Photovoltaic control system based on virtual load

Abstract

The invention relates to the technical field of photovoltaic control, in particular to a photovoltaic control system based on a virtual load, which comprises a solar controller, a lithium battery energy storage subsystem, an inverter, an actual load connecting end and an external mains supply input end, and the inverter is internally provided with a programmable virtual load and is integrated with a power regulation module and a DC / AC conversion module. The system simulates the characteristics of a target load or an energy storage unit through a virtual load, combines the historical data of a memory module and a load learning strategy, rehearses and generates an optimal switching instruction before power supply switching, supports off-grid simulation and closed-loop correction, and improves the power supply continuity, safety and energy utilization efficiency.

Description

Technical Field

[0001] This invention relates to grid photovoltaic systems, and more specifically, to a photovoltaic control system based on virtual loads. Background Technology

[0002] In the field of renewable energy power generation technology, photovoltaic (PV) power generation systems are widely used due to their clean and sustainable characteristics. Their grid connection and power consumption are key links in ensuring the stability and efficiency of the power system. Load forecasting and dynamic matching of power generation constitute the core technical issues of system operation. Distributed PV power consumption systems aim to achieve efficient matching between PV power generation and electricity load through forecasting and scheduling. This system typically involves monitoring, analyzing, and optimizing the control of multiple distributed PV units and electricity loads within a region to improve the local consumption rate of PV energy and reduce the impact on the upper-level power grid. Existing technologies mostly employ centralized forecasting models or single load control strategies, which are difficult to effectively address the spatiotemporal differences in electricity consumption patterns in distributed scenarios. Traditional methods experience a significant decrease in forecast accuracy under conditions of drastic fluctuations in PV output power or complex and variable load demand, and lack a dynamic identification and configuration mechanism for electricity consumption priorities. For example, a PV charging controller with publication number CN103219765B... This patent uses a PWM control circuit to achieve dual-mode switching between MPPT and constant-voltage float charging. It charges at maximum power when the battery is not fully charged and switches to a regulated state after full charge, balancing efficiency and battery protection. However, this solution primarily focuses on static control of a single battery load, failing to fully consider the diversity and dynamic changes of actual loads. It also lacks a virtual load to predict or simulate different power consumption scenarios and lacks an evaluation mechanism for the target load's power consumption characteristics before power supply switching. This limits the adaptability and predictive ability of the switching process, making it difficult to meet the power supply requirements of multiple scenarios and high reliability. For example, a control method for a photovoltaic system (publication number CN111934341A) collects power signal time series data from a power meter, and the system controller analyzes the load's power consumption behavior to achieve coordinated control of the energy conversion module. While it possesses some data acquisition and analysis capabilities, its core still relies on actual operating data from real loads, making it difficult to conduct off-grid simulations or characteristic predictions when there is no load or the load is not connected. Furthermore, the system lacks a programmable virtual load, the ability to proactively construct test conditions to obtain feedback information, and a mechanism for comparing and correcting based on historical data and the current state. As a result, its level of intelligence in areas such as energy storage charging optimization, load identification and learning, and dynamic adjustment of power supply strategies is limited.

[0003] The main problem is the insufficient capacity to absorb renewable photovoltaic (PV) resources. In most cases, PV power needs to be first transmitted to grid-connected energy storage stations for storage before being distributed to various user nodes. This results in significant line losses during PV transmission, and both energy storage and dispatch require resource allocation. While some technologies disclose the idea of ​​localizing PV energy utilization, this is generally limited to the user's own PV power generation unit. If cross-user dispatch is involved, the mismatch between user load variability and the supply capacity of PV power generation units can negatively impact power quality. Therefore, how to ensure power quality while fully utilizing PV resources is a topic that requires further research. Summary of the Invention

