network, becoming a major bottleneck in improving the intelligent operation and management level of the distribution network. Summary of the Invention

[0003] The purpose of this invention is to solve the problems of rigid processes, cumbersome manual operations, and slow response speed in traditional power distribution network simulation analysis.

[0004] To achieve the above objectives, the present invention provides a simulation engine processing system based on a digital power distribution platform, the system comprising:

[0005] The event monitoring and parsing module is used to receive real-time monitoring data from the power distribution digitalization platform, identify event information of the power distribution system based on a predefined event triggering rule library, and parse the event information into a scene description structure. The scene description structure includes event type, simulation target, core area range and simulation time span.

[0006] The model assembly module is used for:

[0007] Simulation model units are invoked based on event type and simulation objective;

[0008] Configure the parameters of the simulation model unit to generate a core area simulation instance for this event type;

[0009] Boundary isopleth modules are used for:

[0010] Determine the boundary region adjacent to the core region;

[0011] Based on event type and real-time monitoring data, the boundary region is simplified by equivalent value processing to generate an equivalent model;

[0012] The task execution module is used for:

[0013] The core area simulation instance is assembled with the equivalent model to obtain the power distribution network simulation calculation model;

[0014] The current event is simulated and calculated based on the aforementioned power distribution network simulation calculation model to obtain simulation results;

[0015] The results processing module is used for:

[0016] The credibility of the simulation results is evaluated, and key issue indicators corresponding to the simulation objectives are extracted from the simulation results.

[0017] Based on the key issue indicators, optimization suggestions are generated, and the optimization suggestions are verified by simulation and compared in terms of effect, and an evaluation report is generated.

[0018] This invention utilizes an event monitoring and parsing module to match and parse real-time monitoring data streams from a digital distribution platform with a predefined rule base, outputting a structured scene description. This transforms ambiguous on-site events into simulation task requirements that machines can accurately understand. The model assembly module and boundary equivalence module work in parallel based on this scene description structure. The model assembly module calls basic simulation model units from a pre-built modular library according to the event type and simulation objective. The boundary equivalence module determines the outer boundary of the core analysis area and selects an appropriate equivalence simplification method based on event characteristics and real-time data to generate an equivalent model representing the external network. Subsequently, the core area simulation instance and the equivalent model are assembled into a closed-boundary distribution network simulation calculation model by the task execution module, and calculations are performed. Finally, the result processing module analyzes the simulation results and generates optimization suggestions. This coherent process encodes expert analytical experience into automatically executable module functions and data flow rules, achieving a fundamental shift from offline operations to online closed-loop operations, greatly improving the agility and automation level of simulation analysis in handling real-time operating scenarios.

[0019] Furthermore, the simulation model unit includes: a source model unit, a network model unit, a load model unit, and a storage model unit;

[0020] Parameter configuration for the simulation model unit specifically includes:

[0021] Based on real-time monitoring data related to the core area, the initial parameters of the simulation model unit are assigned values;

[0022] Based on the simulation time span, historical running data, and predicted data, the time series parameters of the simulation model unit are generated.

[0023] Traditional simulation models often use fixed, typical parameters, leading to time-shifted discrepancies between simulation results and the actual operating state of the power grid, and failing to fully utilize known time-series patterns in subsequent evolution. This proposed solution assigns initial parameters based on real-time monitoring data and uses real measurements provided by the distribution digitization platform to set the initial operating state for each simulation model unit. This ensures the entire simulation instance is fully synchronized with the physical power grid at the initial moment. Then, based on the simulation time span specified for this task, the system matches operating curves from the historical database for the same period, extracts their periodic variation patterns, and incorporates predicted future environmental and market expectations (such as weather forecasts and load forecasts). Using these two types of time-series data, the system calculates and generates a dynamic parameter sequence covering the entire simulation period for each model unit. This enables the model to simulate the continuous behavior of the power grid within a specific time period, thereby improving the simulation credibility of dynamic processes and uncertainties under high-proportion renewable energy integration.

[0024] Furthermore, the types of simulation calculations include: power flow calculation, time-series power flow calculation, reactive power optimization calculation, voltage regulation calculation, harmonic analysis calculation, and short-circuit analysis calculation.

[0025] Among them, power flow calculation is used to analyze the current or a certain moment of operation; time-series power flow calculation is used to analyze dynamic processes; reactive power optimization and voltage regulation calculation focuses on the active control and improvement of power grid operation quality; harmonic analysis calculation is for power quality issues; and short-circuit analysis calculation is specifically used for electrical stress and protection verification under fault conditions.

[0026] Furthermore, the proposed optimization scheme is simulated and its effects compared, and an evaluation report is generated, including:

[0027] Based on the proposed optimization scheme, the parameters of the simulation instance in the core area are adjusted to obtain a comparison scheme model;

[0028] The simulation calculation was re-executed using the aforementioned comparison scheme model to obtain the comparison simulation results;

[0029] The initial simulation results are compared and analyzed with the comparative simulation results to generate an evaluation report.

