New energy storage and flexible load-containing power grid long-process simulation method

Abstract

The invention discloses a long-process simulation method for a power grid containing new energy storage and a flexible load, and belongs to the technical field of digital simulation of a power system. The method comprises the following steps: configuring a built-in control logic module for a new energy, energy storage and flexible load simulation model in a power grid simulation model; during simulation operation, the control logic module obtains an electrical state quantity from a simulation environment in real time, generates a power adjustment instruction based on a preset rule and dynamically updates a power injection value of a simulation model, and forms a closed-loop feedback of state perception-decision-update-solution; and meanwhile, coordination and self-adaptive adjustment among the modules are supported. According to the invention, the simulation of the dynamic collaborative behavior of the source network load storage is realized in the simulation environment, so that the simulation process can reflect the response and interaction of equipment to the state of the power grid.

Description

Long-term simulation method for power grid including new energy storage and flexible loads Technical Field

[0001] This invention relates to the field of digital simulation technology for power systems, and more specifically, to a long-process simulation method for power grids including new energy storage and flexible loads. Background Technology

[0002] With a high proportion of new energy sources, energy storage, and flexible loads being connected to the power grid, the coupling and interaction among various links in the power system's "source-grid-load-storage" are increasingly strengthened. Long-process power grid simulation has become an important tool for studying system operating characteristics and verifying control strategies. Currently, relevant simulation methods are insufficient in simulating the dynamic interactions of these resources.

[0003] Common simulation methods employ open-loop simulation based on preset power curves. For example, Chinese patent CN112260323A proposes a method that simulates power curve deviations by equating new energy sources, energy storage, and flexible loads to loads. In this method, the power output or consumption of the simulated object is determined by a pre-given curve, and its behavior does not adjust with changes in the grid state during simulation. Therefore, it cannot simulate the dynamic adjustment and coordination processes of resources in actual systems based on real-time signals such as voltage and frequency. Other technical solutions focus on the control of actual physical systems. For example, Chinese patent CN121307928A relates to a real-time smooth control method based on edge computing, which achieves coordinated control of physical devices. However, this technical solution operates on actual power systems, and its control logic is separated from the simulation environment, failing to address how to construct a similar dynamic interaction mechanism within the digital simulation.

[0004] Therefore, existing methods are insufficient to realize a dynamic environment that can autonomously respond to virtual grid conditions and simulate multi-resource collaborative interaction in long-term power grid simulations, thus limiting the ability to simulate and verify the operation behavior of highly interactive new power systems. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a long-process simulation method for power grids including new energy storage and flexible loads, in order to solve the problems mentioned in the background art. The aim is to enable the new energy, energy storage and flexible load models in the simulation process to dynamically adjust and interact in coordination according to the real-time operating status of the virtual power grid, so as to improve the limitations of traditional open-loop simulation methods in simulating the aforementioned dynamic behaviors.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a long-process simulation method for a power grid including new energy storage and flexible loads, the method comprising the following steps: S1, Model construction and configuration step: constructing a power grid digital simulation model, the power grid digital simulation model including a new energy generation simulation model, an energy storage simulation model and a flexible load simulation model; activating and associating a control logic module with at least one controlled simulation model selected in the power grid digital simulation model, the control logic module being built into the simulation process and bound to the controlled simulation model, its function being to embed an autonomous decision-making unit into the simulation model to generate control commands according to the state of the simulation environment; the controlled simulation model is one of a new energy generation simulation model, an energy storage simulation model or a flexible load simulation model; S2, Closed-loop simulation execution step: running the power grid digital simulation model to perform long-process time-series simulation, in each global simulation step of the time-series simulation, sequentially executing the following sub-steps to form a dynamic feedback closed loop: S21, State quantity reading step: for each state quantity bound to the control logic module... The controlled simulation model reads at least one electrical state quantity of the simulation bus connected to the controlled simulation model from the real-time solution results of the power grid digital simulation model. This process is used to provide real-time input data for decision-making. S22, Control command generation step: The control logic module executes its built-in calculation rules according to the electrical state quantity read in step S21 to generate a power adjustment command for the controlled simulation model bound to it. This step is the core of calculating the adjustment strategy. S23, Model parameter update step: Based on the power adjustment command generated in step S22, the power setpoint used by the controlled simulation model bound to it for power grid simulation calculation within the global simulation step is corrected. This operation applies the decision result to the simulation model and changes its operating state. S24, Full network state solution step: Based on the corrected power setpoints of all simulation models, power flow calculation or electromechanical transient calculation is performed to update the full network electrical state of the power grid digital simulation model and advance the simulation clock to the next global simulation step. This step updates the system state and drives the next round of feedback loop.

