A combined photovoltaic energy storage system

Abstract

The application discloses a combined photovoltaic energy storage system, and belongs to the technical field of solar power generation and energy storage. The system comprises a photovoltaic input module, a high-frequency isolation power conversion module, an energy storage battery module, a grid-connected inverter output module and a control module. The high-frequency isolation power conversion module internally comprises a three-winding high-frequency transformer, realizing magnetic coupling power transmission and electrical isolation between three ports of photovoltaic input, energy storage battery and grid-connected output. The three-winding transformer has preset leakage inductance between each winding, and forms LC filter circuits with corresponding capacitors, filtering high-frequency ripples and suppressing harmonics. The control module monitors the state of each module in real time and dynamically adjusts the switching frequency and phase shift angle. The application fundamentally solves the power coupling interference and fault propagation problem in the traditional DC coupling scheme through the three-port high-frequency isolation topology, and has the advantages of high integration, high efficiency and good safety.

Description

Technical Field

[0001] This invention belongs to the field of solar power generation and energy storage technology, and specifically discloses a combined photovoltaic energy storage system. Background Technology

[0002] With the deepening of the global energy transition, photovoltaic (PV) power generation has become an important component of renewable energy. However, PV power generation suffers from inherent drawbacks such as intermittency and volatility, making it difficult to meet users' demands for a stable power supply. To overcome this problem, the industry has proposed a "PV + energy storage" technical approach, which uses energy storage systems to smooth out fluctuations in PV output and achieve the spatiotemporal transfer of electricity.

[0003] Currently, the main technical shortcomings of photovoltaic and energy storage systems on the market are as follows:

[0004] First, the integration is limited and the space utilization is low. Traditional photovoltaic and energy storage systems typically arrange components such as photovoltaic inverters, energy storage converters, and battery systems separately, with the equipment connected by cables. This results in problems such as large footprint, complex wiring, and reduced system efficiency.

[0005] Second, power coupling interference is severe. Most existing integrated photovoltaic and energy storage systems use a DC-coupled scheme, directly connecting the photovoltaic modules and energy storage batteries to the same DC bus. Photovoltaic modules are voltage source devices, and their output power fluctuates drastically with sunlight; energy storage batteries require stable charging and discharging voltage and current. The coupling of two DC sources with vastly different characteristics through the DC bus inevitably introduces power coupling interference that is difficult to eliminate, making it difficult to exceed 90% in energy conversion efficiency and shortening battery cycle life.

[0006] Third, fault propagation is not isolated. In traditional DC coupling schemes, when a short circuit or overvoltage fault occurs at either the photovoltaic side or the battery side, the fault current will directly impact the other port via the shared DC bus, triggering a cascading fault.

[0007] Fourth, the thermal management solutions are simplistic and lack sufficient heat dissipation efficiency. Existing integrated photovoltaic and energy storage systems mostly use air cooling, which suffers from uneven heat dissipation, high energy consumption, and significant performance degradation in high-temperature environments.

[0008] Therefore, it is necessary to invent a combined photovoltaic energy storage system to solve the above problems. Summary of the Invention

[0009] To overcome the aforementioned deficiencies in the prior art, this invention provides a combined photovoltaic energy storage system, comprising a photovoltaic input module, a high-frequency isolated power conversion module, an energy storage battery module, a grid-connected inverter output module, and a control module. The high-frequency isolated power conversion module internally includes a three-winding high-frequency transformer, enabling magnetic coupling power transfer and electrical isolation between the photovoltaic input, energy storage battery, and grid-connected output ports. Each winding of the three-winding transformer has a preset leakage inductance, which, together with corresponding capacitors, forms an LC filter circuit to filter out high-frequency ripple and suppress harmonics. The control module monitors the status of each module in real time and dynamically adjusts the switching frequency and phase shift angle. This three-port high-frequency isolation topology fundamentally solves the power coupling interference and fault propagation problems in traditional DC coupling schemes, effectively addressing the issues mentioned in the background art.

[0010] To achieve the above objectives, the present invention provides the following technical solution: a combined photovoltaic energy storage system, comprising:

[0011] Photovoltaic input module: used to receive DC power output from photovoltaic modules, perform maximum power point tracking, and output the first DC power.

[0012] The high-frequency isolated power conversion module has its power input terminal connected to the power output terminal of the photovoltaic input module to receive the first DC power; its first power output terminal connected to the power input terminal of the grid-connected inverter output module; and its second bidirectional power port connected to the bidirectional power port of the energy storage battery module. The high-frequency isolated power conversion module contains a three-winding high-frequency transformer. The first, second, and third windings of the three-winding high-frequency transformer are respectively connected to the photovoltaic input port, the energy storage battery port, and the grid-connected output port. Power is transferred between the three windings through magnetic coupling, and electrical isolation is achieved between the three ports.

[0013] Energy storage battery module: Its bidirectional power port is connected to the second bidirectional power port of the high-frequency isolated power conversion module, and is used to receive the charging energy output by the high-frequency isolated power conversion module for storage, or to output the discharging energy to the high-frequency isolated power conversion module;

[0014] Grid-connected inverter output module: Its power input terminal is connected to the first power output terminal of the high-frequency isolated power conversion module, and is used to receive the DC power output by the high-frequency isolated power conversion module, convert it into AC power and connect it to the power grid or supply it to the local load;

[0015] Control module: Its signal acquisition terminals are respectively connected to the voltage sampling terminal of the photovoltaic input module, the state of charge sampling terminal of the energy storage battery module, the power sampling terminal of the grid-connected inverter output module, and the voltage sampling terminal of the grid connection point, for real-time monitoring of photovoltaic input power, energy storage battery state of charge, load power, and grid status; its control signal output terminal is connected to the switching control terminal of the high-frequency isolated power conversion module, for dynamically adjusting the switching frequency and phase shift angle of the high-frequency isolated power conversion module according to the monitoring results, so as to adjust the power distribution among the three ports.

[0016] The technical effects and advantages of this invention are as follows:

[0017] 1. Innovative three-port high-frequency isolation topology completely solves power coupling interference and achieves intrinsic safety. A single three-winding high-frequency transformer is used to achieve magnetic coupling and electrical isolation between the photovoltaic, battery, and grid connection ports. The power of each port is independently controlled, eliminating the impact of photovoltaic fluctuations on the battery and avoiding multi-stage conversion losses. At the same time, when any port is short-circuited, the transformer's magnetic circuit automatically limits the current, and together with a solid-state circuit breaker, achieves microsecond-level fault isolation, while non-faulty ports continue to operate normally.