[0004] In view of this, the purpose of this invention is to provide a photovoltaic control system based on virtual load.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: A photovoltaic control system based on virtual load includes a solar controller, a lithium battery energy storage subsystem, an inverter, an actual load connection terminal, and an external mains power input terminal. The inverter has a built-in programmable virtual load with a memory module; The photovoltaic output terminal of the solar controller, the external mains input terminal, the charging / discharging terminal of the lithium battery energy storage subsystem, and the actual load connection terminal are all uniquely connected to the corresponding interfaces of the inverter, so that the power transmission in photovoltaic power supply, photovoltaic charging, external mains power supply, external mains charging, and energy storage power supply scenarios all pass through the inverter. The virtual load is integrated with the inverter's DC / virtual load photovoltaic control system C-conversion module, power regulation module, and control unit, and is used to collect power consumption information data for each scenario and feed the data back to the control unit. The control unit is configured with a power supply response strategy and a power supply control strategy. The power supply response strategy responds to power consumption switching requests and connects the programmable virtual load according to the power consumption switching requests to obtain feedback information. The power supply control strategy is used to analyze the feedback information to obtain power supply switching information. Existing technologies rely on the actual load operating state for strategy generation, and cannot pre-simulate power supply behavior when there is no load or the load is not connected, and lack an active identification mechanism for the characteristics of the target load. This solution embeds the programmable virtual load inside the inverter, sharing hardware channels with the power regulation module and DC / AC conversion module, so that all power supply paths are forced to be uniformly scheduled through the inverter. Upon receiving a power switching request, the control unit first drives a virtual load to simulate the electrical characteristics of the target load, obtains feedback information, and then executes the actual switching operation, avoiding voltage drops or current surges caused by sudden load changes. This structural design enables the system to have pre-testing capabilities during multi-mode switching processes such as off-grid, grid-connected, and energy storage charging and discharging, improving power supply continuity and equipment safety.

[0006] Furthermore, it also includes a memory module, which stores power consumption information data under various power supply and charging scenarios, and allows the control unit to call upon historical data in the memory module for comparison and analysis with current detection data. Traditional systems only record operation logs and cannot achieve real-time comparison between historical and current states. This solution drives the initialization process of the virtual load through deviation vectors, making the simulated operating conditions closer to the historical behavior characteristics of the actual load and improving the representativeness of the feedback information.

[0007] Furthermore, the virtual load is configured with a first load group and a second load group. When the target of the power switching request is the actual load, the power supply response strategy executes a comparison sub-strategy. This comparison sub-strategy includes retrieving corresponding power consumption information data based on the power switching request to obtain a first configuration instruction to configure the first load group, obtaining the current actual load's power consumption characteristic data to obtain a second configuration instruction to configure the second load group, and connecting the first and second load groups to the power supply terminals respectively to obtain the feedback information. The power supply current of the two load groups is independently controlled by the power regulation module, and their respective voltage response waveforms, power fluctuation amplitudes, and frequency stability are recorded to form feedback information. This structure, through a dual-load parallel testing mechanism, achieves synchronous verification of historical behavior and current characteristics, avoiding misjudgments caused by a single simulation.

[0008] Furthermore, the control unit is configured with a load learning strategy, which in turn is configured with a load characteristic library. The load learning strategy generates load characteristic information based on electricity consumption data and stores it in the corresponding load characteristic library. After each feedback message is received, the load learning strategy extracts the steady-state power, inrush surge multiple, resonant frequency, and dynamic response time to generate load characteristic information, which is then stored in the corresponding sub-library according to the load type tag. This mechanism allows the system to gradually accumulate load knowledge after multiple switches, enabling the direct call to a matching virtual load configuration template when similar loads are subsequently connected, thus shortening response latency.

[0009] Furthermore, when the target of the power switching request is a lithium battery energy storage subsystem, the power supply response strategy executes a characteristic capture sub-strategy. This characteristic capture sub-strategy includes generating a mapping configuration instruction based on lithium battery data to configure a virtual load, and connecting the virtual load to the power supply terminal to obtain feedback information. This process allows the system to predict the energy storage unit's accepting capacity before formal charging, preventing damage from overcurrent or overvoltage.

[0010] Furthermore, the lithium battery energy storage subsystem includes a hierarchical management system composed of multiple battery cells. The battery cells are classified into different levels based on their health status and charge / discharge performance. This hierarchical information is transmitted to the control unit. The characteristic capture sub-strategy generates corresponding mapping configuration sub-instructions based on the hierarchical information of each different battery cell. The virtual load includes several charging simulation load units, and the mapping configuration sub-instructions are used to configure these charging simulation load units. This structure supports the simulation of unit-level characteristics for multiple series-parallel battery packs, improving the granularity of the charging strategy.

[0011] Furthermore, the power supply control strategy includes a charging control sub-strategy. When the target of the power switching request is the lithium battery energy storage subsystem, the power supply control strategy executes the charging control sub-strategy. This charging control sub-strategy is configured to generate a power supply sequence based on power supply characteristic data and hierarchical information. The power supply sequence reflects the order and timing of battery cell access to the power supply system. The control unit sequentially closes the charging relays of the corresponding battery cells according to the power supply sequence and dynamically adjusts the inverter output voltage. This mechanism prevents high-resistance battery cells from overheating due to forced high-current charging, thus extending the lifespan of the energy storage system.

