Balcony photovoltaic energy storage integrated system

Abstract

The application discloses a balcony photovoltaic energy storage integrated system, and belongs to the technical field of photovoltaic energy storage, which comprises a photovoltaic energy collection assembly, an energy storage and energy management master control box, a safety interface and protection unit, an intelligent energy control unit and an energy interaction and recovery assembly; the intelligent energy control unit executes an energy scheduling algorithm based on model predictive control, and samples photovoltaic power, battery state of charge, household load power and power grid price in real time, and determines the system working mode by solving a constraint optimization problem; all direct current links are safe and low voltage, and the photovoltaic assembly and the energy storage master control box are connected through a waterproof and foolproof quick plug cable to realize plug and play, so that users can complete hanging and mobile phone power distribution within 10 minutes without professional electricians; meanwhile, the energy storage master control box is compactly integrated with modular devices such as clothes drying racks and plant walls, does not occupy the ground activity space of the balcony, and completely solves the pain points of complex wiring and space waste of the existing decentralized system.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic energy storage technology, and more specifically, to an integrated photovoltaic energy storage system for balconies. Background Technology

[0002] Currently, photovoltaic power generation systems for balconies typically employ a simple grid-connected model of "photovoltaic modules + micro-inverters." A few solutions further integrate independent battery energy storage units. However, existing technologies generally suffer from prominent problems such as low energy conversion efficiency, complex installation and poor space utilization, insufficient safety, and low levels of intelligence. Specifically, in AC-coupled schemes, the photovoltaic DC power needs to be converted to AC by a micro-inverter, then rectified back to DC by an energy storage inverter to charge the battery, and then inverted back to AC during discharge. These multiple AC-DC conversions result in energy losses as high as 15%-25%. While DC-coupled schemes reduce some conversion steps, existing products mostly involve independently distributed photovoltaic panels and energy storage units, leading to cumbersome wiring and occupying valuable balcony space. The DC side often uses high voltage (such as above 400V), which poses serious risks of arcing and electric shock in the humid and vibrating environment of balconies. In addition, existing systems lack active thermal management design, and the high temperature of the photovoltaic backsheet is directly conducted to the battery or inverter, accelerating battery aging and reducing system reliability. Moreover, the controllers generally use simple timing or voltage threshold logic, which cannot dynamically optimize scheduling based on electricity price signals, weather forecasts, and household electricity consumption habits. This means that peak-valley arbitrage cannot be achieved, and there is a lack of intelligent functions such as emergency backup power and anti-reverse current. Furthermore, existing balcony photovoltaic and energy storage products are of a single form and have failed to explore the potential for integrating unique functions of balconies, such as clothes drying, green plant cooling, and community energy sharing. This results in the system being disconnected from users' daily lives, with limited user participation and economic benefits.

[0003] Therefore, we have made improvements to this and proposed an integrated photovoltaic energy storage system for balconies. Summary of the Invention

[0004] In view of the above-mentioned problems in the existing technology, the purpose of this invention is to provide an integrated photovoltaic energy storage system for balconies.

[0005] To solve the above problems, the technical solution adopted by the present invention is as follows: It includes: a photovoltaic energy harvesting component, an energy storage and energy management main control box, a safety interface and protection unit, an intelligent energy control unit, and an energy interaction and recovery component; the intelligent energy control unit executes an energy dispatching algorithm based on model predictive control, samples photovoltaic power, battery state of charge, household load power, and grid electricity price in real time, and determines the system operating mode by solving a constrained optimization problem. The intelligent energy control unit internally sets an objective function. Minimize it: ; in, For real-time grid electricity prices, For power exchange with the power grid, This represents the battery's state of charge (0-1). For the target state of charge, These are the weighting coefficients. For integration variables, To predict the time domain, This refers to the current moment.

[0006] Preferably, the photovoltaic energy harvesting component has longitudinal heat dissipation fins and a rear-mounted ventilation slot on its back, forming an air gap between itself and the mounting bracket, utilizing thermal pressure difference. Driven by natural convection, the thermal pressure difference is calculated by the following formula: ; in, air density, It is the acceleration due to gravity. The vertical height of the ventilation duct. The absolute temperature of the photovoltaic backsheet. The absolute temperature of the environment; when When the flow resistance is less than the required level, the intelligent energy control unit activates the axial fan for forced ventilation. The fan duty cycle is determined by a linear mapping of the back panel temperature: the duty cycle is zero when the back panel temperature is not higher than 55 degrees Celsius, increases linearly from zero to one when the temperature is between 55 and 65 degrees Celsius, and is one when the temperature is not lower than 65 degrees Celsius.