[0030] In this process, after the system generates preliminary optimization suggestions based on key issue indicators, it doesn't output them directly. Instead, it transforms these suggestions into modification instructions for the simulation model, adjusting the corresponding parameters of the original core area simulation instance to generate a comparative model. Subsequently, the system uses this comparative model to re-execute the simulation calculation under identical boundary conditions and initial states, obtaining comparative simulation results. Finally, by comparing and analyzing the simulation results of the original problem with these verification results, the actual effect of the optimization scheme is quantitatively calculated, such as the percentage point increase in voltage qualification rate, and an evaluation report is generated. This process enhances the reliability of the analysis conclusions and the scientific rigor of the decision-making.

[0031] Furthermore, the credibility of the simulation results is evaluated, including:

[0032] Based on the preset evaluation rules, the integrity of the input data of the simulation results and the logical consistency of the electrical quantities are checked to obtain the check results;

[0033] The key operational data in the simulation results are compared with historical operational data from the same period to calculate the deviation and obtain the deviation calculation results.

[0034] Based on the inspection results and the deviation calculation results, a reliability score for the simulation results is generated.

[0035] In this invention, a credibility assessment sub-process is implemented in the result processing module. This assessment does not focus on the correctness of the simulation results, but rather on the rationality of the process and data that generated the results. First, based on preset assessment rules, it checks whether the input data on which the simulation results depend is complete, and whether the electrical quantities within the simulation results conform to basic physical logic consistency, such as whether the power is balanced and whether the voltage and power distribution is reasonable. This step verifies the internal logic of the simulation model and calculations. Second, it compares the key operating data calculated in the simulation results with actual monitoring data from the same period and / or under the same operating conditions extracted from historical databases, calculating the deviation. This step verifies from the perspective of external historical experience patterns. Finally, the system integrates the results of these two checks and comparisons to generate a quantitative credibility score, enabling users to judge the reliability of the results.

[0036] Furthermore, the event triggering rule base includes real-time monitoring event triggering rules and custom event triggering rules; the custom events include business expansion load access events, new energy access events, and electric vehicle charging events.

[0037] Furthermore, the task execution module also includes a simulation control unit;

[0038] The simulation control unit is used for:

[0039] Real-time monitoring of intermediate simulation results, and extraction of event evolution feature sequences based on intermediate results;

[0040] When the evolution feature sequence meets the preset triggering conditions, a simulation task is generated and executed based on the distribution network simulation calculation model to obtain future simulation results; wherein, the simulation task is used to superimpose short-term disturbances on the current model state to simulate future scenarios;

[0041] A control strategy is generated based on the future projection results, and the execution strategy of the current simulation task is dynamically adjusted based on the control strategy.

[0042] Within the task execution module, a simulation control unit is set up to monitor and control the ongoing simulation calculation process online. It monitors the intermediate result stream of the simulation calculation in real time, extracting feature sequences that reflect the dynamic evolution trend of events, such as the rate of voltage decline at a node or the acceleration of load growth on a line. When these feature sequences meet preset trigger conditions, indicating that the event may worsen, the control unit initiates a deduction task. Based on the current complete simulation calculation model, a short-term disturbance is superimposed to quickly simulate possible scenarios over a short period of time, thus obtaining a deduction result for the future situation. Based on this deduction result, the control unit generates corresponding control strategies and dynamically adjusts the execution strategy of the currently executing main simulation task. This constructs an online control closed loop of "monitoring-prediction-decision-adjustment," significantly improving the system's analytical depth for complex events and its intelligent scheduling capability of computing resources.

[0043] Furthermore, based on the aforementioned distribution network simulation calculation model, a deduction task is generated and executed, including:

[0044] Based on the aforementioned distribution network simulation model, the electrical sensitivity characteristics of the distribution network are calculated; these electrical sensitivity characteristics are used to quantify the impact of changes in node injected power or branch disconnection on critical electrical quantities.

[0045] Analyze the correlation between the electrical sensitivity characteristics and the evolution characteristic sequence, and determine the disturbance parameters of the short-term disturbance based on the correlation;

[0046] The disturbance parameters are dynamically constructed in the distribution network simulation calculation model, and the simulation task is instantiated and executed.

[0047] This invention, based on current distribution network simulation models, calculates the electrical sensitivity characteristics of the power grid online. These characteristics describe the impact of minute changes in injected power at any node or the opening / closing of any branch on critical electrical quantities across the entire network, characterizing the distribution of current weak points and the transmission paths of their effects. Simultaneously, the system performs correlation analysis between the electrical sensitivity characteristics and evolutionary feature sequences extracted from real-time simulations. For example, if the feature sequence shows a continuous voltage drop in region A, and sensitivity analysis indicates that the injected power at node B is most sensitive to the voltage in region A, the system will decide to apply a disturbance to node B. The type and magnitude of the disturbance are also set based on this correlation. Finally, based on this decision, disturbance parameters with clear physical meaning and exploratory purpose are dynamically constructed and instantiated into a simulation task. This method ensures that each simulation is not random but tailored to the specific operating state and evolution trend of the current power grid, greatly improving the efficiency of simulation in identifying potential critical risks.

[0048] Furthermore, the control strategy includes:

[0049] Control strategies to enhance the depth of analysis, update boundary model data, and optimize the allocation of computing resources.