[0007] Furthermore, the electrical state quantities mentioned in step S21 are one or more of the voltage amplitude of the simulated bus, system frequency, injected active power, or injected reactive power. These quantities are the direct basis for assessing the local power grid status and determining regulation needs.

[0008] Furthermore, the built-in calculation rule described in step S22 includes the following execution process: the control logic module performs electrical state quantity change deduction within the next L global simulation steps based on the current global simulation step size and the electrical state quantity data obtained within the previous K consecutive global simulation steps, and determines the power adjustment command according to the technical parameter boundaries of the controlled simulation model it is bound to; wherein, K and L are pre-set positive integers. This process uses historical and predictive information to make forward-looking decisions in order to cope with system fluctuations in advance.

[0009] Furthermore, the technical parameter boundaries include: when the controlled simulation model is an energy storage simulation model, the absolute value of its power injection value does not exceed its rated power parameter, and the simulation state quantity value representing the energy storage level is maintained between the first preset value and the second preset value to ensure the safe operating life of the energy storage device; when the controlled simulation model is a new energy power generation simulation model, the absolute value of the change in its power injection value between two adjacent global simulation steps does not exceed the preset power change rate parameter to simulate the ramp-up limitations of physical equipment.

[0010] Furthermore, the power setting value correction mentioned in step S23 involves algebraically adding the power change corresponding to the power adjustment command to the value of the pre-stored reference power data at the corresponding simulation time, thereby combining the planned curve with real-time adjustment to form the actual power value executed in the simulation.

[0011] Furthermore, the cyclic execution cycle consisting of steps S22 and S23 is configured to be smaller than the global simulation step size cycle of step S24, so that before a full network state calculation, the control logic module can complete multiple rapid local decisions and parameter pre-adjustments to simulate the characteristics of the control equipment response speed being much faster than the electromechanical dynamics of the power grid in reality.

[0012] Furthermore, during the timing simulation, the control logic module adjusts the internal parameters in its built-in calculation rules based on the correlation between the electrical state data sequence and the power regulation command sequence recorded in the historical global simulation step. This process enables the control logic to slowly self-optimize to adapt to long-term simulation operation characteristics.

[0013] Furthermore, in step S1, the control logic modules bound to the controlled simulation models connected to simulation buses of different voltage levels have built-in calculation rules that are independent of each other and can be different, so as to support the simulation of scenarios in a single simulation case where different network levels or regions adopt differentiated control strategies.

[0014] Furthermore, in step S22, the built-in calculation rule calculates the cumulative power adjustment of the controlled simulation model bound to it within the most recent T global simulation steps when determining the power adjustment command, and keeps the cumulative amount within a preset threshold range as one of the calculation conditions; where T is a positive integer, this constraint is used to limit the device from excessively frequent adjustments in a short period of time, simulating its physical tolerance or contractual constraints.

[0015] Furthermore, in step S22, when the control logic module determines the power adjustment command, it receives the adjustment status information of other control logic modules in the power grid digital simulation model and performs joint calculations according to the pre-set coordination rules. This process realizes cross-device collaborative decision-making to ensure that local adjustment actions do not violate system-level operating limits.

[0016] The technical effects and advantages of this invention are as follows: First, by introducing a control logic module and a closed-loop feedback mechanism, the simulation model acquires autonomous adjustment capabilities based on state awareness. During simulation execution, the control logic module obtains real-time electrical state quantities such as voltage and power from its bound simulation model access points. Based on preset calculation rules, the module generates power adjustment commands based on these state quantities and modifies the power injection value of the simulation model in the next calculation step. After the modified parameters participate in the full-network solution, the resulting new grid state is fed back to the control logic module, driving the next round of decision-making. This closed-loop process enables distributed resources in the simulation to continuously adjust their output or power consumption according to changes in the virtual grid state, changing the traditional simulation model where resource behavior is entirely determined by preset curves, and improving the dynamics of the simulation process.

