Wind power-photovoltaic-energy storage-svg joint simulation system modeling method based on PSModel

Abstract

This invention provides a modeling method for a joint simulation system of wind power, photovoltaic, energy storage, and SVG based on PSModel. First, the control characteristics of the wind power, photovoltaic, energy storage, and SVG systems are analyzed to establish a system model containing the main electrical circuit and control strategy, accurately reflecting the dynamic characteristics and interactive effects of each device after connection. Second, based on the model characteristics, a PSModel encapsulated dynamic library containing initialization, calculation, and termination stages is constructed to form a black-box model, decoupling the internal algorithm from the external environment and facilitating integration and invocation in large-scale networks. Finally, based on modular design principles, the complex system is divided into highly cohesive and loosely coupled independent sub-modules, defining connection relationships and building a joint grid-connected system model with a hierarchical architecture. This invention solves the integration problem of joint simulation of complex new energy systems, exhibiting high dynamic response consistency, strong accuracy and compatibility, and effectively supporting the stability analysis and control strategy verification of high-proportion new energy combined power systems.

Description

Technical Field

[0001] This invention relates to the field of new energy simulation modeling technology, and more specifically, to a modeling method for a wind power-photovoltaic-energy storage-SVG joint simulation system based on PSModel. Background Technology

[0002] With the goal of "carbon peaking and carbon neutrality" being proposed, the proportion of new energy sources in the power system continues to increase. The combined grid connection of wind power, photovoltaics, energy storage, and static synchronous compensators has become an important development direction for the construction of new power systems. This multi-energy and control device combined grid connection operation mode can effectively promote the large-scale consumption of clean energy by the power grid and the low-carbon transformation of the overall energy structure. However, the high proportion of new energy power electronic equipment connected makes the dynamic interaction between various independent systems in the system extremely complex. Traditional simulation modeling methods struggle to accurately reflect the electrical dynamic characteristics and joint fault response patterns under such complex scenarios of multi-physical source coupling, posing a severe challenge to the safe and stable operation and dispatch control of modern power systems.

[0003] Currently, most simulation analyses for new energy grid integration in this field are limited to isolated modeling studies of single power electronic devices. There is a general lack of integrated simulation modeling solutions for wind power-photovoltaic-energy storage-static synchronous compensator (SPCC) joint operation systems. More challenging is the fact that existing mathematical models provided by various equipment manufacturers often suffer from closed underlying code and inconsistent data interface standards, leading to significant integration difficulties and severe compatibility issues in large-scale power grid platforms. When these existing single-device models are forcibly combined, their joint simulation dynamic response characteristics have extremely limited consistency with the actual physical system, completely failing to meet the needs of large power grids for refined joint simulation analysis in the face of grid transient disturbances or complex fault scenarios. Therefore, there is an urgent need to explore and construct an efficient, accurate, and highly compatible multi-new energy device joint simulation modeling method to break down the barriers of existing single-device modeling and provide reliable technical support for the construction of new power systems. Summary of the Invention

[0004] The present invention aims to solve at least one of the aforementioned technical problems existing in the prior art.

[0005] To this end, the first aspect of the present invention provides a modeling method for a joint simulation system of wind power-photovoltaic-energy storage-SVG based on PSModel.

[0006] This invention provides a modeling method for a joint simulation system of wind power-photovoltaic-energy storage-SVG based on PSModel, including: The control characteristics of wind turbines, photovoltaic power generation devices, energy storage power stations, and static synchronous compensators are analyzed separately. Independent system models containing the electrical main circuit topology and corresponding control strategies are established for each device to characterize the electrical dynamic response characteristics of each device. The running parameters and algorithm features of each independent system model are extracted, and PSModel dynamic link libraries encapsulating the corresponding system model code are constructed respectively. Each dynamic link library is configured with an initialization stage interface for allocating memory, a simulation calculation stage interface for executing transient logic, and a termination stage interface for releasing resources. In this way, the independent system models of each device are transformed into black box models decoupled from the external calling environment. The PSModel simulation platform adopts a modular and layered architecture, instantiating the black-box models of each device into independent sub-modules with high cohesion and low coupling. Based on the actual physical topology of the wind-solar-storage grid connection, the external electrical connection relationships and interaction nodes between the independent sub-modules are configured in the simulation platform to assemble a wind power-solar-storage-SVG joint simulation system model.

[0007] The PSModel-based wind power-photovoltaic-energy storage-SVG co-simulation system modeling method according to the above-described technical solution of the present invention may also have the following additional technical features: In the above technical solution, the wind turbine adopts a doubly-fed asynchronous wind turbine model. Its system model includes a wind turbine, a phase-locked loop, a shaft system, a doubly-fed asynchronous motor, a grid-side filter, and a dual PWM converter. The dual PWM converter includes a grid-side converter and a rotor-side converter. The grid-side converter adopts grid voltage-oriented vector dual closed-loop control, with constant DC voltage control on the d-axis and constant current control on the q-axis. The rotor-side converter controls the active and reactive power output of the wind turbine by adjusting the rotor current.

[0008] In the above technical solution, the model of the photovoltaic power generation device includes a three-phase photovoltaic grid-connected power generation inverter with a voltage-source three-phase bridge inverter structure. Its control strategy includes an outer loop d-axis DC voltage control and q-axis reactive power control to achieve maximum power point tracking of the photovoltaic array, and an inner loop PI current control that considers the influence of dq-axis cross-coupling terms.