[0018] 2. Utilizing transformer leakage inductance for integrated filtering eliminates the need for additional components, thereby increasing power density. By precisely designing the windings and core air gap, the transformer windings achieve a predetermined leakage inductance, naturally forming an LC filter network with existing capacitors in the circuit. This simultaneously performs power conversion and ripple / harmonic suppression, eliminating the need for independent filter inductors and capacitors, thus reducing size and cost.

[0019] 3. The synergy between photovoltaic panel-assisted heat dissipation and self-powered operation significantly reduces energy consumption and extends off-grid backup time. The photovoltaic backsheet is thermally connected to the heat dissipation fins of the energy storage cabinet, utilizing the photovoltaic modules' own heat dissipation capacity to passively remove heat from the batteries and power modules, reducing active heat dissipation power consumption. At the same time, a photovoltaic self-powered auxiliary power supply is set up, which directly supplies power when there is sunlight, and switches to intermittent battery power supply when there is no sunlight, resulting in extremely low standby power consumption.

[0020] 4. Intelligent energy management and modular parallel operation enhance economy and scalability. Based on model predictive control, power allocation is optimized by integrating photovoltaic forecasting, load forecasting, electricity price signals, and battery status to improve self-consumption rate and economic benefits; it supports parallel networking of multiple devices and cloud collaboration to achieve remote monitoring, fault early warning, and strategy iteration, flexibly adapting to residential, industrial, and commercial scenarios. Attached Figure Description

[0021] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of the overall structure of the present invention.

[0023] Figure 2 This is a circuit diagram of the three-winding transformer and filter of the present invention.

[0024] Figure 3 This is a flowchart of the parameter adaptive control strategy of the present invention.

[0025] Figure 4 This is a hybrid energy storage topology diagram of the present invention.

[0026] Figure 5 This is a schematic diagram of the heat dissipation structure of the present invention.

[0027] Figure 6 This is a flowchart of the Model Predictive Control (MPC) process of the present invention. Detailed Implementation

[0028] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] This invention provides a combined photovoltaic energy storage system, comprising:

[0030] Photovoltaic input module: used to receive DC power output from photovoltaic modules, perform maximum power point tracking, and output the first DC power.

[0031] The high-frequency isolated power conversion module has its power input terminal connected to the power output terminal of the photovoltaic input module to receive the first DC power; its first power output terminal connected to the power input terminal of the grid-connected inverter output module; and its second bidirectional power port connected to the bidirectional power port of the energy storage battery module. The high-frequency isolated power conversion module contains a three-winding high-frequency transformer. The first, second, and third windings of the three-winding high-frequency transformer are respectively connected to the photovoltaic input port, the energy storage battery port, and the grid-connected output port. Power is transferred between the three windings through magnetic coupling, and electrical isolation is achieved between the three ports.

[0032] Energy storage battery module: Its bidirectional power port is connected to the second bidirectional power port of the high-frequency isolated power conversion module, and is used to receive the charging energy output by the high-frequency isolated power conversion module for storage, or to output the discharging energy to the high-frequency isolated power conversion module;

[0033] Grid-connected inverter output module: Its power input terminal is connected to the first power output terminal of the high-frequency isolated power conversion module, and is used to receive the DC power output by the high-frequency isolated power conversion module, convert it into AC power and connect it to the power grid or supply it to the local load;

[0034] Control module: Its signal acquisition terminals are respectively connected to the voltage sampling terminal of the photovoltaic input module, the state of charge sampling terminal of the energy storage battery module, the power sampling terminal of the grid-connected inverter output module, and the voltage sampling terminal of the grid connection point, for real-time monitoring of photovoltaic input power, energy storage battery state of charge, load power, and grid status; its control signal output terminal is connected to the switching control terminal of the high-frequency isolated power conversion module, for dynamically adjusting the switching frequency and phase shift angle of the high-frequency isolated power conversion module according to the monitoring results, so as to adjust the power distribution among the three ports.

[0035] Example 1: System Overall Structure

[0036] This embodiment provides a combined photovoltaic energy storage system, such as Figure 1 As shown, it includes a photovoltaic input module, a high-frequency isolated power conversion module, an energy storage battery module, a grid-connected inverter output module, and a control module.

[0037] The photovoltaic input module is used to connect to the photovoltaic module array, receive the DC power output from the photovoltaic modules, perform maximum power point tracking (MPPT), and output a stable first DC power. The positive and negative output terminals of the photovoltaic input module are respectively connected to the DC input terminal of the high-frequency isolated power conversion module.

[0038] The high-frequency isolated power conversion module has one power input terminal, one first power output terminal, and one second bidirectional power port. Its power input terminal is connected to the output terminal of the photovoltaic input module; the first power output terminal is connected to the input terminal of the grid-connected inverter output module; and the second bidirectional power port is connected to the bidirectional power port of the energy storage battery module. Internally, the high-frequency isolated power conversion module includes a three-winding high-frequency transformer and three bidirectional power conversion circuits to achieve magnetic coupling power transfer and electrical isolation between the three ports.

[0039] The energy storage battery module uses a lithium-ion battery pack, with its positive and negative terminals connected to the positive and negative terminals of the second bidirectional power port of the high-frequency isolated power conversion module, respectively. The energy storage battery module integrates a battery management system (BMS), which transmits information such as battery voltage, current, temperature, state of charge (SOC), and state of health (SOH) to the control module via CAN communication or analog signal lines.

[0040] The DC input terminal of the grid-connected inverter output module is connected to the first power output terminal of the high-frequency isolated power converter module, and its AC output terminal is connected to the power grid or local load. The grid-connected inverter output module adopts a single-phase or three-phase full-bridge inverter topology and is connected to the grid through an LCL filter.

[0041] The control module is based on a digital signal processor (DSP). Its signal acquisition terminals are connected to the output terminals of the photovoltaic input module, the BMS communication interface of the energy storage battery module, the AC output terminal of the grid-connected inverter output module, and the grid connection point via voltage and current sensors, respectively. The control signal output terminals of the control module are connected to the gates of each power switch in the high-frequency isolated power conversion module via drive circuits.

[0042] It should be further explained that the second bidirectional power port and the bidirectional power port are bidirectional ports, which can both output charging energy to the energy storage battery module and receive discharging energy from the energy storage battery module.