[0017] Secondly, by designing the coordination and adaptive functions of the control logic module, the simulation of system-level collaborative and long-term evolutionary behavior is supported. When making decisions, the control logic module can follow multiple conditions, including equipment operating constraints and cumulative adjustment limits, and can collaboratively calculate adjustment instructions with other modules through information exchange and coordination rules, thereby simulating regional resource aggregation and optimal allocation. Furthermore, the module can adjust its internal parameters based on historical simulation data, simulating the slow self-adjustment process of the control strategy. These designs enable this simulation method not only to simulate the response of a single device but also to simulate the interaction of multiple agents under complex rules and their long-term impact on system operation, providing a more effective digital simulation environment for studying collaborative control strategies and market mechanisms. Attached Figure Description

[0018] Figure 1 is a flowchart of the overall simulation execution of the present invention; Figure 2 is a schematic diagram of the internal decision-making process of the control logic module of the present invention; Figure 3 is a schematic diagram of the branch structure of the power regulation instruction calculation rule of the present invention; Figure 4 is a schematic diagram of the branch structure of the multi-timescale simulation execution of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Example 1, as shown in Figures 1 to 4, illustrates a long-process power grid simulation method incorporating new energy storage and flexible loads, implemented in long-process digital simulation software for power grids. This software can be a secondary development environment based on a commercial software platform or a self-developed dedicated simulation system. The following is a typical implementation case: I. Specific Implementation of Model Construction and Configuration Steps This step is completed during the initialization phase of the simulation project or case. Its main task is to embed the control structure of this invention into the conventional power grid simulation model.

[0021] 1. Basic Construction of Power Grid Digital Simulation Model: Using the graphical interface or modeling scripts of simulation software, construct a complete power system model including transmission and distribution networks. This model should include at least the following traditional components: AC buses, transmission lines, transformers, conventional synchronous generators, and static loads. Based on this, focus on adding the following three types of simulation models: New Energy Power Generation Simulation Models: For example, simulating a 100 MW photovoltaic power plant model and a 150 MW wind farm model. These models need to have their output characteristic curves, grid-connected inverter characteristics, and related protection parameters configured.

[0022] Energy storage simulation model: For example, simulating a lithium-ion battery energy storage system model with a rated power of 50 MW and an energy storage capacity of 100 MWh. This model needs to be configured with parameters such as charge and discharge efficiency, self-discharge rate, rated power, and upper and lower limits of state of charge (e.g., minimum state of charge of 20% and maximum state of charge of 95%).

[0023] Flexible load simulation models: For example, simulating an intermittent industrial load aggregation model with a total capacity of 30 MW, or an electric vehicle charging station cluster model with a total power of 20 MW. These models need to be configured with flexible characteristic parameters such as their base power curve, adjustable power range, maximum interruption duration, and adjustment response time.

[0024] All models are connected to specific nodes of the power grid digital simulation model via their electrical connection parameters. This setup process utilizes existing simulation technology, and its output is a power grid digital simulation model file containing complete component parameters.

[0025] 2. Creation and Binding of the Control Logic Module: In this step, a software functional entity named the control logic module is created in the simulation system through software development. This module is an independent object containing internal state variables, calculation functions, and communication interfaces.

[0026] Selecting a Controlled Simulation Model: The user or configuration script selects the objects to participate in the dynamic co-simulation from the existing models. In this embodiment, we selected the aforementioned 100 MW photovoltaic power plant model, 50 MW / 100 MWh energy storage system model, and 30 MW interruptible load model as controlled simulation models.

[0027] Instantiation and Binding: For each selected controlled simulation model, an independent control logic module is instantiated. The binding relationship between the two is established through the application programming interface or plug-in mechanism provided by the simulation system. Specifically: (a) Data Input Link Establishment: The control logic module is registered as a state listener for the bus connected to the controlled simulation model. After the simulation engine completes its calculations, the electrical state quantities of the bus, such as voltage, frequency, and power, will be automatically transmitted to the corresponding control logic module through callback functions or shared memory areas.