[0050] Among these strategies, the control strategy for enhancing analytical depth is generated when the simulation predicts a potential severe downstream cascading failure. This requires more refined modeling and simulation of the potential failure area to support in-depth mechanistic analysis. The control strategy for updating boundary model data is generated when the simulation indicates that the boundary equivalence assumptions made in the initial simulation are no longer accurate due to the development of the main event, and are about to become distorted. This is achieved by updating the boundary conditions of the simulation model. The control strategy for optimizing computational resource allocation is generated when the simulation results show that the event's evolution trend is clear and will rapidly stabilize. This strategy, while ensuring the reliability of the conclusions, terminates lengthy calculations early to release resources.

[0051] This invention also provides a simulation engine processing method based on a power distribution digitalization platform, the method comprising:

[0052] Receive real-time monitoring data from the power distribution digitalization platform, identify event information of the power distribution system based on a predefined event triggering rule base, and parse the event information into a scene description structure, which includes event type, simulation target, core area range and simulation time span;

[0053] Simulation model units are invoked based on event type and simulation objective;

[0054] Configure the parameters of the simulation model unit to generate a core area simulation instance for this event type;

[0055] Determine the boundary region adjacent to the core region;

[0056] Based on event type and real-time monitoring data, the boundary region is simplified by equivalent value processing to generate an equivalent model;

[0057] The core area simulation instance is assembled with the equivalent model to obtain the power distribution network simulation calculation model;

[0058] The current event is simulated and calculated based on the aforementioned power distribution network simulation calculation model to obtain simulation results;

[0059] The credibility of the simulation results is evaluated, and key issue indicators corresponding to the simulation objectives are extracted from the simulation results.

[0060] Based on the key issue indicators, optimization suggestions are generated, and the optimization suggestions are verified by simulation and compared in terms of effect, and an evaluation report is generated.

[0061] One or more technical solutions provided by this invention have at least the following technical effects or advantages:

[0062] 1. This invention establishes a fully automated framework encompassing event automatic identification, dynamic modeling, simulation execution, and result optimization and verification. It encodes the traditionally experience-dependent simulation work into a series of automatically executable modular functions and data flow rules, thus solving the problems of slow response and rigid processes inherent in traditional methods. This not only significantly improves the response speed and automation level of simulation analysis in real-time operation and planning simulation scenarios, but also provides quantitatively verified optimization suggestions for operational decisions through embedded credibility assessment and solution verification processes, thereby enhancing the reliability of the analysis results and the scientific rigor of decision support.

[0063] 2. An online control layer has been added on top of the basic automated process. During the execution of simulation tasks, evolutionary features are extracted in real time, and extrapolation tasks are intelligently generated based on the current model and electrical sensitivity characteristics to proactively explore future risks. Then, based on the extrapolation results, the execution strategy of the main simulation task is dynamically adjusted. On the one hand, it can detect potential cascading failure risks in advance, guiding the simulation to focus on key issues and enhancing the foresight of complex event analysis; on the other hand, based on the extrapolation conclusions, intelligent decision-making achieves an optimal balance between computational accuracy, data timeliness, and resource efficiency. Attached Figure Description

[0064] The accompanying drawings, which are provided to further illustrate embodiments of the invention and constitute a part of this invention, are not intended to limit the scope of the invention.

[0065] Figure 1This is a schematic diagram of the composition structure of a simulation engine processing system based on a power distribution digital platform in this invention. Detailed Implementation

[0066] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, where there is no conflict, the embodiments of the present invention and the features thereof can be combined with each other.

[0067] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0068] Example 1

[0069] Please refer to Figure 1 Embodiment 1 of the present invention provides a simulation engine processing system based on a power distribution digital platform, the system comprising:

[0070] The event monitoring and parsing module is used to receive real-time monitoring data from the power distribution digitalization platform, identify event information of the power distribution system based on a predefined event triggering rule library, and parse the event information into a scene description structure. The scene description structure includes event type, simulation target, core area range and simulation time span.

[0071] The model assembly module is used for:

[0072] Simulation model units are invoked based on event type and simulation objective;

[0073] Configure the parameters of the simulation model unit to generate a core area simulation instance for this event type;

[0074] Boundary isopleth modules are used for:

[0075] Determine the boundary region adjacent to the core region;

[0076] Based on event type and real-time monitoring data, the boundary region is simplified by equivalent value processing to generate an equivalent model;

[0077] The task execution module is used for:

[0078] The core area simulation instance is assembled with the equivalent model to obtain the power distribution network simulation calculation model;

[0079] The current event is simulated and calculated based on the aforementioned power distribution network simulation calculation model to obtain simulation results;

[0080] The results processing module is used for:

[0081] The credibility of the simulation results is evaluated, and key issue indicators corresponding to the simulation objectives are extracted from the simulation results.

[0082] Based on the key issue indicators, optimization suggestions are generated, and the optimization suggestions are verified by simulation and compared in terms of effect, and an evaluation report is generated.