[0043] Furthermore, in the above technical solution, the three-winding high-frequency transformer has a first leakage inductance between the first and third windings, and a second leakage inductance between the second and third windings. The first and second leakage inductances are obtained at predetermined values ​​by adjusting the winding method and / or the air gap of the core of the three-winding high-frequency transformer. The first leakage inductance and the output capacitor of the photovoltaic input module together constitute an LC filter circuit. The second leakage inductance and the input capacitor of the grid-connected inverter output module together constitute an LC filter circuit. The predetermined value range of the first leakage inductance is 10μH~20μH, and the predetermined value range of the second leakage inductance is 8μH~15μH. The specific values ​​are determined according to the switching frequency and the allowable ripple current. The high-frequency isolated power conversion module distributes the first DC power to the first power output terminal and / or the second bidirectional power port via magnetic coupling by adjusting the phase shift angle and switching frequency of each port, or transfers the discharge energy input by the energy storage battery module through the second bidirectional power port to the first power output terminal.

[0044] Example 2: Specific Implementation of a Three-Winding High-Frequency Transformer and Leakage Inductance Filter Circuit

[0045] This embodiment details the three-winding high-frequency transformer and its leakage inductance filtering structure in the high-frequency isolated power conversion module, such as... Figure 2 As shown.

[0046] The three-winding high-frequency transformer T1 uses a ferrite core (model EE65 or larger). An air gap is formed in the core column, the length of which is determined by the required leakage inductance value, typically ranging from 0.3mm to 1.5mm. The three windings are wound as follows:

[0047] The first primary winding N1 is wound on the left column of the magnetic core and is used to connect to the photovoltaic input port.

[0048] The second primary winding N2 is wound on the right post of the magnetic core and is used to connect to the energy storage battery port.

[0049] The third primary winding N3 is wound on the center post of the magnetic core and is used to connect to the grid-connected inverter output port.

[0050] The number of turns in each winding is determined based on the system voltage level. Taking a 48V battery system with a photovoltaic input voltage range of 60V~150V and a grid-connected output voltage of 400V (DC bus) as an example, N1=20 turns, N2=20 turns, and N3=80 turns. All windings use multi-strand enameled wire wound in parallel to reduce high-frequency skin effect losses.

[0051] By adjusting the coupling distance between N1 and N3 (i.e., the distance between the left column and the center column) and the core air gap, the leakage inductance Lk13 between N1 and N3 is made to 15μH±10%. By adjusting the coupling distance between N2 and N3, the leakage inductance Lk23 between N2 and N3 is made to 12μH±10%. The design basis for the above leakage inductance values ​​is: based on the switching frequency fs and the allowable high-frequency ripple current ΔI = 5% of the photovoltaic rated current, the formula Lk = Vpv is used. 2 The value of leakage inductance is calculated by / (2π·fs·ΔI). The method for measuring leakage inductance is to short-circuit the corresponding winding and measure the inductance of the other winding.

[0052] The first leakage inductance Lk13 and the output filter capacitor C1 of the photovoltaic input module (e.g., a 10μF / 600V CBB capacitor) together form an LC low-pass filter. The cutoff frequency of this filter... When Lk13 = 15μH and C1 = 10μF, the cutoff frequency is approximately 13kHz. The DC current output by the photovoltaic module contains high-frequency ripple (typically between 10kHz and 100kHz) caused by MPPT disturbances and changes in illumination. This LC filter can effectively attenuate these ripples.

[0053] The second leakage inductance Lk23, together with the input filter capacitor C2 (e.g., a 20μF / 800V CBB capacitor) of the grid-connected inverter output module, forms an LC low-pass filter. Cutoff frequency. When Lk23 = 12μH and C2 = 20μF, the cutoff frequency is approximately 10.3kHz. This filter is used to suppress current harmonics generated by the switching frequency of the grid-connected inverter (e.g., 50kHz), reducing the total harmonic distortion (THD) of the grid-connected current to below 3%.

[0054] In the above design, transformer T1 simultaneously performs power conversion and filtering functions, eliminating the need for an additional independent filter inductor, thereby reducing system size and cost.

[0055] The high-frequency isolated power conversion module includes three bidirectional power conversion circuits, denoted as the first power conversion circuit S1, the second power conversion circuit S2, and the third power conversion circuit S3. Specifically, S1 connects the photovoltaic input module to the first primary winding N1, S2 connects the energy storage battery module to the second primary winding N2, and S3 connects the grid-connected inverter output module to the third primary winding N3.

[0056] Furthermore, in the above technical solution, the high-frequency isolated power conversion module adopts a dual active bridge topology, and each arm of the dual active bridge topology is composed of multiple power switches; the control module includes a parameter adaptive controller, which collects in real time the maximum power point tracking voltage of the photovoltaic input module, the optimal charging voltage of the energy storage battery module, and the grid voltage requirement at the grid-connected output terminal, and dynamically adjusts the switching frequency and phase shift angle of the high-frequency isolated power conversion module according to the collection results; the power switches of the high-frequency isolated power conversion module utilize the energy stored in the first leakage inductance, the second leakage inductance, and the magnetizing inductance of the three-winding high-frequency transformer to make the junction capacitance voltage resonate to zero during the turn-off period of the power switches, thereby achieving zero-voltage turn-on.

[0057] Example 3: Implementation of a parameter adaptive controller

[0058] This embodiment describes the parameter adaptive controller in the control module, and the control flow is as follows: Figure 3 As shown.

[0059] The high-frequency isolated power conversion module adopts a dual active bridge (DAB) topology. Each bridge arm in this topology consists of multiple power switches (e.g., MOSFETs (Metal-Oxide-Semiconductor Field-Effect Transistors) or IGBTs (Insulated-Gate Bipolar Transistors)). These power switches are driven by the PWM signal from the control module, achieving high-frequency switching to control power transfer. The parameter adaptive controller is based on a DSP, utilizing its high-precision ADC module and PWM module. The specific implementation steps are as follows:

[0060] Step A: Signal Acquisition

[0061] The photovoltaic input port voltage Vpv is acquired using a voltage divider circuit and an isolation operational amplifier, while the photovoltaic input current Ipv is acquired using a current sensor. The instantaneous power Ppv is then calculated. The MPPT algorithm employs the perturbation-observation method and outputs a reference voltage Vpv. ref .

[0062] The battery pack's total voltage (Vbat) and state of charge (SOC) are read from the BMS. Based on the charging curve provided by the battery manufacturer, the optimal charging voltage (Vbc) corresponding to the current SOC is obtained by referring to a table. ref For example, the charging voltage of a lithium iron phosphate battery is 51.5V when SOC=30%, 55.2V when SOC=80%, and 56.8V when SOC=95% (constant voltage charging stage).