[0012] Furthermore, the power supply control strategy includes a power supply characteristic analysis sub-strategy for generating power supply characteristic data based on feedback information. This sub-strategy is configured with a characteristic analysis library, which pre-stores several power supply characteristic data points. Each power supply characteristic data point is indexed by load test features and feedback information features. The feedback information features are extracted from the feedback information using a preset feature extraction algorithm. The load test features reflect the operating parameters of the load when the feedback information is generated. Wavelet decomposition is performed on the voltage waveform to obtain high-frequency disturbance energy, FFT transformation is performed on the current signal to extract the dominant harmonic components, and sliding window variance calculation is performed on the power curve to quantify the fluctuation intensity. This analysis process transforms the raw feedback information into a structured feature vector, supporting subsequent strategy decisions.

[0013] Furthermore, the control unit is also configured with an off-grid simulation strategy. When the actual load has no power demand, the off-grid simulation strategy is executed. This strategy generates off-grid simulation information and configures the corresponding power supply terminal to connect to the power supply circuit based on the information. Simultaneously, a simulation strategy is generated to configure the corresponding virtual load to collect simulated power consumption data under simulated conditions. Based on the off-grid simulation information, the corresponding virtual load configuration template is selected, the simulated power supply circuit switch inside the inverter is closed, and the virtual load is started. At the same time, the output voltage stability, frequency drift, and harmonic distortion rate under simulated conditions are recorded to form simulated power consumption data. This mechanism maintains the system's active state when there is no real load, verifying off-grid operation capability.

[0014] Furthermore, the control unit also includes a comparison correction strategy. This strategy compares simulated power consumption data with corresponding power consumption information data to generate a comparison deviation, and generates correction parameters based on the comparison deviation to correct the power supply characteristic data. The correction parameters are used to adjust the PWM carrier frequency, dead time, and filter inductor parameters of the DC / AC conversion module, and update the power supply characteristic data of the corresponding index item in the characteristic analysis library. This closed-loop correction mechanism enables the system to continuously optimize power supply quality during long-term operation, adapting to device aging and environmental changes.

[0015] The main technical advantages of this invention are reflected in the following aspects: By embedding a programmable virtual load inside the inverter and integrating it with the power regulation module and DC / AC conversion module, all power supply paths are forced to pass through a unified scheduling node; before power supply switching, the control unit drives the virtual load to simulate the electrical characteristics of the target load or energy storage unit, obtains feedback information, and then executes the actual switching; the memory module stores historical power consumption information data and supports the calculation of the deviation between historical and current states; the load learning strategy constructs a load characteristic library to achieve automatic identification and configuration reuse of load types; for the lithium battery energy storage subsystem, through the coordination of a hierarchical management system and charging simulation load units, it achieves unit characteristic simulation and orderly charging; the off-grid simulation strategy and the comparison and correction strategy constitute a closed-loop optimization mechanism to continuously improve power supply quality. This system solves the problems of existing technologies, such as the inability to perform pre-simulation under no-load conditions, the lack of pre-assessment during switching processes, and the coarse nature of energy storage charging strategies, thereby improving the utilization efficiency of photovoltaic energy and the reliability of power supply. Attached Figure Description

[0016] Figure 1 The overall architecture diagram of the photovoltaic control system based on virtual load of this invention; Figure 2 : Schematic diagram of the integrated structure of virtual load and inverter of this invention; Figure 3 : Power supply switching control flowchart of this invention; Figure 4 : Schematic diagram of lithium battery graded management and charging control in this invention; Figure 5 : Schematic diagram of the off-grid simulation and comparison correction closed loop of this invention. Detailed Implementation

[0017] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings, so that the technical solution of the present invention can be more easily understood and mastered.