[0009] In the above technical solution, according to the modeling method of claim 1, the model of the energy storage power station includes a battery pack external characteristic model composed of individual battery cells connected in series and parallel, a converter with an LCL filter, and a multi-scenario dynamic simulation control model. The battery pack external characteristic model adopts an internal potential and internal resistance structure of individual battery cells considering the state of charge. The multi-scenario dynamic simulation control model includes a plant-level control model for responding to grid-connected regulation, a normal operation control model, a fault ride-through control model, and a current limiting model for preventing converter overcurrent.

[0010] In the above technical solution, the model of the static synchronous compensator is based on a voltage source converter with an H-bridge cascaded structure. The DC side uses a DC capacitor as an energy storage support element. The DC side voltage is converted into an AC voltage with the same frequency as the power grid by the voltage source converter. By comparing the phase and amplitude with the AC voltage at the grid connection point, the continuous generation or absorption of reactive power can be controlled.

[0011] In the above technical solution, the initialization phase interface of the PSModel dynamic link library allocates memory space for the model instance pointer by passing in the total number of model parameters, the total number of input and output variables, and the simulation step size; the simulation calculation phase interface is called within each calculation step to read data from the external public storage area and perform calculations of the model function; the termination phase interface is used to perform memory release operations when the simulation terminates.

[0012] In the above technical solution, the modular layered architecture adopted in the PSModel simulation platform includes: combining and encapsulating generators, motors, converters and control systems into unit-level models, further combining and encapsulating multiple unit-level models into wind power plant-level models, and finally connecting multiple plant-level models to the regional power grid model to form a unified and hierarchical dynamic system topology.

[0013] In the above technical solution, the parameters of the LCL filter of the converter are selected to meet the following conditions: the voltage drop generated by the inductive impedance is less than 10% of the voltage under the rated operating conditions of the power grid, the resonant frequency of the filter is 10 times higher than the power grid frequency and less than half of the switching frequency, and the reactive power absorbed by the filter capacitor is less than 5% of the rated active power of the system.

[0014] In the above technical solution, in the multi-scenario dynamic simulation control model, the power control command generated by the plant-level control model is used to generate a current command through the normal operation control model and the fault ride-through control model. The current limiting model limits the current command according to the current state of charge of the battery pack to form a dq-axis injected current. Finally, the grid connection interface model converts the dq-axis injected current into a dq-axis voltage command and injects it into the AC grid.

[0015] In the above technical solution, the interface function of the PSModel dynamic link library also includes a snapshot state input identifier variable. When the identifier variable is true, the initialization phase interface directly reads the historical cross-sectional data saved at the previous steady state moment from the external public storage area as the initial value of the model's internal state variable, so as to achieve a smooth start-up of the co-simulation system.

[0016] In summary, due to the adoption of the above-mentioned technical features, the beneficial effects of the present invention are: The modeling method proposed in this invention completely breaks through the limitations of traditional single-device simulation modeling, and fully considers the dynamic characteristics and deep interactive effects of the combined system of wind power, photovoltaic, energy storage, and static synchronous compensator at the whole system level. By establishing a system model that includes the underlying electrical main circuit and customized control strategies, this method can accurately reflect the real dynamic characteristics and multi-source coupling correlation of each new energy device after it is connected to the power grid.

[0017] Within this technical framework, this invention creatively constructs a PSModel encapsulated dynamic link library based on model characteristics, comprising three standard execution phases: initialization, calculation, and termination. This forms a complete black-box model structure, perfectly achieving deep decoupling between the internal complex algorithm model and the external physical simulation environment, greatly facilitating rapid integration and seamless invocation in large-scale network environments. By introducing a modular design core concept of high cohesion and low coupling, the large and complex joint system is scientifically divided into independent sub-modules, and the node connection relationships are defined with a clear hierarchical architecture. Under the premise of strictly ensuring the accuracy of electromagnetic transient simulation, this invention solves the long-standing industry pain points of extremely high integration difficulty and severe lack of cross-platform compatibility in complex new energy joint systems.

[0018] This modular, encapsulated black-box design not only simplifies the physical structure of mega-power systems from a macroscopic perspective but also significantly enhances the reusability of the underlying model units, providing extremely flexible system expansion space for future integration of more types of distributed power sources. In harsh fault condition tests, such as three-phase voltage dips in the power grid, the PSModel black-box co-simulation model built in this invention can accurately reproduce the fault ride-through physical characteristics of each physical renewable energy device and more accurately reflect the interactive response mechanism of the wind power-photovoltaic-energy storage-SVG combined system during transient processes. Its output dynamic response curve perfectly matches the original high-precision model provided by the manufacturer. This modeling method fully verifies its absolute accuracy, excellent compatibility, and technological advancement in complex co-simulation scenarios. It can effectively and accurately support in-depth stability analysis and wide-area control strategy verification of high-proportion renewable energy combined power systems under complex conditions, providing indispensable and solid technical support for the construction of new power systems.

[0019] Additional aspects and advantages of the invention will become apparent in the following description or may be learned by practice of the invention. Attached Figure Description