[0083] In actual implementation, the event monitoring and analysis module continuously receives real-time monitoring data streams uploaded by devices such as SCADA and PMU from the power distribution digitization platform. For example, when the voltage at the end node of line ABC is detected to be below 0.93 pu (per unit value) for three consecutive sampling periods, the module identifies the event according to the predefined "voltage limit exceeded" event triggering rules. Subsequently, the event information is parsed into a structured scenario description structure. This structure includes: the event type is "low voltage"; the simulation objective is "analyze the cause and find an optimization solution"; the core area is set as the topology area with line ABC as the main trunk, including its upstream and downstream related lines and nodes; and the simulation time span is set to the next hour.

[0084] The model assembly module receives the scenario description structure. Based on the "low voltage" event type, it retrieves the network model elements and load model elements related to the core area from the pre-built modular simulation model library. Simultaneously, the boundary equivalence module identifies the external network closely connected to the core area as the boundary region. Assuming real-time monitoring data shows that the boundary region transmits power to the core region with minimal fluctuations, this module selects the load equivalence method, equating the external network to a static load to generate an equivalent model.

[0085] The task execution module electricalally assembles the pre-assembled core area simulation instance with the equivalent model to form a complete distribution network simulation calculation model. This module schedules computing resources to perform time-series power flow calculations on the assembled model, simulating voltage changes in the area over the next hour, and obtaining simulation results including information on voltage at each node and line power flow.

[0086] Finally, the results processing module automatically analyzes the simulation results. For example, it extracts key performance indicators such as "the lowest voltage at the end of line ABC is 0.91 pu" and "the main line load rate is 85%". Based on these indicators and a pre-set optimization rule base, such as "prioritizing reactive power compensation for low voltage", the module generates an optimized suggestion: "install a set of 300kVar capacitors at the end nodes of line ABC". The system then automatically creates a verification task: modify the model, install the capacitors, and re-run the simulation. The verification result shows that the voltage recovers to 0.98 pu. The system generates an evaluation report based on this comparison result (voltage increase of 0.07 pu) and the original solution, for operators to make decisions.

[0087] The simulation model unit includes: a source model unit, a network model unit, a load model unit, and a storage model unit;

[0088] Parameter configuration for the simulation model unit specifically includes:

[0089] Based on real-time monitoring data related to the core area, the initial parameters of the simulation model unit are assigned values;

[0090] The time series parameters of the simulation model unit are generated based on the simulation time span, historical running data, and predicted data.

[0091] Among them, source model units include distributed photovoltaic inverter models and diesel generator models; grid model units include π-type equivalent line models and double-winding transformer models; load model units include ZIP static load models and air conditioning aggregated load models; and storage model units include lithium battery energy storage converter models.

[0092] When configuring parameters, let's take configuring a load model unit as an example. Assume the core area contains a commercial area load. First, the system obtains three types of data from the distribution digitization platform: real-time monitoring data, for example, the current total power of the commercial area is P. n =1500kW,Q n =500kvar; Historical load data: load curve for the same period last week; Forecast data: temperature data for the next hour obtained from weather forecasts. The configuration process is as follows: first, use real-time data P... n Q n This serves as the initial power baseline value for the load model. Then, based on the simulation time span, the module reads historical load curves from the same period and calculates the normalized scaling factor K that varies with time. h (t), and simultaneously, a prediction correction coefficient K is calculated using the temperature prediction value through the load-temperature relationship model. p (t), K pThe calculation of (t) can be performed using regression analysis, specifically using existing techniques in this field. Ultimately, the time series parameters of this load model are generated: P(t) = P n *K h (t)*K p Q(t) and Q(t) are calculated similarly. This generates a dynamic load simulation example that reflects the daily cycle and weather influences.

[0093] The types of simulation calculations include: power flow calculation, time-series power flow calculation, reactive power optimization calculation, voltage regulation calculation, harmonic analysis calculation, and short-circuit analysis calculation.

[0094] The optimization proposal is simulated and its effects compared to generate an evaluation report, including:

[0095] Based on the proposed optimization scheme, the parameters of the simulation instance in the core area are adjusted to obtain a comparison scheme model;

[0096] The simulation calculation was re-executed using the aforementioned comparison scheme model to obtain the comparison simulation results;

[0097] The initial simulation results are compared and analyzed with the comparative simulation results to generate an evaluation report.

[0098] For example, during implementation, the system copies a current core area simulation instance, modifies its reactive power compensation capacity parameter at the "end of ABC line" node from 0 to 300kVar, and generates a comparison scheme model.

[0099] The task execution module uses the same boundary equivalent model, simulation time span, and calculation parameters as the original simulation to re-execute the time-series power flow calculation on the model comparison scheme to obtain the comparison simulation results.

[0100] The system automatically compares the original simulation results with the comparative simulation results and extracts key indicators. For example, the lowest voltage in the original results is 0.91 pu, while the lowest voltage in the comparative results is 0.98 pu; the total active power loss of the entire network per hour in the original results is 850 kWh, while in the comparative results it is 820 kWh. Then, the improvement is calculated: voltage increase +0.07 pu, line loss reduction -30 kWh. Finally, an evaluation report is generated, including a description of the solution, a comparison table of key indicators, quantitative results of the improvement, and conclusions, such as "This solution can effectively solve the low voltage problem and also reduce line loss."