[0018] A photovoltaic control system based on virtual load includes a solar controller, a lithium battery energy storage subsystem, an inverter, an actual load connection terminal, and an external mains power input terminal. The inverter incorporates a programmable virtual load with a built-in memory module. It adopts a full-bridge topology, integrating a DC / AC conversion module, a power regulation module, a control unit, and the programmable virtual load. All modules share hardware channels to simplify the structure and ensure consistent power transmission. The DC / AC conversion module uses IGBT power devices to form a full-bridge inverter circuit with a rated output voltage of 220VAC and a rated output power configured from 5kW to 50kW depending on the application scenario. The power regulation module achieves continuous adjustment of output power through PWM pulse width modulation technology. The control unit uses an STM32H7 series microcontroller with a built-in 16-bit ADC acquisition module. The data acquisition sampling frequency is set to 1kHz, used to receive electrical signal data from the virtual load and various interfaces in real time and execute strategy calculations. The memory module uses an industrial-grade SD card as the storage medium, with a storage capacity of ≥32GB, to store power consumption information data under various power supply and charging scenarios. The stored data includes key parameters such as voltage, current, power, power fluctuation amplitude, dynamic response time, and frequency stability.

[0019] The photovoltaic output terminal of the solar controller, the external mains input terminal, the charging / discharging terminal of the lithium battery energy storage subsystem, and the actual load connection terminal are all uniquely connected to the corresponding interfaces of the inverter, ensuring that power transmission in photovoltaic power supply, photovoltaic charging, external mains power supply, external mains charging, and energy storage power supply scenarios all occurs through the inverter. The solar controller is an MPPT type controller, with its photovoltaic output terminal uniquely connected to the inverter's photovoltaic input interface via a shielded cable. The cable cross-section is selected based on the rated current to ensure transmission loss ≤3%. The external mains input terminal is connected to the inverter's mains interface via an air switch. The air switch's rated current is 1.2 times the system's rated current, providing overcurrent protection. The lithium battery energy storage subsystem consists of several 18650 lithium battery cells connected in series and parallel. Each battery cell has a rated voltage of 3.7V and a rated capacity of 2000mAh. Its charging / discharging terminal is connected to the inverter's energy storage interface via a BMS battery management system, which collects data such as voltage, current, and temperature from each battery cell. The actual load connection terminal is equipped with wiring terminals for built-in current and voltage sensors, used to connect various actual loads (such as household appliances, industrial equipment, etc.) and collect the load's power consumption characteristic data in real time. The virtual load is integrated with the inverter's DC / virtual load photovoltaic control system C-conversion module, power regulation module, and control unit, used to collect power consumption information data for various scenarios and feed the data back to the control unit. The control unit is configured with a power supply response strategy and a power supply control strategy; the power supply response strategy responds to power consumption switching requests and connects to the programmable virtual load according to the power consumption switching requests to obtain feedback information, and the power supply control strategy is used to analyze the feedback information to obtain power supply switching information.

[0020] It also includes a memory module, which is used to store power consumption information data under various power supply and charging scenarios, and allows the control unit to call the historical data in the memory module for comparison and analysis with the current detection data.

[0021] Furthermore, the virtual load is configured with a first load group and a second load group. When the target of the power switching request is the actual load, the power supply response strategy executes a comparison sub-strategy. This comparison sub-strategy includes retrieving corresponding power consumption information data based on the power switching request to obtain a first configuration instruction to configure the first load group, obtaining the current actual load's power consumption characteristic data to obtain a second configuration instruction to configure the second load group, and connecting the first and second load groups to the power supply terminals respectively to obtain the feedback information. The power supply current of the two load groups is independently controlled by the power regulation module, and their respective voltage response waveforms, power fluctuation amplitudes, and frequency stability are recorded to form feedback information. This structure, through a dual-load parallel testing mechanism, achieves synchronous verification of historical behavior and current characteristics, avoiding misjudgments caused by a single simulation.

[0022] The comparison sub-strategy is as follows: After receiving a power switching request for the actual load, the control unit parses the load identification information in the request; based on the load identification information, it retrieves the historical power consumption information data corresponding to the load from the memory module, including historical steady-state power. Historical startup surge current Historical dynamic response time Based on these parameters, a first configuration instruction is generated. The parameters in the first configuration instruction are calculated using the following formula: , Where R1 is the equivalent resistance parameter of the first load group, The equivalent inductance parameters for the first load group are... The system's rated output voltage; the control unit collects current load power consumption data, including the current no-load voltage, from the built-in sensor at the actual load connection terminal. Current short-circuit current Current equivalent impedance Based on this data, a second configuration instruction is generated. The parameters of the second configuration instruction are calculated as follows: , ,in The equivalent resistance parameters for the second load group are as follows: The equivalent inductance parameters for the second load group are as follows: The power factor angle is the current actual load. The system's rated output frequency is set; the control unit sends a first configuration command to the first load group and a second configuration command to the second load group, completing the parameter configuration of the two load groups; the inverter's power regulation module connects the first and second load groups to the power supply terminal respectively, and continuously collects the voltage response waveforms after the two load groups are connected. , and current response waveform , Calculate the actual power of the two load groups. , And record the power fluctuation amplitude. , and frequency stability , (The difference between frequency fluctuation and rated frequency) and integrate these data into feedback information.