[0020] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a flowchart of a modeling method for a wind power-photovoltaic-energy storage-SVG co-simulation system based on PSModel, according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a doubly-fed asynchronous wind power generation system in an embodiment of the present invention; Figure 3 This is a schematic diagram of the grid-side converter control strategy in an embodiment of the present invention; Figure 4 This is a schematic diagram of the rotor-side converter control strategy in an embodiment of the present invention; Figure 5 This is a schematic diagram of the topology of a three-phase photovoltaic grid-connected inverter in an embodiment of the present invention; Figure 6 This is a schematic diagram of the three-phase inverter control in an embodiment of the present invention; Figure 7 This is a schematic diagram of a typical energy storage power station in an embodiment of the present invention; Figure 8 This is a schematic diagram of a multi-scenario dynamic simulation model of an energy storage power station in an embodiment of the present invention; Figure 9 This is a schematic diagram of a battery pack model in an embodiment of the present invention; Figure 10 This is a schematic diagram of the equivalent circuit for STATCOM access to the power grid in an embodiment of the present invention; Figure 11 This is a schematic diagram of the SVG power unit connection in an embodiment of the present invention; Figure 12 This is a schematic diagram of a unified current control system, including a STATCOM controller and a main circuit, in an embodiment of the present invention. Figure 13 This is a schematic diagram of modular modeling of a wind power generation system in an embodiment of the present invention; Figure 14 This is a schematic diagram of the modeling topology of the wind power-photovoltaic-energy storage-SVG co-simulation system in an embodiment of the present invention; Figure 15 This is a schematic diagram comparing the phase A voltage of the PSModel black box model and the manufacturer's model in an embodiment of the present invention; Figure 16 This is a schematic diagram comparing the B-phase voltage of the PSModel black box model and the manufacturer's model in an embodiment of the present invention; Figure 17 This is a schematic diagram comparing the C-phase voltage of the PSModel black box model and the manufacturer's model in an embodiment of the present invention. Detailed Implementation

[0021] 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, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0022] 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.

[0023] The following reference Figures 1 to 17 This paper describes a modeling method for a wind power-photovoltaic-energy storage-SVG co-simulation system based on PSModel, provided by some embodiments of the present invention.

[0024] Some embodiments of this application provide a modeling method for a wind power-photovoltaic-energy storage-SVG co-simulation system based on PSModel.

[0025] like Figure 1 As shown, the first embodiment of the present invention proposes a modeling method for a joint simulation system of wind power-photovoltaic-energy storage-SVG based on PSModel, including the following steps S1 to S3.

[0026] S1. Analyze the control characteristics of wind turbines, photovoltaic power generation devices, energy storage power stations, and static synchronous compensators respectively. Establish independent system models for each device, including the electrical main circuit topology and its corresponding control strategy, to characterize the electrical dynamic response characteristics of each device.

[0027] In some embodiments, the wind turbine is modeled using a currently mainstream doubly-fed induction generator (DFIG) on the market. The rotor of this DFIG's wound-rotor induction generator is connected to the grid via back-to-back power inverters employing insulated-gate bipolar transistors (IGBTs). These power inverters can simultaneously control the amplitude and frequency of the rotor current. For example... Figure 2 In the schematic diagram of the doubly fed asynchronous wind power generation system shown, since the stator and rotor windings of the asynchronous generator are connected to the power grid and participate in the energy conversion process, the entire system can be decomposed into modules such as wind turbine, phase-locked loop, shaft system, doubly fed asynchronous motor, grid-side filter, dual PWM converter and interface line.

[0028] The rotor of a doubly-fed induction generator (DFIG) is connected to the grid via a back-to-back converter. Since grid voltage is readily available and relatively stable, grid-side converters (GSCs) typically employ grid voltage-oriented vector control. Figure 3 The diagram shows the structure of a dual closed-loop control system for the grid-side converter. The d-axis uses constant voltage control, and the q-axis uses constant current control (typically, the q-axis current is set to 0). Taking the d-axis as an example, the given DC voltage and the measured DC voltage U... dc After comparison, the current enters the PI regulator, outputting a reference value for the d-axis current. This value is then compared with the measured d-side current value, and the current enters the PI regulator again, outputting a d-axis reference voltage. This voltage is then used to compensate for cross-coupling caused by the d-q axis current and the measured d-axis voltage value u. d The d-axis reference voltage is obtained. The q-axis control is similar, but decoupled from the d-axis control. Finally, the d-axis and q-axis voltage reference values ​​enter the modulation stage to output the drive signal of the converter valve, thereby realizing the control of the grid-side converter output voltage.

[0029] The rotor-side converter also employs dual closed-loop PI control, adjusting the rotor current to control the active and reactive power output of the doubly-fed wind turbine. The control strategy of the rotor-side converter is as follows: Figure 4 As shown. The active power reference value on the rotor side is generally the mechanical power input to the generator shaft of the prime mover, while the reactive power reference value is adjusted according to the grid demand. The inner current loop uses the output of the outer power loop as a setpoint to control the rotor current. Similarly, the cross-coupling term caused by the dq axis current also needs to be considered. Finally, the voltage command is sent to the modulation module to precisely control the output voltage of the rotor-side converter.

[0030] In some embodiments, for photovoltaic power generation devices, the stand-alone system model primarily revolves around a three-phase photovoltaic grid-connected inverter. For example... Figure 5 As shown, this inverter adopts a voltage-source three-phase bridge inverter structure, which has advantages such as simple circuit topology, easy control, and low power switching voltage stress. Its three-phase inverter control strategy is as follows: Figure 6 As shown, the d-axis in the outer loop control is controlled by DC voltage, the purpose of which is to achieve maximum power point tracking of the photovoltaic array. ref u is the reference voltage for the maximum power output of the photovoltaic power source. dc The output voltage of the photovoltaic power source is the actual value; the q-axis represents the reactive power control of the photovoltaic output, providing a reference current for the inner-loop q-axis current control. The inner-loop current control also uses PI control, and incorporates the cross-term effect caused by the dq-axis current to ensure stable operation of the grid-connected inverter.

[0031] In some embodiments, for energy storage power stations, the core of dynamic simulation modeling of energy storage power stations in a large power grid is to study their input and output power characteristics, such as... Figure 7As shown, its typical grid connection structure consists of three parts: a battery pack, a converter (PCS), and a control system. The battery pack is composed of a large number of individual cells. Under the action of the control strategy, the converter converts DC to AC to achieve synchronous grid connection. Generally, dq-axis decoupling control is used to achieve independent control of active and reactive power.