[0101] The credibility assessment of the simulation results includes:

[0102] Based on the preset evaluation rules, the integrity of the input data of the simulation results and the logical consistency of the electrical quantities are checked to obtain the check results;

[0103] The key operational data in the simulation results are compared with historical operational data from the same period to calculate the deviation and obtain the deviation calculation results.

[0104] Based on the inspection results and the deviation calculation results, a reliability score for the simulation results is generated.

[0105] In this embodiment, the total score for input data integrity checking is set at 30 points. The check examines whether the resistance and reactance parameters of all lines in the core topology used in the simulation are complete, and whether there are any missing load power data. If more than 5% of the data is missing, a defect is recorded, and 5 points are deducted for each defect.

[0106] During the logic consistency check, the total score is set at 40 points, and the check items include power balance deviation and voltage over-limit ratio.

[0107] Wherein, power balance deviation = (total generation - total load - total loss) / absolute value of relative error of total generation E b .

[0108] Voltage over-limit ratio P v Specifically, it refers to the proportion of nodes whose voltage exceeds the reasonable range out of the total number of nodes.

[0109] If E b If the percentage is ≤1%, no points will be deducted; if it is 1% <E b ≤5%, deduct 10 points; if E b >5%, deduct 20 points.

[0110] If P v =0, no points deducted; if 0 <P v ≤5%, deduct 10 points; if P v >5%, deduct 20 points.

[0111] When calculating the deviation, the total score is set at 30 points. Actual monitoring data from the same period as the current simulation and under similar meteorological conditions are retrieved from the historical database. Several key indicators are selected, such as main line current and hub node voltage; the specific indicators are determined by those skilled in the art based on the actual situation. The average relative error E between the simulated value and the historical value of each indicator is calculated. i Then calculate the average relative error H of these indicators. e .

[0112] If H e ≤5%, no points deducted; if 5% <H e ≤10%, deduct 15 points; if H e >10%, deduct 30 points.

[0113] Finally, all scores are summed to obtain the credibility score.

[0114] The event triggering rule base includes real-time monitoring event triggering rules and custom event triggering rules; the custom events include business expansion load access events, new energy access events, and electric vehicle charging events.

[0115] Real-time monitoring event triggering rules include: "If the node voltage is <0.95 pu and the duration is >300 seconds, then trigger a low voltage event"; "If the line current is > the long-term allowable current and the duration is >60 seconds, then trigger an overload event".

[0116] User-defined event triggering rules can be manually triggered by planning or scheduling personnel through a human-machine interface or through planned tasks. For example, the trigger condition for a "business expansion load access event" is that a user submits a new application for industrial user access with a capacity of 2000kVA and specifies the access location; the trigger condition for a "new energy access event" is that a 500kW photovoltaic power generation system is planned to be added to a certain transformer area. The system provides different parameter input forms for these two types of events, which are automatically filled into the scene description structure when triggered.

[0117] Example 2

[0118] Based on Embodiment 1, the task execution module further includes a simulation control unit;

[0119] The simulation control unit is used for:

[0120] Real-time monitoring of intermediate simulation results, and extraction of event evolution feature sequences based on intermediate results;

[0121] When the evolution feature sequence meets the preset triggering conditions, a simulation task is generated and executed based on the distribution network simulation calculation model to obtain future simulation results; wherein, the simulation task is used to superimpose short-term disturbances on the current model state to simulate future scenarios;

[0122] A control strategy is generated based on the future projection results, and the execution strategy of the current simulation task is dynamically adjusted based on the control strategy.

[0123] The simulation control unit enables the simulation process to be dynamically adjusted based on the real-time calculated situation.

[0124] The following example illustrates the workflow of this unit, using a scenario where an overhead line L1 in a distribution network experiences a transient ground fault due to a lightning strike, followed by the circuit breaker successfully reclosing automatically.

[0125] When the system performs timing simulation calculations for this fault-reclosing event, the simulation control unit monitors the intermediate result stream of the simulation calculation in real time and collects the instantaneous voltage values ​​U of key nodes, such as nodes A and B near the fault point. A (t), UB The voltage gradient (t) is calculated, along with its rate of change dU / dt and the acceleration d²U / dt² over the simulation time axis. Simultaneously, the spatial distribution of the voltage is analyzed to create a spatial distribution map of the voltage gradient.

[0126] In this event, after the simulation started, the simulation control unit extracted a set of evolution features. After analysis, it was found that the voltage U at node A... A After recovering from near 0 to 0.95 pu, its recovery speed dU A / dt slows down significantly and approaches zero, and the acceleration d²U recovers. A / dt² turns negative; the voltage spatial distribution map shows that the low-voltage area centered on node A has not continued to shrink, but instead shows a slight expansion trend towards adjacent power supply areas. This indicates that the voltage recovery process has stalled, and there is a risk that the affected area will expand.