[0023] The control unit is configured with a load learning strategy, which in turn is configured with a load characteristic library. The load learning strategy generates load characteristic information based on electricity consumption data and stores it in the corresponding load characteristic library. After each feedback information is obtained, the load learning strategy extracts the steady-state power, startup surge multiple, resonant frequency, and dynamic response time to generate load characteristic information, and stores it in the corresponding sub-library according to the load type tag.

[0024] The load learning strategy is an adaptive mechanism by which the control unit extracts load characteristic parameters based on collected electricity consumption information data, builds and updates the load characteristic library, and enables rapid matching and configuration of similar loads in the future. The specific execution steps are as follows: The control unit receives power consumption information data collected from the virtual load, including the steady-state power sequence during load operation. Peak inrush current Dynamic response time Resonant frequency The steady-state power sequence P(t) is subjected to moving average filtering, and the filtering formula is as follows: Where N is the length of the sliding window. The data acquisition time interval (value is 1ms). The filtered steady-state power; extract the load characteristic parameter set. A unique load type label T is assigned to each load, generated by combining the load's power level range and application scenario type. The load characteristic library is then queried to determine if a characteristic parameter set with the same load type label T exists. If not, a new storage entry is created, and the characteristic parameter set S is stored in that entry. If the set already exists, the characteristic parameters under that entry are updated using a weighted average method, with the update formula as follows: ,in For the updated feature parameter set, S_old represents the currently extracted set of feature parameters, while S_old represents the original set of feature parameters in the library. The weighting coefficient is used to store the updated load characteristic library in the memory module, thus completing the load learning process.

[0025] When the target of the power switching request is the lithium battery energy storage subsystem, the power supply response strategy executes a characteristic capture sub-strategy. This characteristic capture sub-strategy includes generating a mapping configuration instruction based on lithium battery data acquisition information to configure a virtual load, and connecting the virtual load to the power supply terminal to obtain feedback information. The characteristic capture sub-strategy is as follows: First, after receiving the power switching request directed to the lithium battery energy storage subsystem, the control unit sends a data acquisition instruction to the BMS system of the lithium battery energy storage subsystem; second, the BMS system acquires the voltage of each battery cell. Current The first step involves collecting lithium battery information such as remaining SOC, SOH, and charge / discharge efficiency η, and transmitting this information to the control unit. The third step involves the control unit generating a mapping configuration instruction based on the lithium battery information. The charging simulation load unit parameters in the mapping configuration instruction are calculated using the following formula: ,in The equivalent resistance parameters of the charging simulation load unit are used. The control unit sends the mapping configuration command to the virtual load's charging simulation load unit to complete the configuration of the simulation load parameters corresponding to the battery unit; the configured charging simulation load unit is then connected to the power supply terminal, and the charging voltage after connection is collected. Charging current Charging time The data is processed to generate feedback information and transmitted to the control unit.

[0026] The lithium battery energy storage subsystem includes a hierarchical management system composed of multiple battery cells. These battery cells are classified into different levels based on their health status and charge / discharge performance. This hierarchical information is transmitted to the control unit. The characteristic capture sub-strategy generates corresponding mapping configuration sub-instructions based on the hierarchical information of each battery cell. The virtual load includes several charging simulation load units, and the mapping configuration sub-instructions are used to configure these charging simulation load units. The hierarchical management system of the lithium battery energy storage subsystem is used to classify battery cells according to their health status and charge / discharge performance, providing a basis for the charging strategy: the BMS system periodically (every 5 minutes) collects the State of Health (SOH) and charge / discharge performance parameters of each battery cell. The charge / discharge performance parameters include the rated charge / discharge power. Maximum charging and discharging current Charge-discharge cycle life The fuzzy comprehensive evaluation method is used to classify each battery cell, and the classification evaluation index and weight are preferably SOH of 0.4. Preferably 0.3, The preferred value is 0.2. The preferred value is 0.1, and the grading evaluation formula is: ,in The overall score for the battery cell. This represents the maximum rated charge and discharge power of all battery cells. The maximum charge / discharge current among all battery cells The maximum charge-discharge cycle life among all battery cells; based on comprehensive rating. Classification by level, with level A corresponding to ≥0.8, Grade B corresponds to 0.6≤ <0.8, corresponding to grade C If the value is less than 0.6, hierarchical information for each battery cell is generated. This hierarchical information is transmitted to the control unit in real time. The control unit establishes a battery cell hierarchical information table and stores it in association with the identifier ID of each battery cell. Furthermore, the corresponding identifier ID is dynamically updated when the battery's state changes.