[0032] The principles of dynamic simulation modeling for energy storage power stations in large power grids: When an energy storage power station is connected to the grid, from the grid's perspective, only its external characteristics are typically considered, treating the energy storage power station as a power source. The core of dynamic simulation modeling is to study its input and output power characteristics. Under normal operating conditions, the energy storage power station has excellent four-quadrant operation capabilities, and its output power can be arbitrarily adjusted, adapting to simulations of different grid scenarios. In terms of active power, it can execute plant-level active power control commands and quickly respond to frequency regulation needs, such as participating in the primary frequency regulation and AGC secondary frequency regulation processes of the system. In terms of reactive power, it can execute plant-level reactive power control commands and quickly respond to the system voltage regulation needs. At this time, the response characteristics of the energy storage power station are mainly determined by the PCS control strategy.

[0033] When a battery pack operates under charging / discharging conditions for an extended period, its charge may reach a fully charged / fully discharged state. At this point, energy flow on the battery side of the energy storage station is obstructed, preventing it from responding to power commands bidirectionally. This characteristic of energy storage stations can be simulated by constructing the state of charge (SOC) of the battery itself.

[0034] When the power command is too high or the system voltage is abnormal, the converter may experience overcurrent, making it difficult for the output power to fully track the power command or control signal. Therefore, a current-limiting model needs to be considered. In particular, when the voltage is too low or too high, the energy storage power station will enter fault ride-through control mode. At this time, the power command will change fundamentally, and the converter's power output will be mainly determined by the fault ride-through control strategy.

[0035] For electrical components in energy storage, the main components of an energy storage PCS are inverters and filters. Among them, LCL filters have gradually replaced traditional L-type filters in practical applications due to their excellent attenuation of high-order harmonics and performance advantages at low switching frequencies. The design principle of filter parameters should be to achieve the best filtering effect while saving inductor core material. At the same time, it is essential to ensure that the resonant frequency is not too low, so as not to place excessive demands on the design of the current controller. The parameters of the LCL filter are selected as follows: a) The voltage drop generated by the inductive impedance of the LCL filter is less than 10% of the voltage under the rated operating conditions of the power grid.

[0036] b) The resonant frequency of the filter should be 10 times higher than the mains frequency, but less than half the switching frequency.

[0037] c) In order to keep the power factor of the grid-connected inverter at a level that is not too low, the reactive power absorbed by the filter capacitor should be less than 5% of the rated active power of the system.

[0038] In one specific embodiment, taking into account the operating conditions, a system was constructed as follows: Figure 8 The multi-scenario dynamic simulation control model shown includes a plant-level control model for simulating the grid-connected regulation function of an energy storage power station, comprising active power control, reactive power control, inertia support, primary frequency regulation, and AGC secondary frequency regulation function models; in the figure, P... branch For tie line power, P set Q set U is the command value for the plant / station. tset This is the AC voltage command value; U t U is the AC voltage, f is the system frequency, and U is the AC voltage. dc I dc This refers to the battery terminal voltage and current. The power control command P generated by the plant-level control model... ref Q ref Current command I is generated by the normal control model and the fault ride-through control model. drefp I qrefp The current limiting model adjusts the current command I based on the current battery pack's SOC state. drefp I qrefp After limiting, a dq-axis injected current I is generated. dref I qref Finally, the grid connection interface model is converted into a dq-axis V model. dref V qref Injected into the AC power grid. Among them, the long-term charging and discharging process of the battery pack and the function of energy storage participating in secondary frequency regulation are medium- and long-term dynamic processes (minutes and above), while other models are electromechanical transient simulation processes, so they can be excluded from the electromagnetic transient time domain.

[0039] A grid-connected energy storage power station with a battery pack model has a large power output. The battery pack consists of hundreds of individual cells connected in series and parallel. A typical topology is as follows: Figure 9 As shown in the figure. N se N represents the number of individual cells connected in series in one branch of the battery pack. sh This indicates the number of parallel branches in the battery pack. For individual cell models, when applied to grid-connected dynamic simulation analysis, the internal electrochemical reaction processes are generally ignored, and only the external output characteristics of the battery are considered. Taking into account computational efficiency, simulation accuracy, and engineering application requirements, Rint is selected to simulate the external characteristics of individual cells. Figure 9 The area within the dashed box represents a single battery cell model, where: E b R is the internal potential of a single cell, close to its resting voltage;b is the internal resistance of the battery and is related to the SOC.

[0040] In fact, since the battery pack is directly (or via a DC / DC converter) connected to the inverter for grid connection, as long as the battery pack operates within the normal operating range, for example, the state of charge satisfies 0.2 < SOC < 0.8, the DC-side voltage U of the inverter can be maintained dc basically unchanged. Then, from the perspective of the power grid, the power external characteristics of the energy storage power station are mainly determined by the power control strategy.

[0041] In some embodiments, for a static synchronous compensator, i.e., SVG or STATCOM, which has a voltage source converter as the core, a DC capacitor is used as the energy storage element on the DC side, and it is connected in parallel to the system through a connecting reactor or a coupling transformer.

[0042] Compared with traditional reactive power compensation devices, STATCOM has the advantages of continuous regulation, low harmonics, low losses, wide operating range, high reliability, fast regulation speed, etc. Since its appearance, it has received extensive attention and developed rapidly. The equivalent circuit of STATCOM connected to the power grid is as Figure 10 shown.