[0127] When the above feature sequence meets the preset triggering condition, for example, dU A / dt is below the stability threshold and d²U A When the duration of a negative / dt² exceeds the time threshold, the simulation control unit immediately adjusts the time based on the current time (T). k The completed distribution network simulation model is used to initiate a simulation task. The purpose of this simulation task is to simulate a possible new disturbance, such as a large nearby motor failing to start or disconnecting from the grid due to voltage instability, under the current state of stagnant voltage recovery, in order to assess whether a cascading effect will occur in the system.

[0128] The system will analyze the current T k Under certain conditions, the system determines which load switching events in the power grid are most sensitive to the voltage stability of critical nodes (i.e., those with high electrical sensitivity) and, combined with voltage spatial distribution characteristics, automatically decides on the most concerning disturbances, their types, and magnitudes. For example, the model simulates a sudden tripping of a 500kW motor at node C. Then, the system uses T... k Starting from the current state, this perturbation is applied to the model to quickly perform the simulation and obtain the future simulation results.

[0129] The simulation control unit analyzes the simulation results. The simulation assumes that after the motor trips, the voltage at node A further drops to 0.85 pu, triggering an alarm on adjacent line protection. Based on this result, the unit generates a clear control strategy: increasing the analytical depth of the current main simulation task, focusing on the risk of cascading failures. The unit dynamically adjusts the execution strategy of the main simulation task: on the one hand, it requests a higher computational priority from the system resource scheduler; on the other hand, it sends instructions to the model assembly module, requesting a more detailed dynamic simulation model for the motor at node C. For example, a model including its rotor motion equations and excitation characteristics, so that the interaction between such dynamic load events and the power grid can be analyzed more accurately in subsequent calculations of the main simulation. Through this process, the simulation engine achieves an upgrade from passive calculation to active prediction and intervention.

[0130] The process includes generating and executing a simulation task based on the distribution network simulation calculation model, including:

[0131] Based on the aforementioned distribution network simulation model, the electrical sensitivity characteristics of the distribution network are calculated; these electrical sensitivity characteristics are used to quantify the impact of changes in node injected power or branch disconnection on critical electrical quantities.

[0132] Analyze the correlation between the electrical sensitivity characteristics and the evolution characteristic sequence, and determine the disturbance parameters of the short-term disturbance based on the correlation;

[0133] The disturbance parameters are dynamically constructed in the distribution network simulation calculation model, and the simulation task is instantiated and executed.

[0134] In practical implementation, firstly, the electrical sensitivity characteristics are calculated, and the system is based on T. k The distribution network simulation model and current operating point at any given time are used to perform online electrical sensitivity calculations. Specifically, the adjoint sensitivity analysis method is adopted. By linearizing the system power flow equations at the current operating point and solving the corresponding adjoint system equations, the L values ​​of all load nodes in the entire network are calculated. i Active power injection P i Even minor changes have a significant impact on the key performance indicators currently of great concern (identified as the voltage stagnation node V in this implementation). A The partial derivative of the voltage amplitude, i.e., the sensitivity coefficient S. i =∂V A / ∂P i Among them, S i The absolute value of reflects the significance of the load change at node i on the current voltage stability problem. After calculation, the system applies |S| to all load nodes. i Sort the nodes in descending order to identify the set of highly sensitive nodes.

[0135] Then, the system performs spatiotemporal correlation analysis between the electrical sensitivity ranking and the voltage spatial distribution characteristics (spatial distribution map of voltage gradient) extracted in real time from the main simulation. A comprehensive attention score is calculated for each high-sensitivity node. The scoring model simultaneously considers the correlation between the node's sensitivity magnitude and its spatial location with the current abnormal situation. For example, a simple scoring function is: Score = |S i |.α+SpatialCorrelation i *β, where SpatialCorrelation i This is a quantitative indicator characterizing whether node i is located in its gradient diffusion direction. It is obtained as follows: Based on real-time monitoring data, the geometric center and range of the voltage anomaly region are determined. The physical distance from each node to be evaluated to the center of this region is calculated. This distance is then mapped to a correlation value between 0 and 1 using a preset distance decay function (such as an exponential or Gaussian decay function), with closer distances indicating higher correlation. α and β are preset weighting coefficients. The node with the highest overall attention score is automatically determined as the most worthy target for perturbation application.

[0136] Determine the object L to be applied C Then, the system determines the disturbance type based on a pre-set domain knowledge rule base. The rule base matches the current event type with possible chain reaction patterns and outputs the most relevant disturbance type suggestion. In this scenario, the rule base indicates that a "sudden load drop" should be simulated to assess the risk of reverse overvoltage that may be caused by a sudden decrease in reactive power demand.

[0137] The determination of the disturbance amplitude adopts an adaptive critical value exploration strategy, with the system using node L... C Current apparent power S C Based on this, generate a set of increasing trial amplitude sequences, for example, k. m =[0.5,0.7,0.9,1.0]*S C Subsequently, the system used T k Using the state at time step as the initial condition, the simplified derivation calculation is performed iteratively multiple times, evaluating different magnitudes k in sequence. m The dynamic response of the system is then analyzed. By analyzing the results of these rapid simulations, the system can identify a safe range (e.g., k=0.7*S). C (When the system is stable) and a dangerous amplitude (e.g., k=1.0*S) C (System crash). Ultimately, the perturbation amplitude in this simulation was set to a value between the two (e.g., 0.85*S). C The simulation task, instantiated to this extent, can effectively reveal the development path of cascading failures and ensure the stability of numerical calculations, thereby achieving a quantitative assessment of the system's stability margin.