[0027] The power supply control strategy includes a charging control sub-strategy. When the target of the power switching request is a lithium battery energy storage subsystem, the power supply control strategy executes the charging control sub-strategy. This charging control sub-strategy is configured to generate a power supply sequence based on power supply characteristic data and hierarchical information. The power supply sequence reflects the order and timing of battery cell access to the power supply system. The control unit sequentially closes the charging relays of the corresponding battery cells according to the power supply sequence and dynamically adjusts the inverter output voltage. The charging control sub-strategy is used to generate a power supply sequence based on power supply characteristic data and hierarchical information when the target of the power switching request is a lithium battery energy storage subsystem, achieving orderly charging: the control unit receives feedback information from the characteristic capture sub-strategy and extracts the power supply characteristic data, including the power supply voltage. Power supply current Power supply Power supply stability The system retrieves the classification information of the lithium battery energy storage subsystem, sorts it from highest to lowest level, prioritizes level A battery cells, and within the same level, sorts them from lowest to highest remaining state of charge (SOC). Based on the power supply characteristic data and the sorting results, a power supply sequence is generated, specifying the connection time of each battery cell in the power supply sequence. The calculation is as follows: ,in At the initial moment of charging start-up, This is the sequence number of the battery cell in the sorting. This is the connection time interval between adjacent battery cells, ranging from 0.5 to 2 seconds, depending on power supply stability. Adjustment, The larger the interval, the longer the interval; the control unit sends closing commands to the charging relays of each battery unit sequentially according to the power supply sequence, and dynamically adjusts the inverter's output voltage based on the power supply characteristic data. The adjustment formula is: ,in This is the actual output voltage of the inverter. This is the voltage compensation amount. The compensation coefficient ranges from 0.8 to 1.2. During the charging process, the charging status of each battery cell is continuously monitored. When the remaining charge (SOC) of a battery cell reaches 100% or an overcurrent or overvoltage abnormality occurs, a trip command is sent to disconnect the charging relay of that battery cell, and the power supply sequence is updated at the same time.

[0028] The power supply control strategy includes a power supply characteristic analysis sub-strategy for generating power supply characteristic data based on feedback information. This sub-strategy is configured with a characteristic analysis library, which pre-stores several power supply characteristic data points. Each power supply characteristic data point is indexed by load test features and feedback information features. The feedback information features are extracted from the feedback information using a preset feature extraction algorithm. The load test features reflect the operating parameters of the load when the feedback information is generated. The control unit receives feedback information from the virtual load, including raw data such as voltage time series U(t), current time series I(t), and power time series P(t). Wavelet decomposition is performed on the voltage time series U(t) using the db4 wavelet basis function, with a decomposition level of 3, to extract high-frequency disturbance energy. The calculation method is as follows: ,in These are the high-frequency coefficients after wavelet decomposition. The scale number after decomposition. Number of sampling points at each scale; for current time series Perform a Fast Fourier Transform to extract the frequency of the dominant harmonic component. and amplitude The FFT transformation formula is: ,in The transformed frequency domain amplitude, For time-domain sampling point index, Here, N is the frequency domain point index, and N is the number of sampling points; for power time series... Perform sliding window variance calculation to quantify power fluctuation intensity. The calculation formula is: ,in This represents the number of sampling points within the sliding window (with a value of 50). For the first in the window Power values ​​at each sampling point The average power value within the window; the extracted , , , The parameters are combined to form power supply characteristic data, which are then stored in the characteristic analysis library using load test characteristics and feedback information characteristics as indexes.