[0007] Preferably, the energy storage and energy management main control box includes a lithium iron phosphate battery cluster, a bidirectional DC-DC converter, and a bidirectional DC-AC inverter. The bidirectional DC-DC converter executes a variable step-size disturbance observation maximum power point tracking algorithm, whose disturbance step size is adaptively adjusted according to the power change rate: the current step size is equal to the maximum step size multiplied by the absolute value of the difference between the current power and the previous power, and then divided by the product of the rated power and the normalization coefficient; when the power fluctuation is greater than 10% of the rated power, the step size is automatically reduced to one-tenth of the original; the maximum power point tracking algorithm is executed once every 20 milliseconds, and is interrupted and switched to constant voltage mode when the voltage drops by more than 5%; the bidirectional DC-AC inverter has a built-in virtual synchronous generator control, and its rotor motion equation is: the moment of inertia multiplied by the differential of the angular frequency is equal to the reference active power divided by the reference angular frequency minus the electromagnetic active power divided by the actual angular frequency minus the damping coefficient multiplied by the difference between the angular frequency and the reference angular frequency; the power angle differential is equal to the angular frequency.

[0008] Preferably, the gravity potential energy recovery type intelligent clothes drying rack includes: a lifting rope, a pulley system, a one-way bearing, a permanent magnet synchronous motor, a reversible clutch device, a tension sensor, and a motor driver; during the descent of the clothes rack crossbar and the total mass of the clothes, the one-way bearing is locked, the gravitational torque drives the motor to generate electricity, and the output power is rectified to charge the battery; the recovered power is equal to the generator efficiency multiplied by the total mass multiplied by the gravitational acceleration and then multiplied by the descent speed; the descent speed is adjusted by a PID controller: the proportional term coefficient multiplied by the difference between the speed set value and the measured value, plus the integral term coefficient multiplied by the error integral, plus the differential term coefficient multiplied by the error differential, and the sum of the three plus the set speed gives the target speed; during the ascent, the clutch disengages, and the user manually pulls the rope; when the tension sensor detects that the tension exceeds the preset threshold and lasts for more than two seconds, the system automatically closes the clutch and starts the electric assist mode, the battery powers the motor to output auxiliary torque, and the auxiliary torque is equal to the assist coefficient multiplied by the difference between the measured tension and the normal ascent tension.

[0009] Preferably, the photovoltaic-storage driven micro-ecological circulation system includes: a plant evaporative cooling wall, a heat exchange duct, a micro PEM electrolyzer, a metal hydride hydrogen storage tank, a micro fuel cell, and a water circulation pipeline; the plant evaporative cooling wall is installed below the photovoltaic modules and consists of a planting box, a capillary water absorption layer, a water storage tank, and a temperature and humidity sensor; plant transpiration lowers the temperature of the air flowing through it, and the temperature drop is equal to the latent heat of vaporization of water multiplied by the transpiration rate, divided by the air density multiplied by the specific heat capacity of the air, and then multiplied by the volumetric flow rate through the plant wall; the heat exchange duct introduces the cool, humid air from the back of the plant wall into the air inlet of the energy storage main control box, and the duct is equipped with an adjustable speed fan and an electric air valve; the intelligent energy control unit adjusts the temperature according to the internal temperature of the energy storage box. Fan speed is controlled by relative humidity: the fan stops when the internal temperature does not exceed 35 degrees Celsius or the relative humidity is not lower than 80%; when the internal temperature is between 35 and 45 degrees Celsius and the relative humidity is lower than 80%, the fan speed is equal to the maximum speed multiplied by the internal temperature minus 30 and then divided by 15; when the internal temperature exceeds 45 degrees Celsius and the relative humidity is lower than 80%, the fan runs at full speed; the electrolytic cell start-up and shutdown conditions are: when the battery state of charge is higher than 70% and the grid electricity price is lower than 0.25 yuan per kilowatt-hour, or when the photovoltaic power is greater than 80% of the rated power and the state of charge is higher than 90%, the electrolysis program is started; the electrolysis power is taken as the smaller value between the system surplus power and the electrolytic cell rated power.

[0010] Preferably, the near-field magnetic coupling wireless shared charging platform includes: a multi-coil array wireless charging transmitter, a capacitive proximity sensor, a foreign object detection coil, an infrared beacon receiver, a payment verification module, and an electrically operated lifting protective cover; the foreign object detection is based on the change in the quality factor of the transmitter coil: the ratio of the difference between the currently measured quality factor and the no-load quality factor to the no-load quality factor is calculated. If this ratio is less than negative 15% and continues for three measurement cycles, it is determined that there is a metallic foreign object and the output is cut off; the charging power is dynamically adjusted according to the request of the external device, taking the minimum value among the device's requested power, the maximum output power, and the safety margin power to ensure that the total power of the household does not exceed the circuit breaker capacity.

