A real-time simulation method for photovoltaic power generation system based on FPGA

Abstract

The application discloses a photovoltaic power generation system real-time simulation method based on FPGA, which comprises the following steps: 1) FPGA modeling of the photovoltaic electrical system: establishing a photovoltaic component output characteristic equation suitable for the FPGA architecture; using a numerical integration method to perform difference on passive elements in the photovoltaic electrical system to obtain an equivalent circuit model in the form of an equivalent calculation admittance and a historical item current source in parallel; 2) FPGA modeling of the photovoltaic control system based on different photovoltaic control strategies. The simulation method has high hardware parallelism and time parallelism, high simulation precision, fast simulation speed, flexible semi-custom integrated circuit design, complete programmable and reconfigurable capability, and reduced simulation cost.

Description

Technical Field

[0001] This invention relates to the field of distributed power system simulation technology, and in particular to a real-time simulation method for photovoltaic power generation systems based on FPGA. Background Technology

[0002] With the large-scale integration of distributed generation into active power distribution networks, electrical equipment models are becoming increasingly complex. The nonlinear characteristics and dynamic properties of electrical equipment at different time scales are intertwined, increasing the difficulty of analysis and placing higher demands on the transient process modeling and real-time simulation of microgrids. To ensure the safe and stable operation of new power systems, real-time simulation research on distributed generation and power system architecture has received increasing attention.

[0003] Currently, the underlying computing hardware of mainstream commercial real-time simulators mainly consists of DSPs and CPUs. In essence, these are still fully custom circuits, and the processors operate serially internally. The computational speed and small step size of serial operation are limited in real-time simulation, making it difficult to meet the real-time simulation requirements of microsecond-level step sizes for high-frequency power electronic equipment. Consequently, their computing speed and simulation accuracy are limited.

[0004] FPGAs, with their high hardware parallelism, memory distribution, and pipelined architecture, can perform detailed simulations of active power distribution networks, improve the efficiency of power system and power electronics research, and enable semi-physical testing of power systems or equipment. Currently, FPGAs are showing great potential in the transient real-time simulation of distributed generation systems. Summary of the Invention

[0005] The purpose of this invention is to provide a real-time simulation method for photovoltaic power generation systems based on FPGA.

[0006] Therefore, the technical solution of the present invention is as follows:

[0007] A real-time simulation method for a photovoltaic power generation system based on FPGA, including...

[0008] 1) Perform FPGA modeling of the photovoltaic electrical system:

[0009] Establish output characteristic equations for photovoltaic modules suitable for FPGA architecture;

[0010] The passive components in the photovoltaic electrical system are differentially analyzed using the numerical integration method to obtain an equivalent circuit model in the form of the equivalent calculated admittance and the parallel connection of the historical term current source;

[0011] 2) FPGA modeling of photovoltaic control systems based on different photovoltaic control strategies.

[0012] Furthermore, the simulation method adopts a global control hierarchy architecture: a global timing controller FSM0 is designed at the top level of the FPGA to precisely control the orderly solution of the electrical system and the control system. At the same time, each module and processing link is controlled by its subsystem timing controller.

[0013] Furthermore, this simulation method employs an interactive solution design for both the electrical and control systems, specifically including:

[0014] Once the electrical system solution for this time step is completed, the collected voltage and current parameters are stored in RAM_E, triggered by the storage end signal. When the read signal is triggered, the control system immediately reads the data from RAM_E and performs the control process solution. Then, the switch state obtained after the solution is completed is stored in RAM_C. When the clock signal for the next time step arrives, the electrical system immediately reads the data from RAM_C and performs the solution for the next time step.

[0015] Furthermore, the output characteristic equation of the photovoltaic module suitable for FPGA architecture is as follows:

[0016] (1)

[0017] in,

[0018] (2)

[0019] in, and These are the output voltage and output current of the photovoltaic cell, respectively. , , , These are the short-circuit current, open-circuit voltage, maximum power point voltage, and maximum power point current of the photovoltaic cell, which are usually provided by the manufacturer.