[0043] STATCOM has a voltage source converter as the core, a DC capacitor is used as the energy storage element on the DC side, and relies on the VSC to convert the DC voltage into an AC voltage with the same frequency as the power grid, and is connected in parallel to the system through a connecting reactor or a coupling transformer. Usually, the AC output voltage Us of the VSC is in the same phase as the power grid voltage Ug. If Us is greater than Ug, then STATCOM sends reactive power to the power grid at this time; if Us is less than Ug, then STATCOM absorbs reactive power from the power grid at this time. The DC-side capacitor of STATCOM only plays a role in voltage support, so the capacity of the AC capacitor in SVC is much smaller. Currently, the main circuit of STATCOM generally adopts a cascaded H-bridge chain structure, which is the main development direction of current technology. The topological structure is as Figure 11 shown. This topological structure adopts a cascaded composition method, with mature technology, easy to modularize, good redundancy, can avoid using a multiple transformer and directly obtain a very high AC output voltage and excellent harmonic characteristics, reduce the floor area, and lower the cost. The unified current control block diagram equivalent to the controller and the main circuit of STATCOM is as Figure 12 shown. Through precise control strategies, SVG can improve the power system function in aspects such as dynamic voltage control and transient stability, and play a role in continuous regulation and fast response.

[0044] S2. Extract the running parameters and algorithm features of each independent system model, and construct PSModel dynamic link libraries that encapsulate the corresponding system model code. Each dynamic link library is configured with an initialization stage interface for allocating memory, a simulation calculation stage interface for executing transient logic, and a termination stage interface for releasing resources. This transforms the independent system models of each device into black-box models decoupled from the external calling environment.

[0045] After completing the analysis and basic model building steps in step S1, step S2 is executed. This involves constructing a PSModel encapsulated dynamic link library based on the characteristics of the independent system model, forming a complete PSModel black-box model structure. This achieves complete decoupling between the internal control model and the external simulation environment, facilitating rapid integration and invocation in large-scale networks. Specifically, the constructed PSModel black-box model structure includes external electrical circuits and an encapsulated dynamic link library. The encapsulated dynamic link library's execution mechanism is strictly divided into three standard execution phases: initialization, calculation, and termination. During the encapsulation process, a special identifier is used to standardize model naming. This identifier consists of letters, numbers, and underscores; for example, a suffix can be formed using simple manufacturer characters, capacity, and model number. Corresponding to the three different execution phases, the dynamic link library contains three standard interface encapsulation functions: an initialization function called during the initialization phase, a calculation function called during the calculation phase, and a termination function called during the termination phase. By building the external electrical system part of the model in the PSModel platform according to the actual grid topology and electrical parameters, and then loading the corresponding dynamic link library according to the black-box unified interface constructed above, the overall call and joint simulation of the internal complex control system part can be successfully completed.

[0046] The initialization phase of a dynamic link library (DLL) primarily involves executing initialization operations for the corresponding functional modules, such as assigning initial values ​​to internal variables and requesting the necessary memory space from the system. Users need to encapsulate all the physical processes and algorithmic logic related to the initialization of the corresponding control system model into the interface function of this phase. This initialization interface function mainly covers creating model instances, checking the passed-in model parameters, and receiving storage addresses allocated by the external platform. After execution, it returns information indicating whether initialization was successful. A return value of zero indicates successful initialization, while a non-zero return value indicates an error corresponding to a specific error number. The interface function in this initialization phase receives multiple standard parameter inputs, primarily including a model instance pointer used to identify multiple different calls to the same model (the user-encapsulated model needs to specify a specific instance address for this pointer during initialization); the total number of model parameters passed from the simulation program to the user-encapsulated model, along with the corresponding model parameter array; and the total number of model input variables and their array, as well as the total number of model output variables and their array passed from the encapsulated model to the simulation program, all determined during initialization. In addition, the initialization interface also receives an identifier variable indicating whether to start from snapshot mode, where a value of one indicates starting from snapshot mode and a value of zero indicates not starting from snapshot mode; it also receives the address of a common storage area passed by the simulation program side, which can be used for user data storage or as a storage area for the snapshot function. Note that its size should be set according to actual needs to avoid out-of-bounds access; it also receives the current simulation step number, which is initially zero at startup, and the model simulation calculation step size in seconds.

[0047] After entering the simulation calculation phase, the dynamic link library mainly executes the core functions of the functional modules, covering transient control processes such as numerical calculation, time delay handling, and complex logic judgments. Users must completely encapsulate the core calculation and solution processes of each new energy device system model into the interface function of this calculation phase. Similar to the initialization function, this calculation interface function also returns information on the success or failure of the simulation calculation; a return value of zero indicates success, while a non-zero value indicates the specific error number. Within each simulation step, this calculation phase interface function is called cyclically by the simulation platform. In addition to the instance pointer, input array, output array, and common memory address mentioned above for identifying model calls, the parameters passed to it also include a system-reserved flag parameter and a control flag indicating whether a snapshot needs to be generated in the current simulation phase. During this phase, as the transient simulation progresses, the current simulation step number parameter passed from the system is automatically incremented by one for each subsequent simulation step, starting from zero, to maintain the time synchronization of the model's internal state calculations.

[0048] When the entire simulation test process ends or an external interrupt command is received, the dynamic link library will enter the termination phase. This phase mainly performs the termination operations of the corresponding functional modules, such as printing the simulation run log and releasing the system memory previously allocated. Users need to encapsulate the corresponding process and cleanup mechanism for completely releasing memory into the interface function of this phase. Although the interface function of this termination phase is defined as a type with no return value, its internal execution logic must still ensure that it returns information on whether the memory release was successful. If successful, the system recognizes it as zero; if unsuccessful, it is recognized as a specific non-zero error number for the underlying layer to perform safety monitoring. The input parameters required for the interface function of this termination phase are extremely concise. It only needs to receive the model instance pointer specified in the initialization phase to identify the specific calling object. Through this unique pointer, the underlying program can accurately locate and safely unload all computer resources occupied by the corresponding instantiated black-box model.