[0011] Preferably, the safety interface and protection unit includes: a DC fast shutdown module, a residual current operated protector, a mechanical disconnect switch, and an insulation resistance detection circuit; the DC fast shutdown module is integrated into the photovoltaic junction box. When the arc current change rate is detected to be greater than 5 amperes per microsecond or the insulation resistance is less than 1 megohm, the series-connected silicon carbide field-effect transistor is disconnected within 100 nanoseconds, and the positive and negative terminals of the photovoltaic string are short-circuited to the discharge resistor; the insulation resistance detection adopts the unbalanced bridge method: two sets of known resistors are connected respectively, the voltage divider is measured twice, and the insulation resistance value is obtained through a rational formula containing the known resistance, the standard resistance, and the two sampled voltages; when the calculated insulation resistance is less than 1 megohm, the protection shutdown is executed.

[0012] Preferably, the intelligent energy control unit executes a self-diagnosis and health management algorithm, automatically running a detection sequence every morning. The battery's ohmic internal resistance is calculated as: the terminal voltage change divided by 0.05 times the rated capacity. If the calculated ohmic internal resistance exceeds 20% of the initial value, the battery is marked as degraded. The photovoltaic module's fill factor is calculated as: the product of the maximum power point voltage and current divided by the product of the open-circuit voltage and short-circuit current. If the fill factor is less than 0.65, an alarm is triggered. All relay contacts are closed sequentially, and the contact voltage drop is detected. If the voltage drop is greater than 0.5 volts, the contacts are judged to be stuck. A 500-volt DC test voltage is applied between the DC bus and ground, and the insulation resistance is measured. If it is less than 2 megohms, a protection shutdown is executed. All detection data is stored locally and uploaded to the cloud to generate a health score. : ; in To score the system's health, This represents the total number of fault items. For fault item index, For the first The weight of each fault item These are measured values. For the corresponding threshold; when Maintenance suggestions will be pushed when the value is less than 0.8.

[0013] Preferably, the intelligent energy control unit also performs load forecasting and photovoltaic power forecasting based on a long short-term memory neural network. The forecasting model structure is a three-layer LSTM network with 64 units per layer. The input features include the photovoltaic power time series of the past 24 hours, cloud map infrared data, irradiance, temperature, humidity, and wind speed. The output is the power curve for the next 4 hours, and the target average absolute percentage error of the forecast is less than 10%. The forecast results are input to the model predictive control optimizer, and the optimization variables are the battery charging and discharging power and the grid exchange power at the next moment. The battery state of charge is updated by subtracting the battery power at the next moment from the current state of charge, multiplying it by the control period, and then dividing it by the total battery energy, where the battery power is positive for discharging and negative for charging. The constraints include: the battery discharge power is between zero and the maximum discharge power, the battery charging power is between the maximum charging power and zero, the state of charge is between the lower limit of 10% and the upper limit of 90%, and the grid exchange power is not less than the negative anti-reverse current upper limit. The model predictive control adopts a quadratic programming algorithm, which is optimized every 5 minutes, and the system working mode is switched according to the optimization results.

[0014] Preferably, the intelligent energy control unit, bidirectional DC-DC converter, bidirectional DC-AC inverter, battery management system, clothes rack motor driver, wireless charging controller, and electrolytic cell controller are interconnected via a bus, using a standard communication protocol with a data refresh rate of no less than 10 Hz. All critical status variables are broadcast to the bus, and a global safety shutdown is immediately triggered when any node detects an anomaly. The logic of the global safety shutdown is as follows: take a logical OR of the fault flags of all nodes in the entire system. If the result is true, simultaneously disconnect all relays, short-circuit the photovoltaic input terminal, shut down the inverter output, and turn on the emergency lights.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. Adopting a unified 48V safe DC bus and DC coupling architecture, photovoltaic power generation directly charges the battery or supplies the load via a single inverter, completely avoiding the 15%-25% energy loss caused by multiple AC-DC conversions in existing AC coupling schemes.