[0020] Furthermore, considering variations in light and temperature, parameters need to be modified to describe the new photovoltaic output characteristic equation. Under non-standard operating conditions, the four parameters provided by the manufacturer can be modified as follows:

[0021] (3)

[0022] in, This represents the difference in relative light intensity between normal and standard operating conditions. The relative temperature difference between normal and standard operating conditions; temperature coefficient of current. Voltage temperature coefficient ; .

[0023] Furthermore, the intermediate variable , , and multiple sets of non-standard operating condition parameters , , , The value is obtained in advance during the compilation process on the host computer, stored in a register, and directly called by the FPGA pipeline.

[0024] Furthermore, the continuous control system is transformed into an explicit solution for each control element, wherein,

[0025] The PI element of the photovoltaic control system is discretized using Euler's formula. The specific processing method is as follows:

[0026] (7)

[0027] in, ; and These are the proportional coefficient and the integral time coefficient, respectively.

[0028] The processing method for the limiting stage of the photovoltaic control system is as follows:

[0029] (8)

[0030] Where y1(t) is the input of the limiting circuit, y2(t) is the output of the limiting circuit, b is the upper limit of the output, and c is the lower limit of the output.

[0031] Furthermore, the PWM modulation stage of the photovoltaic control uses a signal generator design that reads data stored in the ROM of the FPGA, so that the address seeker searches for the ROM address in a pipeline manner, thereby forming a periodic carrier signal.

[0032] Compared with existing technologies, this FPGA-based real-time photovoltaic simulation method boasts high hardware and time parallelism, high simulation accuracy, and fast simulation speed. Furthermore, its semi-custom integrated circuit design offers flexibility, providing full programmability and reconfigurability, thus reducing simulation costs. The model architecture provided by this real-time simulation method can be extended to other distributed generation systems, laying the foundation for real-time simulation of distributed power systems based on small step sizes and offering valuable insights for hardware-in-the-loop simulation. Attached Figure Description

[0033] Figure 1 Inductor model schematic diagram;

[0034] Figure 2 FPGA-based inductance solution model diagram;

[0035] Figure 3Schematic diagram of a photovoltaic system;

[0036] Figure 4 FPGA-based transfer function solution model diagram;

[0037] Figure 5 FPGA-based carrier generation model diagram;

[0038] Figure 6 The operational flow structure of electrical and control systems;

[0039] Figure 7 Interactive solution timing diagram of electrical and control systems;

[0040] Figure 8 CVT power tracking waveform diagram;

[0041] Figure 9 Power tracking waveform diagram using the P&O method. Detailed Implementation

[0042] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but the following embodiments are by no means intended to limit the present invention.

[0043] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0044] A real-time simulation method for photovoltaic power generation systems based on FPGA includes two parts: FPGA modeling of the photovoltaic electrical system and FPGA modeling of the photovoltaic control system. The combination of the two parts enables real-time simulation of the photovoltaic power generation system based on FPGA.

[0045] It should be noted that FPGA modeling of a photovoltaic electrical system involves two steps:

[0046] 1) Establishing output characteristic equations for photovoltaic modules suitable for FPGA architecture; Photovoltaic cells are the basic building blocks of photovoltaic systems. To meet the needs of engineering research, this invention optimizes the four-parameter engineering model and derives output characteristic equations for photovoltaic modules suitable for FPGA architecture:

[0047] (1)

[0048] in,

[0049] (2)

[0050] and These are the output voltage and output current of the photovoltaic cell, respectively. , , , These are the short-circuit current, open-circuit voltage, maximum power point voltage, and maximum power point current of the photovoltaic cell, which are usually provided by the manufacturer.

[0051] It should be noted that, considering variations in light and temperature, parameters need to be adjusted to describe the new photovoltaic output characteristic equation. Under non-standard operating conditions, the four parameters provided by the manufacturer are corrected and can be expressed as follows:

[0052] (3)

[0053] In the formula, This represents the difference in relative light intensity between normal and standard operating conditions. The relative temperature difference between normal and standard operating conditions; temperature coefficient of current. Voltage temperature coefficient ; .