[0063] The effective value of the grid voltage Vgrid is collected through a voltage transformer. rms And phase θgrid.

[0064] Step B: Calculate the optimal switching frequency and phase shift angle

[0065] The three-winding high-frequency transformer employs a phase-shifted full-bridge or dual active bridge (DAB) modulation method. The switching frequency fs and the phase shift angle φ satisfy the following relationship:

[0066] Transmission power P = (V1·V2·φ·(π-φ)) / (2π) 2 ·fs·Lk)

[0067] Where V1 and V2 are port voltages, and Lk is the equivalent leakage inductance. To adapt to a wide voltage range, the controller employs the following adaptive algorithm:

[0068] When Vpv is below a set threshold (e.g., 60V), reduce the switching frequency to 50kHz to maintain sufficient power delivery capability.

[0069] When Vpv is higher than 120V, increase the switching frequency to 150kHz to reduce core loss and switching loss.

[0070] The phase shift angle φ is dynamically adjusted according to the required power command, ranging from 0 to 180°.

[0071] The controller internally stores a two-dimensional lookup table, with inputs Vpv and Vbat, and outputs the optimal switching frequency fs. opt And the phase shift angle range. During operation, the controller also monitors the transformer core temperature in real time (through the embedded NTC thermistor). If the temperature exceeds 90°C, it will automatically reduce the switching frequency by 20% to reduce losses.

[0072] Step C: Soft switching implementation

[0073] In a dual active bridge topology, the energy stored in the transformer leakage inductances Lk13 and Lk23 and the magnetizing inductance Lm is used to make the junction capacitance voltage resonate to zero during the turn-off period of the switching transistor, achieving zero-voltage turn-on (ZVS). The specific condition is: Lk·Ipeak 2 ≥2·Coss·Vdc 2Where Ipeak is the peak value of the leakage inductance current, Coss is the output capacitance of the switching transistor, and Vdc is the bus voltage. By adjusting the phase shift angle and dead time (typically 200ns~500ns), the drain-source voltage of each switching transistor is reduced to near 0V before turn-on. The dead time must be set longer than the time required for the junction capacitance voltage to resonate to zero.

[0074] Furthermore, in the above technical solution, the energy storage battery module includes a hybrid energy storage unit, which is composed of a main energy storage branch and an auxiliary energy storage branch connected in parallel. The main energy storage branch uses a lithium iron phosphate battery pack for long-term energy storage, and the auxiliary energy storage branch uses a supercapacitor pack or a lithium titanate battery pack for high-frequency power absorption. The auxiliary energy storage branch absorbs high-frequency pulsating power from the photovoltaic output through the second bidirectional power port of the high-frequency isolated power conversion module. After the high-frequency component of the photovoltaic power is extracted by the control module, the high-frequency pulsating power is dynamically distributed to the auxiliary energy storage branch by adjusting the phase shift angle of the high-frequency isolated power conversion module, thereby suppressing the impact of power fluctuations on the main energy storage branch.

[0075] Example 4: Specific Implementation of Hybrid Energy Storage Unit

[0076] This embodiment describes a hybrid energy storage unit in an energy storage battery module, such as... Figure 4 As shown.

[0077] The hybrid energy storage unit consists of a main energy storage branch and an auxiliary energy storage branch connected in parallel.

[0078] Main energy storage branch: Utilizing 16 series-connected lithium iron phosphate cells (model: IFR32140, nominal voltage 3.2V, capacity 15Ah, internal resistance ≤5mΩ) to form a 51.2V / 15Ah battery pack. Multiple such battery packs can be connected in parallel according to capacity requirements. The battery pack is equipped with an active balancing BMS with a balancing current of 1A.

[0079] Auxiliary energy storage branch: Six supercapacitors (Model: Maxwell BCAP3000, 2.7V / 3000F) are connected in series, with a rated voltage of 16.2V and a capacitance of 500F (after series connection). The supercapacitor bank is connected in parallel with the main energy storage branch through a bidirectional DC / DC converter (e.g., half-bridge topology, switching frequency 100kHz). The control signal for this DC / DC converter is provided by the control module to regulate the charging and discharging current of the supercapacitors.

[0080] The work process is as follows:

[0081] The photovoltaic output power experiences high-frequency fluctuations (frequency 0.1Hz~10Hz, amplitude up to 50% of rated power) due to cloud cover. The control module extracts the fluctuation component through a high-pass filter and controls the bidirectional DC / DC converter to allow the supercapacitor bank to absorb or release this fluctuating power. The auxiliary energy storage branch absorbs the high-frequency pulsating power from the photovoltaic output through the second bidirectional power port of the high-frequency isolated power converter module. This high-frequency pulsating power is extracted as the high-frequency component of the photovoltaic power by the control module and dynamically distributed to the auxiliary energy storage branch by adjusting the phase shift angle of the high-frequency isolated power converter module. Reference value for fluctuating power. Ppv lp Let be the value of the photovoltaic power after passing through a first-order low-pass filter. The transfer function of this low-pass filter is: The time constant τ = 1.59 seconds (corresponding to a cutoff frequency of 0.1Hz). The low-pass filter can be implemented using a first-order hysteresis algorithm in a digital signal processor. , where α=T s / (T s +τ), T s The sampling period.

[0082] The response time of a supercapacitor is less than 5ms, while that of a lithium iron phosphate battery is approximately 100ms. By using a supercapacitor to handle high-frequency components, frequent charging and discharging of the battery can be avoided, thereby extending the battery's cycle life.

[0083] When the voltage of the supercapacitor bank deviates from the set range (e.g., 8V~16.2V), the bidirectional DC / DC converter automatically replenishes or recovers energy from the main energy storage branch to maintain the voltage of the supercapacitor bank near the intermediate value (12V).

[0084] Furthermore, the above technical solution also includes a photovoltaic panel-integrated thermal energy storage passive heat dissipation module. This module includes a thermally conductive medium layer attached to the backsheet of the photovoltaic module and a heat dissipation fin array disposed on the top of the energy storage cabinet. The thermally conductive medium layer is in thermal contact with the heat dissipation fin array, and the heat dissipation fin array is in thermal contact with the heat dissipation substrate of the energy storage battery module and the high-frequency isolated power conversion module. When the system is running, the heat generated by the high-frequency isolated power conversion module and the energy storage battery module is conducted to the backsheet of the photovoltaic module through the heat dissipation fin array and the thermally conductive medium layer, and then dissipated into the environment by the photovoltaic module.