[0016] 2. All DC links are at a safe low voltage, and the photovoltaic modules and energy storage main control box are connected to each other by a waterproof and foolproof quick-connect cable. Users can complete the installation and mobile network pairing within 10 minutes without a professional electrician. At the same time, the energy storage main control box is compactly integrated with modular devices such as clothes racks and plant walls, without occupying the activity space on the balcony floor, which completely solves the pain points of complex wiring and wasted space in the existing distributed system.

[0017] 3. The system has a built-in fast shutdown module, insulation resistance detection, and unbalanced bridge method for real-time monitoring. Once the arc current change rate is detected to exceed 5A / μs or the insulation resistance is lower than 1MΩ, the silicon carbide switch can be disconnected and the photovoltaic input short-circuited within 100ns, eliminating the risk of arcing and electric shock in the humid and vibrating environment of the balcony. At the same time, the active heat dissipation air duct and plant evaporative cooling wall keep the battery operating temperature in the optimal range of 25-35℃.

[0018] 4. The intelligent energy control unit executes an energy dispatching algorithm based on model predictive control, combined with a long short-term memory neural network to predict photovoltaic power and household load. It automatically performs optimal charging and discharging management based on time-of-use electricity prices, weather warnings, and user habits, achieving fully automatic operation such as peak shaving and valley filling, anti-reverse current, and emergency backup power without manual intervention. The self-diagnosis and health management algorithm automatically detects key parameters such as battery internal resistance, fill factor, and contact voltage drop every morning, generates a health score, and pushes maintenance suggestions accurate to the module level.

[0019] 5. It integrates a gravity potential energy recovery type smart clothes drying rack, a photovoltaic energy storage driven micro-ecological circulation system and a near-field magnetic coupling wireless shared charging platform, transforming the daily act of drying clothes into a micro-power generation behavior, using plant transpiration to assist in heat dissipation and produce oxygen and hydrogen, supporting community point-to-point energy sharing and drone charging, and upgrading the balcony from a single energy device into a future energy hub that integrates energy harvesting, mechanical energy recovery, ecological regulation and social interaction. Attached Figure Description

[0020] Figure 1 A block diagram of a photovoltaic energy storage integrated system for balconies provided in this application. Detailed Implementation

[0021] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.

[0022] A photovoltaic energy storage integrated system for balconies includes: a photovoltaic energy harvesting component, an energy storage and energy management main control box, a safety interface and protection unit, an intelligent energy control unit, and an energy interaction and recovery component. The intelligent energy control unit executes an energy dispatching algorithm based on model predictive control, samples photovoltaic power, battery state of charge, household load power, and grid electricity price in real time, and determines the system operating mode by solving a constrained optimization problem. The intelligent energy control unit internally sets an objective function. Minimize it: ; in, For real-time grid electricity prices, Power exchanged with the power grid (kW, positive values ​​for power intake, negative values ​​for power feed-in). This represents the battery's state of charge (0-1). For the target state of charge, These are the weighting coefficients. For integration variables, To predict the time domain, The current time; all DC links in the system are unified to a safe voltage bus, and plug-and-play connection between photovoltaic modules and the main control box is achieved through quick-connect cables; the safe voltage is 48V DC, and the quick-connect cables are waterproof and foolproof aviation plugs; the energy interaction and recovery components are selected from one or more of the following: gravity potential energy recovery clothes rack, photovoltaic energy storage driven micro-ecological cycle system, and near-field magnetic coupling wireless shared charging platform; the typical value of the target state of charge is 0.6, and the weighting coefficient is... The value range is 0.1 to 0.5; the constrained optimization problem includes battery power limiting, upper and lower limits of state of charge, and anti-reverse current constraints; prediction time domain. Use 300 seconds as the timeframe and optimize the cycle every 5 minutes.

[0023] Furthermore, the photovoltaic energy harvesting module has longitudinal heat dissipation fins and a rear-mounted ventilation slot on the back, forming an air gap between itself and the mounting bracket, utilizing thermal pressure difference. Driven by natural convection, the thermal pressure difference is calculated by the following formula: ; in, Air density (unit: kg / m³). The acceleration due to gravity is 9.8 m / s². Vertical height of the ventilation duct (in meters). The absolute temperature of the photovoltaic backsheet (in K). The absolute temperature of the environment (in K); when When the flow resistance is less than the required level, the intelligent energy control unit activates the axial fan for forced ventilation. The fan duty cycle is determined by a linear mapping of the back panel temperature: the duty cycle is zero when the back panel temperature is not higher than 55 degrees Celsius, increases linearly from zero to one between 55 and 65 degrees Celsius, and remains one when the temperature is not lower than 65 degrees Celsius; the ventilation slot height h ≥ 30 mm; the air density is 1.2 kg / m³; the flow resistance is obtained through experimental calibration; the axial fan power is ≤ 2 W, controlled by a PWM signal.