[0054] Compared to traditional photovoltaic models, the output characteristic equation provided by this invention requires only three algebraic operations. While ensuring the accuracy of the model, it improves the solution speed of the entire photovoltaic model, which is beneficial for realizing small-step simulation of the system.

[0055] It should be further noted that during the compilation process on the host computer, multiple sets of non-standard operating parameters were involved. , , , and intermediate variables , , The results can be obtained in advance and stored in a register, and then directly called by the FPGA pipeline, avoiding the direct solution of the correction process on the FPGA and improving the solution efficiency of the FPGA photovoltaic model.

[0056] 2) The passive components in the photovoltaic electrical system are differentially analyzed using the numerical integration method to obtain an equivalent circuit model in the form of the equivalent calculated admittance and the parallel connection of the historical term current source;

[0057] Passive components in an electrical system include resistive elements R, inductive elements L, capacitive elements C, and basic components composed of them. Taking an inductive branch as an example, its basic volt-ampere relationship equation is shown in equation (4). Applying the backward Euler integral, we can obtain the difference equation in equation (5), and its equivalent circuit model is as follows. Figure 1 As shown.

[0058] (4)

[0059] (5)

[0060] in,

[0061] (6)

[0062] in, This is for simulating step size; and These are the inductance parameters and equivalent admittance, respectively. and These are the inductor currents for the current time step and the previous time step, respectively. and They are respectively Nodes and Node-local time-step voltage; It is a historical current source.

[0063] FPGA-based inductor solution model as follows Figure 2 As shown, ROM is Read-Only Memory and RAM is Random Access Memory. When the start signal is activated, the current time step voltage... , and equivalent admittance It can read data from both RAM and ROM simultaneously, and perform multiplication operations concurrently using two floating-point multipliers. Then, through two floating-point subtraction operations, combined with historical current sources... Calculate the inductor current at this time step. Finally, deposit In this context, it is used for calculation in the next time step. Among them, the historical quantity current source... From current The inverse is taken, and then delayed by one simulation step. The inductance model solution is implemented with a pipelined architecture. When the first multiplication operation is processed, the subsequent multiplication units also begin processing simultaneously, and their output times differ by one clock cycle, which reflects the data flow and spatial parallelism of the FPGA.

[0064] It should be further noted that the FPGA modeling of photovoltaic control systems is based on different photovoltaic control strategies. The photovoltaic control strategies used in this invention are all existing technologies, and commonly used control strategies include constant voltage control, power disturbance method, and BP neural network adaptive tracking method. Unlike electrical systems, the component characteristics of a large class of models, including limiting elements and transfer functions, are described by input-output relationships. Therefore, the components of a control system have specific input-output relationships. In real-time simulation of active power distribution networks, electrical systems and control systems have different characteristics in terms of modeling and solution. This invention solves the control system model based on the connection relationships between various components (PI element, limiting element, etc.) in the control system, and models the functional relationships of the component inputs and outputs.

[0065] by Figure 3 Taking the constant voltage control (CVT) model as an example, this paper illustrates the modeling of a control system based on FPGA. and These are the output voltage and output current values ​​of the photovoltaic cell, respectively. This is the reference voltage for the photovoltaic cell. The CVT method model circuit is mainly solved serially. The modules ①PI stage, ②limiting stage, and ③PWM modulation stage are all implemented in the order of the control system, but the implementation process of the underlying sub-modules of each stage can make full use of spatial parallelism.

[0066] In an FPGA simulator, a continuous control system can be transformed into an explicit solution for each control element. The invention discretizes the system based on the connection order of each control element and solves for each element sequentially. Accumulation loops and feedback loops are decoupled by inserting a one-step delay. This method avoids solving the linear equations of a continuous system while also providing a degree of parallelism.

[0067] First, the PI element is discretized. The discretized expression after using Euler's integral method is as follows:

[0068] (7)

[0069] In the formula, ; and These are the proportional coefficient and the integral time coefficient, respectively.