[0028] (b) Control Output Link Establishment: Grant the control logic module write permissions to the power setpoint attribute of the controlled simulation model it is bound to. The control logic module can directly modify the power injection value used by the model in the simulation.

[0029] Built-in and Activated: The running code of the control logic module is compiled or interpreted and executed within the main process space of the simulation software. Its lifecycle is bound to the simulation case, and its calculations are driven by simulation clock events. After configuration, its state is set to active.

[0030] II. Specific Implementation of Closed-Loop Simulation Steps After the simulation case starts, it enters the main loop of long-process time-series simulation. The global simulation step size is set to... Seconds. This means that the entire network state is refreshed every 5 seconds the simulation time advances. Within each 5-second step, the following closed-loop process is executed: Step S21, Specific implementation of the simulation state acquisition step: At the simulation time... After the simulation engine completes the power flow calculation or electromechanical transient calculation of the entire network, it obtains the data for all buses in the system. The steady-state or dynamic solution at time t.

[0031] Assume the control logic module, bound to the 50 MW energy storage model, has its controlled simulation model connected to bus numbered BUS305. The simulation engine publishes updated bus data via its internal data bus.

[0032] The status acquisition function of the control logic module is triggered, which accesses the bus data and reads the instantaneous values ​​of the following electrical status quantities of the BUS305 bus: voltage amplitude. pu; system frequency Hz; Active power injected into the bus MW (negative values ​​indicate power absorbed from the bus, i.e., charging state); reactive power injected into the bus. Mvar; the control logic module stores these instantaneous values ​​in its internal data buffer, which stores historical data from the most recent global simulation steps in chronological order, for example... .

[0033] Step S22, Specific implementation of the adjustment command calculation step: Next, the control logic module calls its core calculation rule function to generate power adjustment commands.

[0034] As a specific implementation method, the calculation rule performs the following process: 1. Data preparation and change prediction: Set parameters , The module retrieves the current and past data from the buffer. The busbar of each step size always injects active power into the sequence. Using first-order extrapolation to predict the future. Power value per step: , , .

[0035] Calculate the predicted power and a desired reference value. Deviation sequence: .

[0036] in, Can be set to current power Or through first-order hysteresis filtering A smoothed reference value is obtained. For filter coefficients ( ).

[0037] 2. Constraint Boundary Check: The module queries the technical parameter boundaries of the controlled simulation model it is bound to.

[0038] For energy storage simulation models: Check the current simulated state of charge. Because it is within the safe range of 20% to 95%, the charging and discharging power command is... MW to Valid within the MW (rated power) range. The positive direction of power injection is set to output power to the grid.

[0039] For simulation models of new energy power generation (such as photovoltaics): check the power change rate parameter. Assuming the limit is MW / min. Converted to The step size is seconds, and the maximum allowable variation between adjacent step sizes is... MW.

[0040] 3. Instruction Solving: To minimize future prediction bias With the sum of squares as the objective, and under the constraints of step 2, solve for the power adjustment command that should be issued at the current step size. .

[0041] This optimization problem can be formalized as: finding , so that the objective function Minimum, while satisfying and power change rate constraints.

[0042] The physical meaning of this objective function is: to find a regulation command. This allows the instruction, when applied to the system, to offset predicted power deviations at multiple future time points. .in, This is a regularization weight coefficient (typically ranging from 0.001 to 0.1) used to avoid the calculated instructions... A larger amplitude enhances the stability of the solution.

[0043] As a simplified implementation, a proportional control rule can be used: ,in The preset proportional coefficient is based on the model's adjustment capability (typically ranging from 0.1 to 2.0, and can be set to 0.5 in this example). This is the preliminary calculated power adjustment amount, and its sign is conventionally consistent with the positive direction of power injection.

[0044] 4. Cumulative quantity check: Set parameters (That is, the most recent 12 steps, corresponding to 60 seconds). The module calculates its bound controlled simulation model in the past... Cumulative power adjustment within each step: ,in It is a historical instruction sequence. If... It has approached the upper limit of the preset threshold range. (For example, 15MWh), then it will be reduced proportionally in this calculation. Size, ensure .

[0045] 5. Coordinated Calculation: After completing their local preliminary calculations, all control logic modules will share their respective results. It is published to a shared communication area. The module reads adjustment commands published by other control logic modules within the same coordination area from this area. .