[0024] Furthermore, the energy storage and energy management main control box contains lithium iron phosphate battery clusters, a bidirectional DC-DC converter, and a bidirectional DC-AC inverter. The bidirectional DC-DC converter executes a variable-step perturbation observation maximum power point tracking algorithm, whose perturbation step size is adaptively adjusted according to the power change rate: the current step size equals the maximum step size multiplied by the absolute value of the difference between the current power and the previous power, then divided by the product of the rated power and the normalization coefficient; when the power fluctuation exceeds 10% of the rated power, the step size is automatically reduced to one-tenth of the original; the maximum power point tracking algorithm is executed every 20 milliseconds and is interrupted when the voltage drops by more than 5%. Switch to constant voltage mode; the bidirectional DC-AC inverter has a built-in virtual synchronous generator control, and its rotor motion equation is: the moment of inertia multiplied by the differential of the angular frequency equals the reference active power divided by the reference angular frequency minus the electromagnetic active power divided by the actual angular frequency minus the damping coefficient multiplied by the difference between the angular frequency and the reference angular frequency; the power angle differential equals the angular frequency; the typical maximum step size is 0.01; the normalization coefficient ranges from 0.2 to 0.5; the lithium iron phosphate battery cluster adopts a modular plug-in design; the virtual moment of inertia and damping coefficient are adaptively adjusted according to the grid frequency fluctuation rate, and when the frequency fluctuation rate exceeds 0.5 Hz per second, the moment of inertia increases to twice the default value.

[0025] Furthermore, the gravity potential energy recovery type intelligent clothes drying rack includes: a lifting rope, a pulley system, a one-way bearing, a permanent magnet synchronous motor, a reversible clutch device, a tension sensor, and a motor driver; during the descent of the clothes rack crossbar and the total mass of the clothes, the one-way bearing locks, and the gravitational torque drives the motor to generate electricity, which is then rectified to charge the battery; the recovered power is equal to the generator efficiency multiplied by the total mass multiplied by the gravitational acceleration and then multiplied by the descent speed; the descent speed is adjusted by a PID controller: the proportional term coefficient multiplied by the difference between the speed setpoint and the measured value, plus the integral term coefficient multiplied by the error integral, plus the differential term coefficient multiplied by the error derivative, and the sum of these three plus the set speed yields the target speed; During the lifting process, the clutch disengages, and the user manually pulls the rope. When the tension sensor detects that the tension exceeds the preset threshold and lasts for more than two seconds, the system automatically closes the clutch and activates the electric assist mode. The battery powers the motor to output auxiliary torque, which is equal to the assist coefficient multiplied by the difference between the measured tension and the normal lifting tension. The generator efficiency ranges from 0.6 to 0.8. The descent speed PID setting is 0.2 meters per second. The preset threshold is 80 Newtons. The normal lifting tension is 30 Newtons. The assist coefficient is calibrated and determined. The clothes rack also integrates a weather-linked automatic opening and closing function: when the predicted wind speed is greater than 10 meters per second or the probability of rainfall is greater than 70%, it automatically lowers to recover energy.

[0026] Furthermore, the photovoltaic-storage driven micro-ecological circulation system includes: a plant evaporative cooling wall, a heat exchange duct, a micro PEM electrolyzer, a metal hydride hydrogen storage tank, a micro fuel cell, and a water circulation pipeline; the plant evaporative cooling wall is installed below the photovoltaic modules and consists of a planting box, a capillary water absorption layer, a water storage tank, and temperature and humidity sensors; plant transpiration lowers the temperature of the air flowing through it, and the temperature drop is equal to the latent heat of vaporization of water multiplied by the transpiration rate, divided by the air density multiplied by the specific heat capacity of the air, and then multiplied by the volumetric flow rate through the plant wall; the heat exchange duct introduces the cool, humid air from the back of the plant wall into the air inlet of the energy storage main control box, and the duct is equipped with an adjustable speed fan and an electric air valve; the intelligent energy control unit controls the fan speed according to the internal temperature and relative humidity of the energy storage box: when the internal temperature and relative humidity are within the storage box, the fan speed is adjusted accordingly. The fan stops when the internal temperature does not exceed 35 degrees Celsius or the relative humidity is not lower than 80%. When the internal temperature is between 35 and 45 degrees Celsius and the relative humidity is lower than 80%, the fan speed is equal to the maximum speed multiplied by the internal temperature minus 30 and then divided by 15. When the internal temperature exceeds 45 degrees Celsius and the relative humidity is lower than 80%, the fan runs at full speed. The electrolytic cell starts and stops under the following conditions: when the battery's state of charge is higher than 70% and the grid electricity price is lower than 0.25 yuan per kilowatt-hour, or when the photovoltaic power is greater than 80% of the rated power and the state of charge is higher than 90%, the electrolysis program is started. The electrolysis power is the smaller of the system's surplus power and the electrolytic cell's rated power. The latent heat of vaporization of water is 2260 kJ / kg. The air density is 1.2 kg / m³, and the air specific heat capacity is 1.005 kJ / kg·K. The micro fuel cell has a rated power of 5 watts. When the battery's state of charge is lower than 10%, there is no light, and the grid is cut off, the fuel cell automatically starts to supply power to the control unit. The water circulation pipeline automatically drains water every seven days to prevent algae growth.