[0085] Example 5: Photovoltaic panel-integrated thermal storage passive heat dissipation structure

[0086] This embodiment describes a passive heat dissipation module, such as... Figure 5 As shown.

[0087] Photovoltaic module backsheets are typically aluminum frames or polymer backsheets. In this embodiment, a heat-conducting structure is installed between the photovoltaic module backsheet and the top of the energy storage cabinet, specifically including:

[0088] Thermal conductive layer: High thermal conductivity silicone pads are used. One side of the silicone pad is attached to the photovoltaic backsheet, and the other side is attached to the heat dissipation fin array. The silicone pads are compressible and can absorb assembly tolerances.

[0089] Heat sink fin array: Extruded aluminum fins, substrate thickness 5mm, fin height 40mm, fin spacing 6mm, length the same as the width of the energy storage cabinet. The fin substrate contacts the heat sink substrate inside the energy storage cabinet via thermal grease. The heat sink substrate connects to the heat dissipation surface of the power module (such as IGBT or MOSFET) and the aluminum shell of the battery module.

[0090] Heat pipe assistance (optional): A flat heat pipe (6mm in diameter, with the evaporation section attached to the power module and the condensation section embedded in the fin substrate) is embedded between the power module and the heat sink fins to further reduce thermal resistance.

[0091] Heat transfer path: Power module heats up → heat pipe → heat sink fin substrate → fins → thermally conductive silicone pad → photovoltaic backsheet → photovoltaic module radiates / convection heat to the environment.

[0092] Furthermore, the above technical solution also includes a photovoltaic self-powered auxiliary power supply module. The input end of this module is connected to the high-voltage DC bus of the photovoltaic input module, and the output end is connected to the power supply end of the control module. The photovoltaic self-powered auxiliary power supply module has a first power supply path and a second power supply path. The first power supply path draws power from the photovoltaic DC bus, and the second power supply path draws power from the energy storage battery module. When the photovoltaic input module has power output, the first power supply path is turned on and the second power supply path is turned off. When the photovoltaic input module has no power output, the first power supply path is turned off and the second power supply path is turned on intermittently.

[0093] Example 6: Photovoltaic self-powered auxiliary power supply

[0094] This embodiment describes a photovoltaic self-powered auxiliary power supply module.

[0095] The auxiliary power supply provides low-voltage DC (+12V, +5V, +3.3V) to the control module, drive circuit, communication module, etc. Its input has two sources:

[0096] The first power supply path draws power from the photovoltaic input bus. A flyback DC / DC converter is used, with an input voltage range of 60V~450V and an output of 15V / 10W. The flyback transformer design features a primary inductance of 1mH, a turns ratio of 10:1, and a switching frequency of 100kHz. The control chip employs a flyback controller with integrated power MOSFETs. This path automatically activates when the photovoltaic input voltage exceeds 60V, simultaneously outputting a voltage detection signal (Vpv). ok (To the control module.)

[0097] The second power supply path draws power from the energy storage battery module, passing through an isolated DC / DC converter (input voltage 40V~60V, output 15V / 5W). This DC / DC converter is controlled by the EN signal from the control module. When EN=0, the converter is in sleep mode, with a quiescent current <10μA. When EN=1, the converter operates in burst mode, waking up every 30 seconds and outputting a 100ms pulse to power the control module, allowing the control module to quickly detect the system status. If no abnormality is found, it returns to sleep mode.

[0098] Work logic:

[0099] When Vpv ok When the light is sufficient, the control module disconnects the second power supply path (EN=0) and is powered by the first power supply path.

[0100] When Vpv ok When the battery is ineffective and the SOC is greater than 10%, the control module outputs an EN=1 pulse every 30 seconds for 100ms. During this period, the second power supply path provides power. After waking up, the control module (or dedicated low-power standby MCU) performs the following operations:

[0101] The first 50ms: waiting for the system clock to stabilize and peripherals to initialize;

[0102] The middle 30ms: Collect key parameters such as battery voltage, temperature, and SOC;

[0103] The last 20ms: Check the status. If there is no abnormality, immediately enter deep sleep; if a fault is detected, remain awake and issue an alarm.

[0104] To avoid increased power consumption due to frequent wake-ups of the main DSP, this system uses an independent ultra-low power MCU in standby mode to handle periodic wake-ups and status monitoring. The main DSP is only woken up when an anomaly is detected or when a charging / discharging operation is required.

[0105] Furthermore, the above technical solution also includes a fault isolation and fast protection module. This module includes three solid-state circuit breakers connected in series between the three ports of the high-frequency isolation power conversion module and the corresponding power conversion circuits, and a fault detection unit. The fault detection unit monitors the voltage and current of each port and the flux change rate of the three-winding high-frequency transformer in real time. The flux change rate is monitored by adding an auxiliary winding to the three-winding high-frequency transformer. The induced voltage of the auxiliary winding is proportional to the flux change rate. When the induced voltage change rate exceeds a preset threshold, it is determined to be a flux change. When a short-circuit fault occurs at any port, the fault detection unit triggers the corresponding solid-state circuit breaker to open after detecting the fault characteristics, and at the same time uses the magnetic circuit of the three-winding high-frequency transformer to limit the propagation of the fault current to other ports.

[0106] Example 7: Fault Isolation and Fast Protection Circuit

[0107] This embodiment describes the fault isolation and rapid protection module.

[0108] The module includes three solid-state circuit breakers Q1, Q2, and Q3, and a fault detection unit DET.

[0109] Solid-state circuit breakers: Each solid-state circuit breaker consists of two power MOSFETs connected in reverse series and a drive circuit. During normal conduction, both MOSFETs are turned on; during fault disconnection, both MOSFETs are turned off. The on-resistance of the solid-state circuit breaker is less than 2mΩ, the rated current is 200A, and the turning time is less than 100ns.

[0110] Fault detection unit: Implemented using a high-speed comparator and an FPGA (Field-Programmable Gate Array) (or CPLD (Complex Programmable Logic Device)). Detection signals include:

[0111] Current at each port (via shunt or Hall sensor, bandwidth > 1MHz).

[0112] Voltage at each port.

[0113] The induced voltage on the transformer's auxiliary winding Naux (reflecting the rate of change of magnetic flux in the core). The auxiliary winding typically has 10 turns, and the output voltage is fed into a comparator after being filtered by an RC filter. The rate of change of magnetic flux is dΦ / dt = Vaux / N. aux Where Vaux is the induced voltage of the auxiliary winding, N aux The number of turns in the auxiliary winding is used. When dΦ / dt exceeds the normal range by ±50%, it is considered a sudden change in magnetic flux.