[0070] Then to Amplitude limiting is applied, and the limiting logic is as follows:

[0071] (8)

[0072] In the formula: b is the upper limit of output; c is the lower limit of output. The FPGA-based transfer function solution model is as follows: Figure 4 As shown.

[0073] The PWM modulation stage is a crucial component of photovoltaic control systems, primarily responsible for generating control signals for switching elements. Its main challenge lies in generating the carrier signal. Traditional PWM modulation stages are based on finite state machines, using a counter and the product of amplitude to generate the carrier signal at each time step, placing higher demands on the FPGA's DSP module resources. This invention, considering the initial phase of the carrier signal, employs a signal generator design that reads data from the FPGA's ROM storage. This allows the address searcher to cyclically search the ROM address in a pipelined manner, thereby forming a periodic carrier signal. The FPGA-based carrier generator is shown below. Figure 5 As shown.

[0074] This invention designs a photovoltaic electrical system and a control system using decoupled parallel computing, and innovatively designs a timing controller to control the orderly operation of the entire system, such as... Figure 6 As shown, this paper constructs a global control hierarchy architecture: a global unified control + a low-level parallel processing architecture.

[0075] First, a global timing controller FSM0 was designed at the top level of the FPGA to precisely control the ordered solution of the electrical and control systems. Furthermore, each module and processing stage is constrained by its subsystem timing controllers FSM1 and FSM2. The photovoltaic system operates globally according to the instructions of the timing controllers, while the underlying submodules of each stage operate independently, fully utilizing the spatial parallel architecture of the FPGA.

[0076] Taking the voltage and current solving module as an example: when the module starts, the N branch voltage solving sub-modules execute simultaneously and solve in parallel, requiring a solution time T; while traditional CPU-based simulators operate serially, solving branch by branch, requiring a solution time NT. Therefore, this parallel operation architecture between sub-modules provides fundamental support for FPGA real-time simulation, greatly improving the model solving speed.

[0077] Figure 7 An interactive solution design for the electrical and control systems is presented. After the electrical system solution for the current time step is completed, the collected voltage and current parameters are stored in RAM_E, triggered by a storage end signal. When a read signal is triggered, the control system immediately reads data from RAM_E and performs the control process solution, then stores the calculated switch states in RAM_C. When the clock signal for the next time step arrives, the electrical system immediately reads data from RAM_C and performs the solution for the next time step. The FPGA can allocate independent RAM resources to the data to be interacted with. When the trigger signal arrives, the system can simultaneously and in parallel read data stored in each RAM, reducing the solution time per time step and lowering the implementation difficulty of small-step simulations.

[0078] The embodiments of the present invention are as follows:

[0079] This invention uses an Altera Stratix V series FPGA development board in photovoltaic power generation systems (such as...). Figure 2 As shown in the figure, capacitor C1 is 0.001F, capacitor C2 is 0.0002F, and inductor L is 0.004H. Considering that the simulation can quickly adjust to achieve MPPT control, the sampling time of MPPT is set to 0.001s, and the frequency of the PWM generator module is 10kHz.

[0080] To verify the real-time performance and accuracy of FPGA simulation, this invention compares the simulation results of the FPGA platform with those of PSCAD and Simulink software under different illumination and temperature conditions using the CVT and P&O photovoltaic control strategies. The FPGA simulator and the offline simulation software Simulink and PSCAD maintain a consistent simulation step size, uniformly set to 2. .

[0081] First, the CVT method was used for testing. The photovoltaic system temperature was maintained at T=25°C. o With C remaining constant, the initial light intensity is 1000 W / m². 2 It suddenly dropped to 800W / m at 0.06s. 2 At 0.08s, it decreased from 800W / m 2 It suddenly rose to 1000W / m 2 The FPGA simulation results were compared with those obtained using PSCAD and Simulink software. The output power waveform tracked by the photovoltaic system is shown below. Figure 8 As shown.

[0082] Then, the P&O method was used to test the FPGA simulator, and the test conditions were kept consistent with the CVT method. The output power waveform tracked by the P&O method is as follows: Figure 9 As shown.