[0049] In one specific embodiment, step S2 is performed as follows: The constructed PSModel black-box model structure includes external electrical circuits and encapsulated dynamic link libraries, where the encapsulated dynamic libraries consist of three stages: Initialization phase: Perform initialization operations on the functional modules, such as assigning initial values, allocating memory, etc.

[0050] Calculation phase: This phase involves executing the specific functions of the functional modules, such as calculation, delay, and logical judgment.

[0051] Termination phase: Perform termination operations on the functional modules, such as logging, releasing memory, etc.

[0052] The “$MODEL” below is a special identifier for a model, consisting of letters, numbers, and underscores “_”. It can be composed of simple manufacturer information, capacity, model number, etc.

[0053] The three interface wrapper functions corresponding to the different stages mentioned above are, in order: Init$MODEL(), Step$MODEL(), and Terminate$MODEL(). The Init$MODEL() function will be called and executed during the initialization stage of program execution, the Step$MODEL() function will be called and executed during the calculation stage, and the Terminate$MODEL() function will be called and executed during the termination stage.

[0054] Initialization phase: Users need to encapsulate the initialization process into this stage. The functions in the initialization stage mainly include: creating instances, checking model parameters, and passing in storage addresses. The functions return information indicating whether initialization was successful. Success: 0; Failure: a non-zero value representing an error number.

[0055] The interface functions for the initialization phase are: intInit$MODEL(void pModelInfo,intParameter_N,intNin,intNout,double Parameter, double Val_In, double Val_out, intSnapshot_Inp, char Store_Add,doubleTnow,doubleDT) $MODEL: The name of the encapsulation model. Different encapsulation models with different functions must have different names.

[0056] pModelInfo: <out>A model instance pointer is used to identify multiple different calls to the same model. During initialization, the user needs to encapsulate the model and assign it to... pModelInfo specifies the address of a model instance.

[0057] Parameter_N: <in>The total number of model parameters can be 0. During initialization, these parameters are passed from the simulation program to the user-encapsulated model.

[0058] Nin: <in>The total number of input variables for the model. During initialization, this number is passed from the simulation program to the user-encapsulated model.

[0059] Nout: <in>The total number of output variables in the model. This number is passed from the simulation program to the user-encapsulated model during initialization.

[0060] Parameter: <in>The model parameter array has dimension Parameter_N. During initialization, it is passed from the simulation program to the user-encapsulated model.

[0061] Val_In: <in>The model input array has a dimension of Nin, which is determined during initialization. During initialization, it is passed from the simulation program to the user-encapsulated model.

[0062] Val_Out: <out>The model output array has dimensions NOut, which are determined during initialization. During initialization, it is passed from the user-encapsulated model to the simulation program.

[0063] Snapshot_Inp: <in>Whether to start from Snapshot mode. 1 indicates starting from Snapshot, 0 indicates not starting from Snapshot. This is passed from the simulation program to the user-packaged model during initialization.

[0064] Store_Add: <in>Public storage area. During initialization, this is transferred from the simulation program to the user-defined model. This public storage area can be used for user data storage or as a storage area for snapshots. Users can set the size of Store_Add according to their needs to avoid cross-access.

[0065] Tnow: <in>Simulation steps. This number is passed from the simulation program to the user-encapsulated model at startup and is 0.

[0066] DT: <in>The simulation step size is measured in seconds (s). During initialization, it is passed from the simulation program to the user-encapsulated model.

[0067] Simulation calculation stage: Users need to encapsulate the corresponding calculation process into this stage. The function returns information indicating whether the simulation calculation was successful: success – 0; failure – a non-zero value representing the error number.

[0068] The interface functions for the computation phase are: intStep$MODEL(void pModelInfo, intMark, double Val_In, double Val_Out, intSnapshot_Inp, char Store_Add, doubleTnow) $MODEL: The name of the encapsulation model. Different encapsulation models with different functions must have different names.

[0069] pModelInfo: <out>A model instance pointer is used to identify multiple different calls to the same model. During initialization, the user needs to encapsulate the model and assign it to... pModelInfo specifies the address of a model instance.

[0070] Mark: Reserved, not used.

[0071] Val_In: <in>The model input array has a dimension of Nin, which is determined during initialization. During initialization, it is passed from the simulation program to the user-encapsulated model.

[0072] Val_Out: <out>The model output array has dimensions NOut, which are determined during initialization. During initialization, it is passed from the user-encapsulated model to the simulation program.

[0073] Snapshot_Inp: <in>Should a snapshot be started? During the simulation phase, the simulation program passes the information to the user-packaged model.

[0074] Store_Add: <in>Common storage area. During initialization, this is transferred from the simulation program to the user-defined model. The common storage area can be used for user data storage and also as the storage area for snapshots in the snapshot function.

[0075] Tnow: <in>Simulation steps. Initializes to 0, increments by 1 for each simulation step.

[0076] Simulation ended: Users need to encapsulate the memory release process within this stage. The function returns information indicating whether the memory release was successful. Success – 0; Failure – a non-zero value representing an error number.

[0077] The interface function for the final stage is: voidTerminate$MODEL(void pModelInfo) $MODEL: The name of the encapsulation model. Different encapsulation models with different functions must have different names.

[0078] pModelInfo: <out>A model instance pointer is used to identify multiple different calls to the same model. During initialization, the user needs to encapsulate the model and assign it to... pModelInfo specifies the address of a model instance.