[0114] Fault diagnosis logic:

[0115] Overcurrent fault: When the current at any port exceeds twice the rated value for more than 5μs, the solid-state circuit breaker corresponding to that port is triggered to turn off.

[0116] Short circuit fault: A two-level criterion is adopted to improve anti-interference capability.

[0117] Level 1: When the current rise rate di / dt is detected to be >50A / μs (this value can be reliably measured by a conventional Hall sensor) and lasts for more than 2μs, the pre-judgment flag is triggered;

[0118] Level 2: Simultaneously detect the rate of change of induced voltage dΦ / dt of the transformer auxiliary winding. If it exceeds the normal range of ±50%, it is determined to be a port short circuit.

[0119] When both conditions are met simultaneously, the corresponding solid-state circuit breaker is shut off within 20μs. If only the first level is met but the second level is not triggered, it is determined that the load is starting normally or there is external interference, and no protection action is performed, but the event is recorded for subsequent analysis. Through the above two-level criteria, the detection sensitivity can be adjusted to a practically feasible range (50A / μs~200A / μs) while avoiding false triggering.

[0120] Overvoltage fault: When the port voltage exceeds 1.3 times the rated value, the corresponding solid-state circuit breaker will be shut down.

[0121] Because the ports of the three-winding high-frequency transformer are magnetically isolated, when one port is short-circuited, the magnetic flux is limited to the vicinity of the faulty winding, and the induced voltage at other ports only rises briefly (less than twice the rated value), after which the solid-state circuit breaker trips to clear the fault. For example, when a short circuit occurs at the photovoltaic input port, the transformer's magnetic circuit limits the short-circuit current to less than three times the rated current (compared to more than ten times the rated current in traditional non-isolated solutions), and the voltage fluctuation at non-faulty ports is less than 20%.

[0122] Furthermore, in the above technical solution, the power conversion circuits corresponding to the photovoltaic input module, the energy storage battery module, and the grid-connected inverter output module are all bidirectional high-frequency inverter circuits, and they share the same digital signal processor as the core of the control module. The digital signal processor dynamically allocates the power flow between the three ports based on the real-time sampling values ​​of the photovoltaic input power, load power, and energy storage battery state of charge. The dynamic allocation is based on the power balance equation of a three-winding high-frequency transformer: Ppv + Pbat = Pgrid + Ploss, where Ppv is the photovoltaic input power, Pbat is the battery charging and discharging power (discharging is positive, charging is negative), and Pgrid is the grid-connected power. In grid-connected mode, the grid-connected inverter output module is controlled to output an AC current with the same frequency and phase as the grid. In off-grid mode, the grid-connected inverter output module is controlled to establish an independent AC voltage reference and supply power to local critical loads.

[0123] Example 8: Digital Signal Processor and Control Strategy

[0124] This embodiment describes the specific control strategy of the DSP in the control module.

[0125] The DSP model is TMS320F28379D, with a main frequency of 200MHz, and includes three independent high-resolution PWM modules and four 16-bit ADCs. The DSP runs the main program, performing the following tasks in each control cycle (100μs):

[0126] 1. Data Acquisition and Preprocessing

[0127] The photovoltaic voltage Vpv and photovoltaic current Ipv are read by ADC, the average values ​​are calculated, and digital filtering (first-order low-pass, cutoff frequency 1kHz) is applied.

[0128] Battery voltage, current, temperature, SOC, and SOH are read through the BMS.

[0129] The grid-connected voltage Vgrid and grid-connected current Ig are read by the ADC and used for phase-locked loop (PLL) and power calculation.

[0130] Read the load current Iload using the ADC (if in off-grid mode).

[0131] 2. Pattern Judgment

[0132] Detect the grid voltage frequency and amplitude: if the effective voltage value is between 198V and 242V and the frequency is between 49.5Hz and 50.5Hz, it is determined to be in grid-connected mode; otherwise, it is in off-grid mode.

[0133] 3. Power Distribution Calculation

[0134] The power balance equation based on a three-winding high-frequency transformer is: Ppv + Pbat = Pigrid + Ploss, which is approximated as Ppv + Pbat = Pigrid when losses are ignored.

[0135] In grid-connected mode:

[0136] Objective: Prioritize local load, and use any remaining power to connect to the internet or charge the battery.

[0137] Calculate the load power Pload = Vgrid rms *Iload rms .

[0138] If Ppv > Pload, then the excess power Pex = Ppv - Pload. Based on the battery SOC: if SOC < 95%, Pex is allocated to charge the battery; if SOC >= 95%, Pex is fed into the grid.

[0139] If Ppv < Pload, the power deficit Pshort = Pload - Ppv. First, the battery is discharged to supplement (if SOC > 15%), otherwise power is taken from the grid.

[0140] In off-grid mode:

[0141] Control the third power conversion circuit S3 to establish a constant voltage and constant frequency alternating current (220V / 50Hz).

[0142] According to the difference between Ppv and Pload, dynamically adjust the battery charge and discharge. If Ppv > Pload, the excess power is used to charge the battery; if Ppv < Pload, the battery discharges to supplement.

[0143] 4. PWM modulation

[0144] According to the above power distribution instructions, calculate the phase shift angles φ1, φ2, φ3 and the switching frequency fs required for the three bidirectional power conversion circuits. The PWM module of the DSP outputs six (or eight) drive signals, which drive the power switch tubes after passing through the isolation drive chip.

[0145] 5. Communication and monitoring

[0146] [[ID=2,2]]The DSP communicates with the WiFi / 4G module through UART and uploads the operation data to the cloud. At the same time, it receives the scheduling instructions from the cloud (such as electricity price periods, battery charge and discharge strategies, etc.).

[0147] Furthermore, in the above technical solution, the digital signal processor also performs multi-objective model predictive control, and the process includes: constructing a cost function that includes a photovoltaic power prediction model, a load prediction model, a real-time electricity price signal, the state of charge and health state of the energy storage battery; in each control cycle, solving the optimal power distribution sequence that minimizes the cost function; and outputting the first set of control quantities of the sequence to each power conversion circuit.

[0148] Example 9: Specific algorithm of multi-objective model predictive control

[0149] This example further describes the multi-objective model predictive control (MPC) algorithm running in the DSP, and the predictive control process is as Figure 6 shown.