[0027] Furthermore, the near-field magnetic coupling wireless shared charging platform includes: a multi-coil array wireless charging transmitter, a capacitive proximity sensor, a foreign object detection coil, an infrared beacon receiver, a payment verification module, and an electrically operated lifting protective cover; foreign object detection is based on changes in the quality factor of the transmitter coil: the ratio of the difference between the currently measured quality factor and the no-load quality factor to the no-load quality factor is calculated. If this ratio is less than negative 15% and continues for three measurement cycles (100 milliseconds per cycle), a metallic foreign object is detected, and the output is cut off; the charging power is dynamically adjusted according to external device requests, taking the minimum value among the device's requested power, the maximum output power (100 watts), and the safety margin power to ensure that the total household power does not exceed the circuit breaker capacity; when the proximity sensor detects a human or animal approaching at a distance of less than 30 centimeters, the power is automatically reduced to below 10 watts; the system also integrates a dedicated charging platform for drones or delivery robots. Drones verify their identity through an infrared beacon handshake protocol, and the charging power is dynamically adjusted according to the drone's battery management system requests; when the wind speed is greater than 8 meters per second or a thunderstorm warning is issued, the intelligent energy control unit rejects the landing request.

[0028] Furthermore, the safety interface and protection unit includes: a DC fast shutdown module, a residual current operated protective device (RCD), a mechanical disconnect switch, and an insulation resistance detection circuit. The DC fast shutdown module is integrated into the photovoltaic junction box. When the arc current change rate is detected to be greater than 5 amps per microsecond or the insulation resistance is less than 1 megohm, it disconnects the series-connected silicon carbide MOSFET within 100 nanoseconds and simultaneously short-circuits the positive and negative terminals of the photovoltaic string to the bleed resistor. The insulation resistance detection uses the unbalanced bridge method: two sets of known resistors are connected respectively, and the voltage divider is measured twice. The insulation resistance value is obtained through a rational formula containing the known resistance, the standard resistance, and the two sampled voltages. When the calculated insulation resistance is less than 1 megohm, the protection shutdown is executed. The bleed resistor has a resistance of 100 ohms and a power of 10 watts. The residual current operated protective device has an operating current of 30 mA. The mechanical disconnect switch is a manually operated knife switch with a rated voltage of 60 volts and a rated current of 30 amps. The insulation resistance detection cycle is 10 minutes.

[0029] Furthermore, the intelligent energy control unit executes self-diagnosis and health management algorithms, automatically running a detection sequence every morning. The battery's ohmic internal resistance is calculated as: the terminal voltage change divided by 0.05 times the rated capacity (where the terminal voltage change is the voltage difference during 5 minutes of charging and discharging at 0.05 times the rated current rate). If the calculated ohmic internal resistance exceeds 20% of the initial value, the battery is marked as degraded. The photovoltaic module's fill factor is calculated as: the product of the maximum power point voltage and current divided by the product of the open-circuit voltage and short-circuit current. If the fill factor is less than 0.65, an alarm is triggered. All relay contacts are closed sequentially, and the contact voltage drop is detected. If the voltage drop is greater than 0.5 volts, the contacts are considered stuck. A 500-volt DC test voltage is applied between the DC bus and ground, and the insulation resistance is measured. If it is less than 2 megohms, a protective shutdown is initiated. All detection data is stored locally and uploaded to the cloud to generate a health score. : ; in To score the system's health, This represents the total number of fault items. For fault item index, For the first The weight of each fault (the sum of all weights is one). These are measured values. For the corresponding threshold; when Maintenance suggestions are pushed when the battery capacity is less than 0.8 ohms; the detection time is fixed at 2:00 AM; the rated battery capacity is in ampere-hours; the maintenance suggestions include specific fault module indications, such as "the internal resistance of the B channel battery module is too high".