[0046] According to preset coordination rules, such as ensuring that the algebraic sum of the adjustment commands of all modules (with the increase of grid output as the positive direction) does not exceed the system-level adjustment limit. MW, perform joint calculations. If Then, each module is reduced proportionally according to its preset weights, resulting in the final module. .

[0047] It should be noted that the above parameters ( , , , , , , The specific values ​​(etc.) are for illustrative purposes only. In practical applications, those skilled in the art can configure and adjust them according to the scale, dynamic characteristics, control objectives, and accuracy requirements of the simulated power grid.

[0048] Step S23, Specific Implementation of Model Input Update Step: The control logic module will calculate and determine the final power adjustment command. (For example, +15.3MW means a request to increase output by 15.3MW or reduce load by the same amount.) This is sent to the controlled simulation model it is bound to via the write interface.

[0049] Update logic: The controlled simulation model has a pre-stored reference power curve. It originates from historical data or forecast plans. After receiving the instruction, the model performs an update operation: This calculation yielded... That is, the next global simulation step size of the model ( The power injection value used when participating in the network-wide calculation at any given time.

[0050] Step S24, Specific Implementation of the Full Network Solving Advancement Step: The simulation engine collects the updated power injection values ​​of all simulation models. (For unbound models, the values ​​are taken directly from the baseline power curve.)

[0051] The engine uses these power injection values ​​as boundary conditions and invokes its built-in numerical solver to perform full-network calculations. The solver's inputs are the power injection values ​​for all buses, the grid network topology, and parameters; its output is... The latest electrical status of the entire network, including the voltage, phase angle, frequency of all buses, and power of each branch, is recorded at all times.

[0052] Simulated clock from Advance to time.

[0053] The updated network status then becomes the real-time calculation result obtained by each control logic module in the next step S21, thereby driving the execution of the next closed loop.

[0054] III. Implementation of Multi-Level Execution Cycles As an extended implementation scheme, to simulate the characteristic that the response of actual control equipment is faster than the slow dynamics of the power grid, this invention supports multi-timescale configuration. Using seconds as the global simulation step size balances computational efficiency and accuracy for simulating slow-motion electromechanical processes in power grids (typically with time constants ranging from several seconds to several minutes). The local cycle of the control logic is set to... The second is used to simulate the fast response capability of actual controllers (such as energy storage converters and load controllers), which is typically in the range of milliseconds to seconds.

[0055] Global simulation step size: set to Seconds are used for the entire network state calculation step (S24).

[0056] Local control cycle: The adjustment instruction calculation step (S22) and model input update step (S23) within the control logic module are as follows: Executes at a frequency of seconds.

[0057] Implementation method: In Within a global step of seconds, the simulation engine drives the time after completing one S24. Advance to However, internally, it can be divided into... indivual The sub-interval is in seconds. In each sub-interval, the control logic module, based on the latest bus state (obtainable through interpolation, or approximating that the state remains unchanged),... Continuous operation with a period of seconds The next fast local computation (S22) and parameter pre-update (S23) are performed.

[0058] However, only when the simulation clock arrives At this global step point, the controlled simulation model will formally adopt the most recent (or a combination of local calculations over a period of time) local calculation results as the power injection value for the next global step. This simulates the process of the controller performing high-speed calculations but executing commands in a slow, dynamic synchronization with the power grid.

[0059] IV. Implementation of Adaptive and Cooperative Mechanisms: Adaptive Adjustment: During simulation, the control logic module continuously records the historical power regulation command sequences it generates. And the corresponding sequence of actual bus active power changes after the instruction takes effect. .

[0060] in, ,and It was precisely in Power regulation commands were applied at all times. Afterwards, the simulation engine at any time The calculated actual power injection value.

[0061] By analyzing the relationships between these data (e.g., calculation instructions) With subsequent actual power changes (Regarding the gain relationship), the module can call a slow parameter self-tuning algorithm (such as gradient descent) to dynamically adjust the internal parameters (such as the scaling factor) in its calculation rules. This allows the actual control effect to be closer to the expectation.

[0062] Heterogeneous strategy configuration: In the initialization step S1, different built-in calculation rules can be configured for control logic modules that are bound to different types of controlled simulation models or connected to buses of different voltage levels.