[0150] MPC performs the following steps in each control cycle (Ts = 1ms):

[0151] Step 1: State sampling and prediction <oo00352>

[0152] Obtain the state at the current time k: Ppv(k), SOC(k), Pload(k), Pgrid(k), real-time electricity price Price buy(k) (Electricity purchase price, yuan / kWh) and Price sell (k) (Electricity sales price, yuan / kWh, usually lower than the electricity purchase price).

[0153] Since photovoltaic power and load power vary significantly on a second-level scale but are relatively stable on a millisecond-level scale, the prediction step size is set to N=10, with a step size Δt=0.1 seconds, meaning the prediction targets power changes within the next second. The prediction model employs a simplified linear extrapolation combined with low-pass filtering.

[0154] A photovoltaic power prediction model is used: based on the rate of change of irradiance collected in the most recent second (via a small irradiance sensor), the current trend is linearly extrapolated to the next second, and a first-order inertial element (time constant 0.5 seconds) is superimposed to smooth the fluctuations. The formula is:

[0155]

[0156] Where dPpv / dt is calculated from the difference between the two most recent sampling points.

[0157] Load forecasting model: A simple average based on the same period in history (sliding window of the previous 10 seconds) is used. If a sudden load change is detected (such as starting a motor), the current instantaneous value is used instead of the forecast value.

[0158] Step 2: Construct the cost function

[0159]

[0160] in:

[0161] The first item: electricity purchase cost (when Pgrid>0), weight w1=1.0.

[0162] Second item: Electricity sales revenue (when Pgrid < 0), due to Price sell If the value is positive, this item is negative (minus revenue), encouraging surplus electricity to be sold to the grid. In the actual system, Price... sell Usually Price buy 0.5 to 0.8 times.

[0163] Third item: Battery SOC deviation, target SOC ref = 80%, weight w2=0.5.

[0164] The fourth item is the battery power fluctuation penalty item, which is used to smooth charging and discharging and extend battery life. Its weight is w3=0.2.

[0165] Fifth item: Power balance constraint (soft constraint), penalty coefficient λ=0.8.

[0166] Step 3: Optimize the solution

[0167] Solving for the optimal control sequence using the gradient descent method [Pbat] opt (k+1),...,Pbat opt [(k+N)] minimizes J. Since the prediction time domain is only 1 second (N=10), the variable dimensionality is low, and gradient descent can complete 20 iterations within a 1ms control period, meeting the real-time requirements.

[0168] To prevent drastic fluctuations in the optimization results, a first-order low-pass filter (cutoff frequency 10Hz) is added to the battery power command. Constraints are also set as follows:

[0169] Battery charge / discharge power limits: ;

[0170] SOC limit: (For example, 10% to 95%).

[0171] Step 4: Implement the first control variable

[0172] The optimal battery power instruction Pbat at time k+1 opt The (k+1) value is output to the second power conversion circuit S2, which adjusts its phase shift angle to achieve the required power. The remaining predicted values ​​(k+2…k+N) are used for rolling optimization in the next cycle, but are not directly output.

[0173] Additional notes: Offline pre-calculation and online fast correction

[0174] To reduce the computational burden on the online system, the optimal control table for typical operating conditions is pre-stored in the DSP's Flash memory. The input is (Ppv, Pload, SOC), and the output is Pbat. During online operation, the initial solution is first obtained by looking up the table, and then convergence is achieved quickly after 1-2 gradient corrections.

[0175] Furthermore, in the above technical solution, the system supports the parallel operation of multiple devices, with each device exchanging operating status and power commands through a high-speed communication bus, and the control module of one of the devices acting as the host to coordinate and allocate cluster power, or unified scheduling by a cloud platform.

[0176] Example 10: Implementation of multi-machine parallel operation

[0177] This embodiment describes a method for operating multiple systems in parallel.

[0178] When the load power exceeds the rated power of a single unit (e.g., 5kW), multiple units of this system can be connected in parallel to the same AC bus. Each system is interconnected via CAN bus or RS485. The communication protocol is defined as follows:

[0179] Master: Address 0x01, responsible for collecting data from slaves and issuing cluster power commands.

[0180] Slave device: Address 0x02~0x0F, reports its own status (operating mode, current power, SOC, fault flag).

[0181] Cluster control strategy:

[0182] After the system is powered on, each device determines the host through a competition mechanism (the device with the smallest address becomes the host).

[0183] The host broadcasts a synchronization frame every 100ms, requiring all slave devices to report their status.

[0184] The host calculates the power that should be allocated to each machine based on the total load power and the total photovoltaic power, and sends power commands to each slave machine.

[0185] When the host fails, the remaining devices will automatically elect a new host.

[0186] Power allocation principle: Prioritize discharging batteries with higher SOC and charging batteries with lower SOC to achieve balanced aging of each battery pack.

[0187] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A combined photovoltaic energy storage system, characterized in that, include: Photovoltaic input module: used to receive DC power output from photovoltaic modules, perform maximum power point tracking, and output the first DC power. The high-frequency isolated power conversion module has its power input terminal connected to the power output terminal of the photovoltaic input module to receive the first DC power; its first power output terminal connected to the power input terminal of the grid-connected inverter output module; and its second bidirectional power port connected to the bidirectional power port of the energy storage battery module. The high-frequency isolated power conversion module contains a three-winding high-frequency transformer. The first, second, and third windings of the three-winding high-frequency transformer are respectively connected to the photovoltaic input port, the energy storage battery port, and the grid-connected output port. Power is transferred between the three windings through magnetic coupling, and electrical isolation is achieved between the three ports. Energy storage battery module: Its bidirectional power port is connected to the second bidirectional power port of the high-frequency isolated power conversion module, and is used to receive the charging energy output by the high-frequency isolated power conversion module for storage, or to output the discharging energy to the high-frequency isolated power conversion module; Grid-connected inverter output module: Its power input terminal is connected to the first power output terminal of the high-frequency isolated power conversion module, and is used to receive the DC power output by the high-frequency isolated power conversion module, convert it into AC power and connect it to the power grid or supply it to the local load; Control module: Its signal acquisition terminals are respectively connected to the voltage sampling terminal of the photovoltaic input module, the state of charge sampling terminal of the energy storage battery module, the power sampling terminal of the grid-connected inverter output module, and the voltage sampling terminal of the grid connection point, for real-time monitoring of photovoltaic input power, energy storage battery state of charge, load power, and grid status; its control signal output terminal is connected to the switching control terminal of the high-frequency isolated power conversion module, for dynamically adjusting the switching frequency and phase shift angle of the high-frequency isolated power conversion module according to the monitoring results, so as to adjust the power distribution among the three ports.