[0083] In terms of simulation accuracy, Figure 8-9 A comparison of the simulation results from the FPGA real-time simulator with those from commercial software PSCAD and Simulink offline simulation is presented. Overall, under two different MPPT control strategies, the output power curves of the three simulation systems basically overlap, and the simulation results are also almost identical under environmental disturbances of 0.6s and 0.8s, thus verifying the simulation accuracy of the FPGA simulator.

[0084] To ensure the safe and stable operation of new distributed power systems, the development of high-precision, small-step real-time simulation tools has become an inevitable trend. This invention develops an FPGA-based photovoltaic real-time simulator, which possesses high hardware and time parallelism, fully demonstrating the advantages and potential of FPGAs in simulation accuracy and scale. Furthermore, this real-time simulation model architecture is also applicable to other distributed power generation systems, laying the foundation for real-time simulation of distributed power systems based on small-step simulation.

Claims

1. A real-time simulation method for a photovoltaic power generation system based on FPGA, characterized in that, include 1) Perform FPGA modeling of the photovoltaic electrical system: Establish output characteristic equations for photovoltaic modules suitable for FPGA architecture; The passive components in the photovoltaic electrical system are differentially analyzed using the numerical integration method to obtain an equivalent circuit model in the form of the equivalent calculated admittance and the parallel connection of the historical term current source; 2) FPGA modeling of photovoltaic control systems based on different photovoltaic control strategies; The output characteristic equation of the photovoltaic module suitable for FPGA architecture is as follows: （1） in, （2） in, and These are the output voltage and output current of the photovoltaic cell, respectively. , , , These are the short-circuit current, open-circuit voltage, maximum power point voltage, and maximum power point current of the photovoltaic cell, which are usually provided by the manufacturer. Considering variations in light and temperature, parameters need to be modified to describe the new photovoltaic output characteristic equation. Under non-standard operating conditions, the four parameters provided by the manufacturer are modified, and can be expressed as follows: （3） in, This represents the difference in relative light intensity between normal and standard operating conditions. The relative temperature difference between normal and standard operating conditions; temperature coefficient of current. Voltage temperature coefficient ; ; The intermediate variable , , and multiple sets of non-standard operating condition parameters , , , The value is obtained in advance during the compilation process on the host computer, stored in a register, and directly called by the FPGA pipeline.

2. The real-time simulation method for a photovoltaic power generation system based on FPGA according to claim 1, characterized in that, The simulation method adopts a global control hierarchy architecture: a global timing controller FSM0 is designed at the top level of the FPGA to precisely control the orderly solution of the electrical system and the control system. At the same time, each module and processing link is controlled by its subsystem timing controller.

3. The real-time simulation method for a photovoltaic power generation system based on FPGA according to claim 2, characterized in that, This simulation method employs an interactive solution design for both the electrical and control systems, specifically including: Once the electrical system solution for this time step is completed, the collected voltage and current parameters are stored in RAM_E, triggered by the storage end signal. When the read signal is triggered, the control system immediately reads the data from RAM_E and performs the control process solution. Then, the switch state obtained after the solution is completed is stored in RAM_C. When the clock signal for the next time step arrives, the electrical system immediately reads the data from RAM_C and performs the solution for the next time step.

4. The real-time simulation method for a photovoltaic power generation system based on FPGA according to claim 3, characterized in that, The continuous control system is transformed into an explicit solution for each control element, where, The PI element of the photovoltaic control system is discretized using Euler's formula. The specific processing method is as follows: （7） in, ; and These are the proportional coefficient and the integral time coefficient, respectively. The processing method for the limiting stage of the photovoltaic control system is as follows: （8） in, This is the input for the amplitude limiting process. is the output of the limiting circuit, b is the upper limit of the output; c is the lower limit of the output.

5. The real-time simulation method for a photovoltaic power generation system based on FPGA according to any one of claims 1-4, characterized in that, The PWM modulation stage of the photovoltaic control uses a signal generator design that reads data stored in the ROM of the FPGA, so that the address seeker searches for the ROM address in a pipeline manner, thereby forming a periodic carrier signal.

Images