[0079] The electrical system part of the model was built in the PSModel platform according to the topology and electrical parameters. Then, the dynamic link library was called according to the constructed PSModel black box unified interface to complete the call of the control system part.

[0080] S3. In the PSModel simulation platform, a modular layered architecture is adopted to instantiate the black box model of each device into independent sub-modules with high cohesion and low coupling. Based on the actual physical topology of the wind-solar-storage grid connection, the external electrical connection relationship and interaction node between each independent sub-module are configured in the simulation platform to assemble a wind power-photovoltaic-energy storage-SVG joint simulation system model.

[0081] Specifically, after completing the dynamic link library encapsulation corresponding to step S2 above, the method of this invention executes step S3, which involves dividing the complex system into independent, highly cohesive, and loosely coupled sub-modules based on modular design principles, defining the connection relationships between modules, simplifying the system structure with a hierarchical architecture, and building a joint grid-connected system model including wind power, photovoltaics, energy storage, and SVG. As a key methodology for addressing the high proportion of new energy sources integrated into new power systems, modular design refers to dividing the entire dynamic system into relatively independent small modules according to certain principles, and specifying the inputs, outputs, states, and relationships between variables of each module. Each module consists of algebraic equations, differential equations, or difference equations, and is itself a basic dynamic system. The topological connections between modules symbolize the interrelationships within this complex system; all modules and connections together construct a complete dynamic system. The models in these modules are derived from actual production, and are obtained through abstract generalization, assumption derivation, and repeated verification of actual production systems, maintaining a high degree of consistency with the actual system.

[0082] The core concept of modular design is high cohesion and low coupling, aiming to simplify the structure of power systems, enhance the reusability of modular units, and support flexible system expansion. For example... Figure 13 The schematic diagram of modular modeling for a wind power generation system illustrates how, for complex power systems composed of multiple subsystems, subsystem modules can be combined and encapsulated. By combining existing modules, the overall power system model maintains a hierarchical structure. Specifically, generators, motors, converters, and control systems are combined and encapsulated into a unit-level model, then further combined and encapsulated into a wind farm-level model, and finally connected to the regional power grid model.

[0083] In the PSModel simulation platform, the external electrical connections and interaction nodes between each independent submodule are configured according to the actual physical topology of the wind-solar-storage integrated grid connection. To achieve effective interaction of the black-box model under electromagnetic transient conditions, this embodiment establishes a multi-port Norton equivalent controlled source interface in the simulation platform. Each independent submodule transmits its internally calculated historical injection current and equivalent admittance to the interaction node at each simulation step. The simulation platform combines the node admittance matrix of the entire system to solve the network equations, obtains the terminal voltage of each interaction node, and then feeds it back as an input variable to the simulation calculation stage interface of each independent submodule. At the same time, the wind-solar-storage-SVG joint simulation system model also includes an upper-level coordination controller to collect the bus voltage and frequency deviation of the interaction nodes. When a voltage sag fault occurs in the grid, the controller prioritizes triggering the static synchronous compensator to output reactive current for transient voltage support; if the voltage drop exceeds a set threshold, it further triggers the wind turbine and photovoltaic power generation device to enter the low voltage ride-through mode, and simultaneously controls the converter of the energy storage power station to redistribute active and reactive power commands according to the set reactive power priority strategy. Through the aforementioned low-level interface interaction and high-level coordination mechanism, a system such as [example system name] was successfully built in the PSModel platform. Figure 14 The modeling topology of the wind power-photovoltaic-energy storage-SVG co-simulation system is shown.

[0084] To verify the advancement and accuracy of the PSModel-based wind power-photovoltaic-energy storage-SVG co-simulation system modeling method of this invention, a corresponding co-grid-connected simulation system was built for testing. The initial operating state of the wind turbine in the system was set as 16MW active power and 0MVar reactive power; the photovoltaic device operated in maximum power point tracking mode; the energy storage system's state of charge was maintained between 0.5 and 0.6; and the SVG device was set to dynamic reactive power compensation mode. During the simulation, a three-phase symmetrical voltage drop fault was triggered at 4s, with the power supply voltage dropping from the rated value of 1.0pu to 0.2pu, and the fault lasting for 0.625s. The dynamic response data of the ABC three-phase voltages at the wind turbine terminals in the co-simulation system were recorded simultaneously. The PSModel black-box co-simulation model constructed by this invention was compared with the native model provided by new energy manufacturers in actual engineering projects. Figure 15 The diagram showing the voltage comparison of phase A at the generator terminals is shown below. Figure 16 The diagram showing the voltage comparison of phase B at the generator terminals and Figure 17 As shown in the comparison diagram of the C-phase voltage at the turbine terminals, the dynamic response curves of the three-phase voltage at the turbine terminals of the PSModel black-box co-simulation model established in this invention perfectly match the manufacturer's original model under grid fault conditions. This not only accurately reproduces the fault ride-through characteristics of various new energy devices, but also accurately reflects the interactive response process of the wind power-photovoltaic-energy storage-SVG combined system. It fully verifies the accuracy, compatibility, and advancement of the modeling method of this invention in co-simulation scenarios, and can effectively support the stability analysis and control strategy verification of high-proportion new energy combined power systems.