2. The integrated photovoltaic energy storage system according to claim 1, characterized in that: The three-winding high-frequency transformer has a first leakage inductance between the first and third windings and a second leakage inductance between the second and third windings. The first and second leakage inductances are obtained at predetermined values ​​by adjusting the winding method and / or the air gap of the core of the three-winding high-frequency transformer. The first leakage inductance and the output capacitor of the photovoltaic input module together form an LC filter circuit. The second leakage inductance and the input capacitor of the grid-connected inverter output module together form an LC filter circuit. The predetermined value range of the first leakage inductance is 10μH~20μH, and the predetermined value range of the second leakage inductance is 8μH~15μH. The specific values ​​are determined according to the switching frequency and the allowable ripple current. The high-frequency isolated power conversion module distributes the first DC power to the first power output terminal and / or the second bidirectional power port via magnetic coupling by adjusting the phase shift angle and switching frequency of each port, or transfers the discharge energy input by the energy storage battery module through the second bidirectional power port to the first power output terminal.

3. The integrated photovoltaic energy storage system according to claim 2, characterized in that: The high-frequency isolated power conversion module adopts a dual active bridge topology, in which each bridge arm consists of multiple power switches. The control module includes a parameter adaptive controller, which collects in real time the maximum power point tracking voltage of the photovoltaic input module, the optimal charging voltage of the energy storage battery module, and the grid voltage requirement at the grid-connected output terminal, and dynamically adjusts the switching frequency and phase shift angle of the high-frequency isolated power conversion module based on the collected results. The power switches of the high-frequency isolated power conversion module utilize the energy stored in the first leakage inductance, the second leakage inductance, and the magnetizing inductance of the three-winding high-frequency transformer to make the junction capacitance voltage resonate to zero during the power switch turn-off period, thereby achieving zero-voltage turn-on.

4. The integrated photovoltaic energy storage system according to claim 1, characterized in that: The energy storage battery module includes a hybrid energy storage unit, which consists of a main energy storage branch and an auxiliary energy storage branch connected in parallel. The main energy storage branch uses a lithium iron phosphate battery pack for long-term energy storage, while the auxiliary energy storage branch uses a supercapacitor pack or a lithium titanate battery pack for high-frequency power absorption. The auxiliary energy storage branch absorbs high-frequency pulsating power from the photovoltaic output through the second bidirectional power port of the high-frequency isolated power conversion module. This high-frequency pulsating power is extracted by the control module from the high-frequency component of the photovoltaic power and dynamically distributed to the auxiliary energy storage branch by adjusting the phase shift angle of the high-frequency isolated power conversion module, thereby suppressing the impact of power fluctuations on the main energy storage branch.

5. A combined photovoltaic energy storage system according to claim 1, characterized in that: It also includes a photovoltaic panel-integrated thermal storage passive heat dissipation module, which includes a thermally conductive medium layer attached to the backsheet of the photovoltaic module and a heat dissipation fin array disposed on the top of the energy storage cabinet; the thermally conductive medium layer is in thermal contact with the heat dissipation fin array, and the heat dissipation fin array is in thermal contact with the heat dissipation substrate of the energy storage battery module and the high-frequency isolated power conversion module; when the system is running, the heat generated by the high-frequency isolated power conversion module and the energy storage battery module is conducted to the backsheet of the photovoltaic module through the heat dissipation fin array and the thermally conductive medium layer, and then dissipated to the environment by the photovoltaic module.

6. A combined photovoltaic energy storage system according to claim 1, characterized in that: It also includes a photovoltaic self-powered auxiliary power supply module. The input end of this module is connected to the high-voltage DC bus of the photovoltaic input module, and the output end is connected to the power supply end of the control module. The photovoltaic self-powered auxiliary power supply module has a first power supply path and a second power supply path. The first power supply path draws power from the photovoltaic DC bus, and the second power supply path draws power from the energy storage battery module. When the photovoltaic input module has power output, the first power supply path is turned on and the second power supply path is turned off. When the photovoltaic input module has no power output, the first power supply path is turned off and the second power supply path is turned on intermittently.

7. A combined photovoltaic energy storage system according to claim 1, characterized in that: It also includes a fault isolation and fast protection module, which includes three solid-state circuit breakers connected in series between the three ports of the high-frequency isolation power conversion module and the corresponding power conversion circuits, and a fault detection unit. The fault detection unit monitors the voltage and current of each port and the flux change rate of the three-winding high-frequency transformer in real time. The flux change rate is monitored by adding an auxiliary winding to the three-winding high-frequency transformer. The induced voltage of the auxiliary winding is proportional to the flux change rate. When the induced voltage change rate exceeds a preset threshold, it is determined to be a flux change. When a short-circuit fault occurs at any port, the fault detection unit triggers the corresponding solid-state circuit breaker to open after detecting the fault characteristics. At the same time, the magnetic circuit of the three-winding high-frequency transformer is used to limit the propagation of the fault current to other ports.

8. A combined photovoltaic energy storage system according to claim 1, characterized in that: The power conversion circuits corresponding to the photovoltaic input module, energy storage battery module, and grid-connected inverter output module are all bidirectional high-frequency inverter circuits, and they share the same digital signal processor as the core of the control module. The digital signal processor dynamically allocates the power flow between the three ports based on the real-time sampled values ​​of photovoltaic input power, load power, and energy storage battery state of charge. The dynamic allocation is based on the power balance equation of a three-winding high-frequency transformer: Ppv + Pbat = Pgrid + Ploss, where Ppv is the photovoltaic input power, Pbat is the battery charging and discharging power (discharging is positive, charging is negative), and Pgrid is the grid-connected power. In grid-connected mode, the grid-connected inverter output module is controlled to output AC current with the same frequency and phase as the grid. In off-grid mode, the grid-connected inverter output module is controlled to establish an independent AC voltage reference and supply power to local critical loads.

9. A combined photovoltaic energy storage system according to claim 8, characterized in that: The digital signal processor also performs multi-objective model predictive control, the process of which includes: constructing a cost function that includes a photovoltaic power prediction model, a load prediction model, a real-time electricity price signal, and the state of charge and health of the energy storage battery; solving for the optimal power allocation sequence that minimizes the cost function in each control cycle; and outputting the first set of control quantities of the sequence to each power conversion circuit.

10. A combined photovoltaic energy storage system according to claim 1, characterized in that: The system supports the parallel operation of multiple devices. The devices exchange operating status and power commands through a high-speed communication bus. The control module of one of the devices acts as the host to coordinate and allocate cluster power, or the cloud platform can perform unified scheduling.

Images