[0138] Based on the above decision results, the system at T k In the simulation model copy at time step, node L C The load power parameter was modified to 15% of its original value (i.e., simulating 85% load shedding). Thus, a simulation model with clear physical meaning and exploratory purpose was instantiated. This model was then submitted to the computational kernel to perform a detailed simulation of a set duration. The output of the future simulation results is used to support the simulation control unit in generating subsequent control strategies.

[0139] The control strategy includes:

[0140] Control strategies to enhance the depth of analysis, update boundary model data, and optimize the allocation of computing resources.

[0141] For example, when the simulation results identify potential serious cascading failure risks, the simulation control unit generates a control strategy to enhance the depth of analysis. The corresponding adjustments include: increasing the priority of computing resources for the main simulation task; and supplementing or switching the key equipment involved in the risk to a more refined simulation model, such as switching from a static impedance model to a dynamic motor model that can describe electromagnetic transient processes, so as to conduct a more in-depth mechanism analysis in subsequent simulations.

[0142] When the development of the main simulation events causes the initial assumptions about the external equivalent boundaries of the power grid to deviate significantly from reality, a control strategy is generated to update the boundary model data. The adjustment process is as follows: the simulation control unit pauses the main simulation task, triggers the boundary equivalent module, and uses the latest real-time power flow and voltage data from the boundary tie lines to recalculate the equivalent parameters of the external network online. The newly generated equivalent model replaces the original boundary model in the main simulation task, and the simulation calculation is seamlessly resumed from the pause point, thereby ensuring the timeliness and accuracy of the simulation boundary conditions.

[0143] When simulation results indicate that the current trend of the simulated event is very clear and will quickly reach a stable state, eliminating the need to complete the entire preset simulation duration, a control strategy for optimizing computational resource allocation is generated. For example, in a simulation of a capacitor switching event, the simulation shows that the voltage stabilizes completely within one second after the switching. In this case, the adjustment is as follows: the simulation control unit continuously monitors the intermediate results of the main simulation. Once it confirms that key indicators such as voltage have entered and remained within the stable range, it automatically terminates the subsequent calculation cycle of the simulation task ahead of schedule. The system will then organize and output complete process data and analysis conclusions from the start of the event to the stabilization point, and release computational resources for other tasks, thereby achieving intelligent optimization of computational efficiency.

[0144] Example 3

[0145] Embodiment 3 of the present invention provides a simulation engine processing method based on a digital power distribution platform, the method comprising:

[0146] Receive real-time monitoring data from the power distribution digitalization platform, identify event information of the power distribution system based on a predefined event triggering rule base, and parse the event information into a scene description structure, which includes event type, simulation target, core area range and simulation time span;

[0147] Simulation model units are invoked based on event type and simulation objective;

[0148] Configure the parameters of the simulation model unit to generate a core area simulation instance for this event type;

[0149] Determine the boundary region adjacent to the core region;

[0150] Based on event type and real-time monitoring data, the boundary region is simplified by equivalent value processing to generate an equivalent model;

[0151] The core area simulation instance is assembled with the equivalent model to obtain the power distribution network simulation calculation model;

[0152] The current event is simulated and calculated based on the aforementioned power distribution network simulation calculation model to obtain simulation results;

[0153] The credibility of the simulation results is evaluated, and key issue indicators corresponding to the simulation objectives are extracted from the simulation results.

[0154] Based on the key issue indicators, optimization suggestions are generated, and the optimization suggestions are verified by simulation and compared in terms of effect, and an evaluation report is generated.

[0155] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0156] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A simulation engine processing system based on a digital power distribution platform, characterized in that, The system includes: The event monitoring and parsing module is used to receive real-time monitoring data from the power distribution digitalization platform, identify event information of the power distribution system based on a predefined event triggering rule library, and parse the event information into a scene description structure. The scene description structure includes event type, simulation target, core area range and simulation time span. The model assembly module is used for: Simulation model units are invoked based on event type and simulation objective; Configure the parameters of the simulation model unit to generate a core area simulation instance for this event type; Boundary isopleth modules are used for: Determine the boundary region adjacent to the core region; Based on event type and real-time monitoring data, the boundary region is simplified by equivalent value processing to generate an equivalent model; The task execution module is used for: The core area simulation instance is assembled with the equivalent model to obtain the power distribution network simulation calculation model; The current event is simulated and calculated based on the aforementioned power distribution network simulation calculation model to obtain simulation results; The results processing module is used for: The credibility of the simulation results is evaluated, and key issue indicators corresponding to the simulation objectives are extracted from the simulation results. Based on the aforementioned key issue indicators, optimization suggestions are generated, and the optimization suggestions are verified through simulation and compared in terms of effectiveness, resulting in an evaluation report. The task execution module includes a simulation control unit; The simulation control unit is used for: Real-time monitoring of intermediate simulation results, and extraction of event evolution feature sequences based on intermediate results; When the evolution feature sequence meets the preset triggering conditions, a simulation task is generated and executed based on the distribution network simulation calculation model to obtain future simulation results; wherein, the simulation task is used to superimpose short-term disturbances on the current model state to simulate future scenarios; A control strategy is generated based on the future projection results, and the execution strategy of the current simulation task is dynamically adjusted based on the control strategy. A simulation task is generated and executed based on the aforementioned distribution network simulation calculation model, including: Based on the aforementioned distribution network simulation model, the electrical sensitivity characteristics of the distribution network are calculated; these electrical sensitivity characteristics are used to quantify the impact of changes in node injected power or branch disconnection on critical electrical quantities. Analyze the correlation between the electrical sensitivity characteristics and the evolution characteristic sequence, and determine the disturbance parameters of the short-term disturbance based on the correlation; The disturbance parameters are dynamically constructed in the distribution network simulation calculation model, and the simulation task is instantiated and executed.