[0029] The control unit is also configured with an off-grid simulation strategy. When the actual load has no power demand, the off-grid simulation strategy is executed. This strategy generates off-grid simulation information and configures the corresponding power supply terminal to connect to the power supply circuit based on this information. Simultaneously, it generates a simulation strategy to configure a corresponding virtual load to collect simulated power consumption data under simulated conditions. The off-grid simulation strategy is used to simulate off-grid operation scenarios and collect data through virtual loads when the actual load has no power demand. The control unit detects the actual load's power demand through sensors at the actual load connection point. If the detected current value of the actual load remains below a threshold for 5 seconds, the control unit will take action. Preferred When the value is 0.1A, it is determined that there is no electricity demand, triggering the off-grid simulation strategy. The control unit generates off-grid simulation information, including parameters such as the simulated power supply mode (photovoltaic power supply, energy storage power supply, or hybrid power supply), the simulated load type (resistive load, inductive load, or composite load), and the simulated power level. The simulated power supply mode is selected randomly. Based on the off-grid simulation information, the corresponding power supply terminal is configured to connect to the power supply circuit. If the photovoltaic power supply mode is selected, the switch between the solar controller and the inverter is closed; if the energy storage power supply mode is selected, the switch between the lithium battery energy storage subsystem and the inverter is closed; if the hybrid power supply mode is selected, both switches are closed simultaneously. In the fourth step, the simulation strategy is generated. Based on the simulated load type and power level, the corresponding load characteristic parameters are retrieved from the load characteristic library, and the equivalent resistance of the virtual load is configured. Equivalent inductance Equivalent capacitance The parameters are configured using the following formula: , , ,in This is the power value corresponding to the simulated power level. To simulate the inductive reactance of an inductive load; set according to the load type, inductive load. =10Ω, composite load =5Ω, To simulate the capacitive reactance of a capacitive load, the capacitive load... =10Ω, composite load =5Ω; Start the virtual load and continuously collect simulated power consumption data under simulated conditions, including simulated voltage. Simulated current Analog power Frequency drift, which reflects the difference between the analog frequency and the rated frequency. Harmonic distortion rate, which reflects the ratio of total harmonic content to fundamental frequency content. The power consumption simulation data will be stored in the memory module.

[0030] The control unit also includes a comparison correction strategy. This strategy compares simulated power consumption data with corresponding power consumption information data to generate a comparison deviation, and generates correction parameters based on the deviation to correct the power supply characteristic data. The correction parameters adjust the PWM carrier frequency, dead time, and filter inductor parameters of the DC / AC converter module, and update the power supply characteristic data of the corresponding index item in the characteristic analysis library. This closed-loop correction mechanism enables the system to continuously optimize power supply quality during long-term operation, adapting to device aging and environmental changes. The comparison correction strategy corrects the power supply characteristic data and optimizes system parameters by comparing simulated power consumption data with historical power consumption information data. The specific execution steps are as follows: The control unit retrieves the simulated electricity consumption data and corresponding historical electricity consumption information data generated by the off-grid simulation strategy from the memory module. The historical electricity consumption information data consists of actual operating data under the same simulated power supply mode and simulated load type. It then calculates the comparison deviation between the simulated electricity consumption data and the historical electricity consumption information data, including the voltage deviation. Current deviation: Power deviation: Frequency deviation: ,in , , , These are the average voltage, average current, average power, and average frequency from the electricity consumption simulation data. , , , These are the corresponding average values ​​from historical electricity consumption data; correction parameters are generated based on the comparison deviation, including the voltage correction coefficient. Current correction factor Power correction factor Frequency correction factor ,in The rated output current of the system is calculated from the rated power and rated voltage. The system's rated output power is used; the power supply characteristic data is corrected using correction parameters, and the high-frequency disturbance energy is included in the corrected power supply characteristic data. Dominant harmonic amplitude Power fluctuation intensity Frequency-related parameters The corrected power supply characteristic data is updated to the characteristic analysis library, and the corrected parameters are sent to the DC / AC conversion module and the power regulation module to adjust the PWM carrier frequency of the DC / AC conversion module. ( (Initial PWM carrier frequency, valued at 10kHz), dead time , Adjust the filter inductor parameters of the power regulation module to set the initial dead time. , The initial filter inductor parameter is set to 1mH.

[0031] Of course, the above are just typical examples of the present invention. In addition, the present invention may have many other specific embodiments. All technical solutions formed by equivalent substitution or equivalent transformation fall within the scope of protection claimed by the present invention.