[0063] For example, configuring rules for energy storage modules on the transmission network side with the core objective of mitigating regional power fluctuations, Derived from regional tie-line planned values; rules for configuring photovoltaic modules on the distribution network side with the core objective of maintaining stable voltage at the connection point. These rules are calculated based on voltage deviation. They are implemented by loading different algorithm function libraries.

[0064] System-level collaboration: As another extended implementation scheme, a dedicated collaborative computing module can be deployed in the simulation system. In step S22, after completing their local preliminary calculations, each control logic module sends its local adjustment requirements and capacity limits to this collaborative computing module.

[0065] The collaborative computing module aggregates all information and performs optimization calculations with the global goals of avoiding limit violations and achieving optimal economy. It then distributes the optimized and allocated instructions to each control logic module. Each module executes the final instruction based on this allocation result.

[0066] V. Simulation Verification and Comparative Analysis To evaluate the effectiveness of the method proposed in this invention, a simulation test environment was built based on the above embodiments. The test system is based on a simplified model of a local power grid, including a 220kV main grid and multiple 10kV distribution networks. At one of the distribution network nodes (numbered N10), a 100MW photovoltaic power plant (PV), a 50MW / 100MWh energy storage system (ESS), and a 30MW interruptible industrial load (IL) are connected.

[0067] The simulation software uses the PandaPower library in the Python environment for power flow calculations, and a custom-developed control logic module plugin was also included. The total simulation duration was set to 24 hours (0:00-24:00) to simulate a complete daily operating cycle. Global simulation step size... Set to 5 seconds, local control cycle Set to 1 second.

[0068] The photovoltaic output and base load curves are based on measured data from a typical day in the region, and are superimposed with random fluctuations that conform to their characteristics (the standard deviation of photovoltaic fluctuations is set to 5% of the rated output, and the standard deviation of load fluctuations is set to 2%).

[0069] For comparison, two simulation scenarios were designed: 1. Baseline scenario (open-loop mode, BS): PV, ESS, and IL all operate strictly according to their predefined baseline power curves. ESS executes a fixed "two-charge, two-discharge" schedule (charging from 10:00 to 14:00 and discharging from 18:00 to 22:00) and does not respond to the real-time status of the power grid. IL does not participate in any regulation.

[0070] 2. Scenario of this invention (closed-loop mode, PMS): According to the method of this invention, control logic modules are configured for the PV, ESS, and IL models respectively. The control objective is set to minimize the net load power of node N10. Fluctuations.

[0071] The parameter settings for each module are as follows: , , , , , , , MW, the safe range of energy storage SOC is [20%, 95%], and the maximum charge / discharge power is MW, gradeability limit is MW / min, cumulative adjustable energy limit MWh.

[0072] Both scenarios were run under identical initial conditions, network parameters, and external perturbation sequences. Key operational metrics were recorded and analyzed, and the comparison results are shown in the table below: Simulation results show that the system's operating parameters changed after applying the method of this invention. The node voltage qualification rate improved, while the standard deviation of voltage fluctuation, the peak-to-valley difference of net load, and the volatility all decreased. The regulation capabilities of energy storage and flexible loads were utilized, and the tracking error of photovoltaic output decreased.

[0073] This simulation demonstrates that the method proposed in this invention can simulate the dynamic coordination and closed-loop control of source, network, load, and storage resources in a digital simulation environment. The data is derived from a typical run of the specific test case, and the actual results may vary depending on the specific network structure, parameter configuration, and disturbance scenario.

[0074] Through the detailed and specific implementation methods and simulation verification described above, those skilled in the art can clearly understand how to fully construct and run the long-process power grid simulation method protected by this invention, which includes endogenous, closed-loop, and collaborative feedback mechanisms, within an existing power grid digital simulation platform through software module development, interface integration, and algorithm implementation. This method enables the simulation models of new energy sources, energy storage, and flexible loads to respond to changes in the virtual power grid's state, thereby reproducing the dynamic process of coordinated operation of power generation, grid, load, and storage in the digital space.