[0085] In this specification, the illustrative expressions of the terms used do not necessarily refer to the same embodiments or examples. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0086] Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this invention shall be included within the scope of protection of this invention.< / out> < / in> < / in> < / in> < / out> < / in> < / out> < / in> < / in> < / in> < / in> < / out> < / in> < / in> < / in> < / in> < / in> < / out>

Claims

1. A modeling method for a wind power-photovoltaic-energy storage-SVG co-simulation system based on PSModel, characterized in that, include: The control characteristics of wind turbines, photovoltaic power generation devices, energy storage power stations, and static synchronous compensators are analyzed separately. Independent system models containing the electrical main circuit topology and corresponding control strategies are established for each device to characterize the electrical dynamic response characteristics of each device. The running parameters and algorithm features of each independent system model are extracted, and PSModel dynamic link libraries encapsulating the corresponding system model code are constructed respectively. Each dynamic link library is configured with an initialization stage interface for allocating memory, a simulation calculation stage interface for executing transient logic, and a termination stage interface for releasing resources. In this way, the independent system models of each device are transformed into black box models decoupled from the external calling environment. The PSModel simulation platform adopts a modular and layered architecture, instantiating the black-box models of each device into independent sub-modules with high cohesion and low coupling. Based on the actual physical topology of the wind-solar-storage grid connection, the external electrical connection relationships and interaction nodes between the independent sub-modules are configured in the simulation platform to assemble a wind power-solar-storage-SVG joint simulation system model.

2. The modeling method for a wind power-photovoltaic-energy storage-SVG co-simulation system based on PSModel according to claim 1, characterized in that, The wind turbine adopts a doubly-fed asynchronous wind turbine model. Its system model includes a wind turbine, a phase-locked loop, a shaft system, a doubly-fed asynchronous motor, a grid-side filter, and a dual PWM converter. The dual PWM converter includes a grid-side converter and a rotor-side converter. The grid-side converter adopts grid voltage-oriented vector dual closed-loop control, with constant DC voltage control on the d-axis and constant current control on the q-axis. The rotor-side converter controls the active and reactive power output of the wind turbine by adjusting the rotor current.

3. The modeling method for a wind power-photovoltaic-energy storage-SVG co-simulation system based on PSModel according to claim 1, characterized in that, The model of the photovoltaic power generation device includes a three-phase photovoltaic grid-connected inverter with a voltage-source three-phase bridge inverter structure. Its control strategy includes an outer loop d-axis DC voltage control and q-axis reactive power control to achieve maximum power point tracking of the photovoltaic array, and an inner loop PI current control that considers the influence of dq-axis cross-coupling terms.

4. The modeling method for a wind power-photovoltaic-energy storage-SVG co-simulation system based on PSModel according to claim 1, characterized in that, The energy storage power station model includes an external characteristic model of a battery pack composed of individual cells connected in series and parallel, a converter with an LCL filter, and a multi-scenario dynamic simulation control model. The external characteristic model of the battery pack adopts an internal potential and internal resistance structure of individual cells that considers the state of charge. The multi-scenario dynamic simulation control model includes a plant-level control model for responding to grid-connected regulation, a normal operation control model, a fault ride-through control model, and a current limiting model for preventing converter overcurrent.

5. The modeling method for a wind power-photovoltaic-energy storage-SVG co-simulation system based on PSModel according to claim 1, characterized in that, The static synchronous compensator model is based on a voltage source converter with an H-bridge cascade structure. The DC side uses DC capacitors as energy storage support elements. The voltage source converter converts the DC side voltage into an AC voltage with the same frequency as the power grid. By comparing the phase and amplitude with the AC voltage at the grid connection point, the continuous generation or absorption of reactive power can be controlled.

6. The modeling method for a wind power-photovoltaic-energy storage-SVG co-simulation system based on PSModel according to claim 1, characterized in that, The initialization phase interface of the PSModel dynamic link library allocates memory space for the model instance pointer by passing in the total number of model parameters, the total number of input and output variables, and the simulation step size; The simulation calculation phase interface is called within each calculation step to read data from the external public storage area and perform calculations for the model functions; the termination phase interface is used to perform memory release operations when the simulation terminates.

7. The modeling method for a wind power-photovoltaic-energy storage-SVG co-simulation system based on PSModel according to claim 1, characterized in that, The modular and layered architecture adopted in the PSModel simulation platform includes: combining and encapsulating generators, motors, converters and control systems into unit-level models, further combining and encapsulating multiple unit-level models into wind power plant-level models, and finally connecting multiple plant-level models to the regional power grid model to form a unified and hierarchical dynamic system topology.

8. The modeling method for a wind power-photovoltaic-energy storage-SVG co-simulation system based on PSModel according to claim 4, characterized in that, The parameters of the LCL filter of the converter are selected to meet the following conditions: the voltage drop generated by the inductive impedance is less than 10% of the voltage under the rated operating conditions of the power grid; the resonant frequency of the filter is 10 times higher than the power grid frequency and less than half of the switching frequency; and the reactive power absorbed by the filter capacitor is less than 5% of the rated active power of the system.

9. The modeling method for a wind power-photovoltaic-energy storage-SVG co-simulation system based on PSModel according to claim 4, characterized in that, In the multi-scenario dynamic simulation control model, the power control command generated by the plant-level control model is used to generate a current command through the normal operation control model and the fault ride-through control model. The current limiting model limits the current command according to the current state of charge of the battery pack to form a dq-axis injected current. Finally, the grid-connected interface model converts the dq-axis injected current into a dq-axis voltage command and injects it into the AC grid.

10. The modeling method for a wind power-photovoltaic-energy storage-SVG co-simulation system based on PSModel according to claim 1, characterized in that, The interface functions of the PSModel dynamic link library also include a snapshot state input flag variable. When the flag variable is true, the initialization phase interface directly reads the historical cross-sectional data saved at the previous steady state moment from the external public storage area as the initial value of the model's internal state variables, so as to achieve a smooth start-up of the co-simulation system.

Images