Claims

1. A photovoltaic control system based on virtual load, comprising a solar controller, a lithium battery energy storage subsystem, an inverter, an actual load connection terminal, and an external mains input terminal, characterized in that: The inverter has a built-in programmable virtual load with a memory module; The photovoltaic output terminal of the solar controller, the external mains input terminal, the charging and discharging terminal of the lithium battery energy storage subsystem, and the actual load connection terminal are all uniquely connected to the corresponding interface of the inverter, so that the power transmission in photovoltaic power supply, photovoltaic charging, external mains power supply, external mains charging and energy storage power supply scenarios all pass through the inverter. The virtual load is integrated with the inverter's DC / virtual load photovoltaic control system C-conversion module, power regulation module, and control unit. It is used to collect power consumption information data for each scenario and feed the data back to the control unit. The control unit is configured with a power supply response strategy and a power supply control strategy. The power supply response strategy responds to the power consumption switching request and accesses the programmable virtual load according to the power consumption switching request to obtain feedback information. The power supply control strategy is used to analyze the feedback information to obtain power supply switching information.

2. The photovoltaic control system based on virtual load according to claim 1, characterized in that: It also includes a memory module, which is used to store power consumption information data under various power supply and charging scenarios, and allows the control unit to call the historical data in the memory module for comparison and analysis with the current detection data.

3. A photovoltaic control system based on virtual load according to claim 2, characterized in that: The virtual load is configured with a first load group and a second load group. When the target of the power switching request is the actual load, the power supply response strategy executes a comparison sub-strategy. The comparison sub-strategy includes retrieving the corresponding power consumption information data according to the power switching request to obtain a first configuration instruction to configure the first load group, obtaining the power consumption characteristic data of the current actual load to obtain a second configuration instruction to configure the second load group, and connecting the first load group and the second load group to the power supply end respectively to obtain the feedback information.

4. A photovoltaic control system based on virtual load according to claim 2, characterized in that: The control unit is configured with a load learning strategy, and the load learning strategy is configured with a load characteristic library. The load learning strategy is used to generate load characteristic information based on electricity consumption information data and store it in the corresponding load characteristic library.

5. A photovoltaic control system based on virtual load according to claim 2, characterized in that: When the target of the power switching request is a lithium battery energy storage subsystem, the power supply response strategy executes a feature capture sub-strategy. The feature capture sub-strategy includes generating a mapping configuration instruction based on lithium battery acquisition information to configure a virtual load, and connecting the virtual load to the power supply end to obtain feedback information.

6. A photovoltaic control system based on virtual load according to claim 5, characterized in that: The lithium battery energy storage subsystem includes a hierarchical management system composed of multiple battery cells. The battery cells are classified into different levels of hierarchical information according to their health status and charge / discharge performance. The hierarchical information is transmitted to the control unit. The feature capture sub-strategy generates corresponding mapping configuration sub-instructions based on the hierarchical information of each different battery cell. The virtual load includes several charging simulation load units. The mapping configuration sub-instructions are used to configure the charging simulation load units.

7. A photovoltaic control system based on virtual load according to claim 6, characterized in that: The power supply control strategy includes a charging control sub-strategy. When the target of the power switching request is a lithium battery energy storage subsystem, the power supply control strategy executes the charging control sub-strategy. The charging control sub-strategy is configured to generate a power supply sequence based on power supply characteristic data and hierarchical information. The power supply sequence reflects the order and time of battery cells connecting to the power supply system.

8. A photovoltaic control system based on virtual load according to claim 1, characterized in that: The power supply control strategy includes a power supply characteristic analysis sub-strategy for generating power supply characteristic data based on feedback information. The power supply characteristic analysis sub-strategy is configured with a characteristic analysis library, which pre-stores a number of power supply characteristic data. Each power supply characteristic data is indexed by load test features and feedback information features. The feedback information features are extracted from the feedback information using a preset feature extraction algorithm. The load test features reflect the operating parameters of the load when the feedback information is generated.

9. A photovoltaic control system based on virtual load according to claim 8, characterized in that: The control unit is also configured with an off-grid simulation strategy. When the actual load has no power demand, the off-grid simulation strategy is executed. The off-grid simulation strategy generates off-grid simulation information and configures the corresponding power supply terminal to connect to the power supply circuit according to the off-grid simulation information. At the same time, a simulation strategy is generated to configure the corresponding virtual load to collect power consumption simulation data in the simulation state.

10. A photovoltaic control system based on virtual load according to claim 9, characterized in that: The control unit also includes a comparison correction strategy, which is used to compare the power consumption simulation data and the corresponding power consumption information data to generate a comparison deviation, and generate correction parameters based on the comparison deviation to correct the power supply characteristic data.

Images