[0075] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A long-process simulation method for a power grid incorporating new energy storage and flexible loads, characterized in that, The method includes the following steps: S1, Model Construction and Configuration Step: Constructing a power grid digital simulation model, which includes a new energy power generation simulation model, an energy storage simulation model, and a flexible load simulation model; configuring and activating a control logic module for at least one selected controlled simulation model in the power grid digital simulation model, the control logic module being built into the simulation process and bound to the controlled simulation model; the controlled simulation model is one of the new energy power generation simulation model, the energy storage simulation model, or the flexible load simulation model; S2, Closed-Loop Simulation Execution Step: Running the power grid digital simulation model for long-process time-series simulation, in each global simulation step of the time-series simulation, the following sub-steps are executed sequentially: S21, Simulation State Acquisition Step: For each controlled simulation model bound to the control logic module, From the solution results of the power grid digital simulation model at the current global simulation step, obtain at least one electrical state quantity of the simulation bus connected to the controlled simulation model; S22, Adjustment command calculation step: The control logic module executes its built-in calculation rules according to the electrical state quantity obtained in step S21 to generate a power adjustment command for the controlled simulation model bound to it; S23, Model input update step: Based on the power adjustment command generated in step S22, update the power injection value used by the controlled simulation model bound to it for power grid simulation calculation in the next global simulation step; S24, Full network solution advancement step: Based on the updated power injection values ​​of all simulation models, execute the power grid network equation solution, update the state of all electrical quantities in the power grid digital simulation model, and advance the simulation time to the next global simulation step.

2. The long-process simulation method for power grids including new energy storage and flexible loads according to claim 1, characterized in that, The electrical state quantities mentioned in step S21 are one or more of the following: voltage amplitude of the simulated bus, system frequency, injected active power, or injected reactive power.

3. The long-process simulation method for power grids including new energy storage and flexible loads according to claim 1, characterized in that, The built-in calculation rule in step S22 includes the following execution process: the control logic module performs electrical state quantity change deduction in the next L global simulation steps based on the current global simulation step size and the electrical state quantity data obtained in the previous K consecutive global simulation steps, and determines the power adjustment command according to the technical parameter boundary of the controlled simulation model it is bound to; wherein K and L are pre-set positive integers.

4. The long-process simulation method for power grids including new energy storage and flexible loads according to claim 3, characterized in that, The technical parameter boundaries include: when the controlled simulation model is an energy storage simulation model, the absolute value of its power injection value does not exceed its rated power parameter, and the simulation state quantity value representing the energy storage level is maintained between the first preset value and the second preset value; when the controlled simulation model is a new energy power generation simulation model, the absolute value of the change in its power injection value between two adjacent global simulation steps does not exceed the preset power change rate parameter.

5. The long-process simulation method for power grids including new energy storage and flexible loads according to claim 1, characterized in that, The updated power injection value in step S23 is to algebraically add the power change corresponding to the power adjustment command to the value of the pre-stored reference power data at the corresponding simulation time.

6. The long-process simulation method for power grids including new energy storage and flexible loads according to claim 1, characterized in that, The execution cycle of the loop consisting of steps S22 and S23 is smaller than the global simulation step size cycle of step S24.

7. The long-process simulation method for power grids including new energy storage and flexible loads according to claim 1, characterized in that, During the timing simulation, the control logic module adjusts the internal parameters in its built-in calculation rules based on the correlation between the electrical state data sequence and the power regulation command sequence recorded in the historical global simulation step.

8. The long-process simulation method for power grids including new energy storage and flexible loads according to claim 1, characterized in that, In step S1, the control logic modules bound to the controlled simulation models connected to simulation buses of different voltage levels have independent and different built-in calculation rules.

9. The long-process simulation method for power grids including new energy storage and flexible loads according to claim 1, characterized in that, In step S22, the built-in calculation rule calculates the cumulative power adjustment of the controlled simulation model bound to it within the most recent T global simulation steps when determining the power adjustment command, and keeps the cumulative amount within a preset threshold range as one of the calculation conditions; where T is a positive integer.

10. The long-process simulation method for power grids including new energy storage and flexible loads according to claim 1, characterized in that, In step S22, when determining the power regulation command, the control logic module receives the regulation status information of other control logic modules in the power grid digital simulation model and performs joint calculations according to the pre-set coordination rules.

Images