2. The simulation engine processing system based on a digital power distribution platform according to claim 1, characterized in that, The simulation model unit includes: source model unit, network model unit, load model unit, and storage model unit; Parameter configuration for the simulation model unit specifically includes: Based on real-time monitoring data related to the core area, the initial parameters of the simulation model unit are assigned values; The time series parameters of the simulation model unit are generated based on the simulation time span, historical running data, and predicted data.

3. The simulation engine processing system based on a digital power distribution platform according to claim 1, characterized in that, The types of simulation calculations include: power flow calculation, time-series power flow calculation, reactive power optimization calculation, voltage regulation calculation, harmonic analysis calculation, and short-circuit analysis calculation.

4. The simulation engine processing system based on a digital power distribution platform according to claim 1, characterized in that, The proposed optimization scheme is simulated and its effects compared, and an evaluation report is generated, including: Based on the proposed optimization scheme, the parameters of the simulation instance in the core area are adjusted to obtain a comparison scheme model; The simulation calculation was re-executed using the aforementioned comparison scheme model to obtain the comparison simulation results; The initial simulation results are compared and analyzed with the comparative simulation results to generate an evaluation report.

5. The simulation engine processing system based on a digital power distribution platform according to claim 1, characterized in that, The credibility of the simulation results is evaluated, including: Based on the preset evaluation rules, the integrity of the input data of the simulation results and the logical consistency of the electrical quantities are checked to obtain the check results; The key operational data in the simulation results are compared with historical operational data from the same period to calculate the deviation and obtain the deviation calculation results. Based on the inspection results and the deviation calculation results, a reliability score for the simulation results is generated.

6. The simulation engine processing system based on a digital power distribution platform according to claim 1, characterized in that, The event triggering rule base includes real-time monitoring event triggering rules and custom event triggering rules; the custom events include business expansion load access events, new energy access events, and electric vehicle charging events.

7. The simulation engine processing system based on a digital power distribution platform according to claim 1, characterized in that, The control strategy includes: Control strategies to enhance the depth of analysis, update boundary model data, and optimize the allocation of computing resources.

8. A simulation engine processing method based on a digital power distribution platform, characterized in that, The method includes: Receive real-time monitoring data from the power distribution digitalization platform, identify event information of the power distribution system based on a predefined event triggering rule base, and parse the event information into a scene description structure, which includes event type, simulation target, core area range and simulation time span; Simulation model units are invoked based on event type and simulation objective; Configure the parameters of the simulation model unit to generate a core area simulation instance for this event type; Determine the boundary region adjacent to the core region; Based on event type and real-time monitoring data, the boundary region is simplified by equivalent value processing to generate an equivalent model; The core area simulation instance is assembled with the equivalent model to obtain the power distribution network simulation calculation model; The current event is simulated and calculated based on the aforementioned power distribution network simulation calculation model to obtain simulation results; The credibility of the simulation results is evaluated, and key issue indicators corresponding to the simulation objectives are extracted from the simulation results. Based on the aforementioned key issue indicators, optimization suggestions are generated, and the optimization suggestions are verified through simulation and compared in terms of effectiveness, resulting in an evaluation report. Real-time monitoring of intermediate simulation results, and extraction of event evolution feature sequences based on intermediate results; When the evolution feature sequence meets the preset triggering conditions, a simulation task is generated and executed based on the distribution network simulation calculation model to obtain future simulation results; wherein, the simulation task is used to superimpose short-term disturbances on the current model state to simulate future scenarios; A control strategy is generated based on the future projection results, and the execution strategy of the current simulation task is dynamically adjusted based on the control strategy. A simulation task is generated and executed based on the aforementioned distribution network simulation calculation model, including: Based on the aforementioned distribution network simulation model, the electrical sensitivity characteristics of the distribution network are calculated; these electrical sensitivity characteristics are used to quantify the impact of changes in node injected power or branch disconnection on critical electrical quantities. Analyze the correlation between the electrical sensitivity characteristics and the evolution characteristic sequence, and determine the disturbance parameters of the short-term disturbance based on the correlation; The disturbance parameters are dynamically constructed in the distribution network simulation calculation model, and the simulation task is instantiated and executed.

Images