[0030] Furthermore, the intelligent energy control unit also performs load forecasting and photovoltaic power forecasting based on a long short-term memory neural network. The forecasting model structure is a three-layer LSTM network with 64 units per layer. Input features include the photovoltaic power time series of the past 24 hours, cloud map infrared data, irradiance, temperature, humidity, and wind speed. The output is the power curve for the next 4 hours, with a target average absolute percentage error of less than 10%. The forecast results are input into the model predictive control optimizer, where the optimization variables are the battery charging and discharging power and grid exchange power at the next moment. The battery state of charge is updated by subtracting the battery power at the next moment from the current state of charge, multiplying by the control period, and then dividing by the total battery energy. The system is characterized by positive charge and negative charge; constraints include: battery discharge power between zero and maximum discharge power, battery charging power between maximum charging power (negative value) and zero, state of charge between 10% and 90%, and grid exchange power not less than the negative anti-reverse current limit (50 watts); model predictive control uses a quadratic programming algorithm, which is optimized every 5 minutes and the system operating mode is switched based on the optimization results; the long short-term memory neural network is pre-trained in the cloud and then downloaded to the local intelligent energy control unit for operation; the quadratic programming algorithm uses open-source solvers such as OSQP or interior-point method; the anti-reverse current limit is set at 50 watts to avoid reverse power feeding to the grid; the control cycle is 300 seconds.

[0031] Furthermore, the intelligent energy control unit, bidirectional DC-DC converter, bidirectional DC-AC inverter, battery management system, clothes rack motor driver, wireless charging controller, and electrolytic cell controller are interconnected via a bus, using a standard communication protocol with a data refresh rate of no less than 10 Hz. All critical status quantities (including voltage, current, temperature, and fault codes) are broadcast to the bus. When any node detects an anomaly, a global safety shutdown is immediately triggered. The logic of the global safety shutdown is as follows: a logical OR is performed on the fault flags of all nodes in the entire system. If the result is true, all relays are simultaneously disconnected, the photovoltaic input terminal is shorted, the inverter output is turned off, and the emergency light is turned on. The bus is a CAN bus or RS485 bus; the communication protocol is the Modbus RTU protocol; the global safety shutdown action also includes triggering a mechanical disconnect switch; the emergency light is a 3-watt LED with a battery life of no less than 2 hours.

[0032] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of protection of this application is limited to these examples; within the framework of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of one or more embodiments of this application as described above, which are not provided in detail for the sake of brevity.

[0033] One or more embodiments in this application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of this application. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of one or more embodiments in this application should be included within the protection scope of this application.

Claims

1. A photovoltaic energy storage integrated system for balconies, characterized in that, include: The system comprises a photovoltaic energy harvesting module, an energy storage and energy management main control box, a safety interface and protection unit, an intelligent energy control unit, and an energy interaction and recovery module. The intelligent energy control unit executes a model predictive control-based energy scheduling algorithm, samples photovoltaic power, battery state of charge, household load power, and grid electricity price in real time, and determines the system operating mode by solving a constrained optimization problem. The intelligent energy control unit internally sets an objective function. Minimize it: ； in, For real-time grid electricity prices, For power exchange with the power grid, This represents the battery's state of charge (0-1). For the target state of charge, These are the weighting coefficients. For integration variables, To predict the time domain, The current moment; The energy interaction and recovery components are selected from one or more of the following: a gravitational potential energy recovery clothes rack, a photovoltaic energy storage driven micro-ecological cycle system, and a near-field magnetic coupling wireless shared charging platform.

2. The photovoltaic energy storage integrated system for balconies according to claim 1, characterized in that, The photovoltaic energy harvesting module has longitudinal heat dissipation fins and a rear-mounted ventilation slot on its back, forming an air gap between itself and the mounting bracket, utilizing thermal pressure difference. Driven by natural convection, the thermal pressure difference is calculated by the following formula: ； in, air density, It is the acceleration due to gravity. The vertical height of the ventilation duct. The absolute temperature of the photovoltaic backsheet. The absolute temperature of the environment; when When the flow resistance is less than the required level, the intelligent energy control unit activates the axial fan for forced ventilation. The fan duty cycle is determined by a linear mapping of the back panel temperature: the duty cycle is zero when the back panel temperature is not higher than 55 degrees Celsius, increases linearly from zero to one when the temperature is between 55 and 65 degrees Celsius, and is one when the temperature is not lower than 65 degrees Celsius.

3. The photovoltaic energy storage integrated system for balconies according to claim 1, characterized in that, The energy storage and energy management main control box contains a lithium iron phosphate battery cluster, a bidirectional DC-DC converter, and a bidirectional DC-AC inverter. The bidirectional DC-DC converter executes a variable step-size disturbance observation maximum power point tracking algorithm, whose disturbance step size is adaptively adjusted according to the power change rate: the current step size is equal to the maximum step size multiplied by the absolute value of the difference between the current power and the previous power, and then divided by the product of the rated power and the normalization coefficient; when the power fluctuation is greater than 10% of the rated power, the step size is automatically reduced to one-tenth of the original; the maximum power point tracking algorithm is executed once every 20 milliseconds, and is interrupted and switched to constant voltage mode when the voltage drops by more than 5%; the bidirectional DC-AC inverter has a built-in virtual synchronous generator control, and its rotor motion equation is: the moment of inertia multiplied by the differential of the angular frequency is equal to the reference active power divided by the reference angular frequency minus the electromagnetic active power divided by the actual angular frequency minus the damping coefficient multiplied by the difference between the angular frequency and the reference angular frequency; the power angle differential is equal to the angular frequency.

4. The photovoltaic energy storage integrated system for balconies according to claim 1, characterized in that, The gravity potential energy recovery type intelligent clothes drying rack includes: a lifting rope, a pulley system, a one-way bearing, a permanent magnet synchronous motor, a reversible clutch device, a tension sensor, and a motor driver. During the descent of the clothes rack crossbar and the total mass of the clothes, the one-way bearing locks, and the gravitational torque drives the motor to generate electricity. The output power is rectified and then used to charge the battery. The recovered power is equal to the generator efficiency multiplied by the total mass multiplied by the gravitational acceleration and then multiplied by the descent speed. The descent speed is adjusted by a PID controller: the proportional term coefficient multiplied by the difference between the speed set value and the measured value, plus the integral term coefficient multiplied by the error integral, plus the derivative term coefficient multiplied by the error derivative, and the sum of the three plus the set speed gives the target speed. During the ascent, the clutch disengages, and the user manually pulls the rope. When the tension sensor detects that the tension exceeds the preset threshold and lasts for more than two seconds, the system automatically closes the clutch and starts the electric assist mode. The battery powers the motor to output auxiliary torque, which is equal to the assist coefficient multiplied by the difference between the measured tension and the normal ascent tension.

5. The photovoltaic energy storage integrated system for balconies according to claim 1, characterized in that, The photovoltaic-storage driven micro-ecological circulation system includes: a plant evaporative cooling wall, a heat exchange duct, a micro PEM electrolyzer, a metal hydride hydrogen storage tank, a micro fuel cell, and a water circulation pipeline. The plant evaporative cooling wall, installed below the photovoltaic modules, consists of a planting box, a capillary water absorption layer, a water storage tank, and temperature and humidity sensors. Plant transpiration lowers the temperature of the air flowing through it; the temperature drop is equal to the latent heat of vaporization of water multiplied by the transpiration rate, divided by the air density, multiplied by the air specific heat capacity, and then multiplied by the volumetric flow rate through the plant wall. The heat exchange duct introduces the cool, humid air from the back of the plant wall into the air inlet of the energy storage main control box; the duct is equipped with an adjustable-speed fan and an electric air valve. The intelligent energy control unit adjusts the airflow based on the internal temperature and phase of the energy storage box. Humidity control of fan speed: The fan stops when the internal temperature does not exceed 35 degrees Celsius or the relative humidity is not lower than 80%. When the internal temperature is between 35 and 45 degrees Celsius and the relative humidity is lower than 80%, the fan speed is equal to the maximum speed multiplied by the internal temperature minus 30 and then divided by 15. When the internal temperature exceeds 45 degrees Celsius and the relative humidity is lower than 80%, the fan runs at full speed. The start-up and shutdown conditions of the electrolytic cell are: when the battery state of charge is higher than 70% and the grid electricity price is lower than 0.25 yuan per kilowatt-hour, or when the photovoltaic power is greater than 80% of the rated power and the state of charge is higher than 90%, the electrolysis program is started. The electrolysis power is the smaller value between the system surplus power and the rated power of the electrolytic cell.

6. The photovoltaic energy storage integrated system for balconies according to claim 1, characterized in that, The near-field magnetic coupling wireless shared charging platform includes: a multi-coil array wireless charging transmitter, a capacitive proximity sensor, a foreign object detection coil, an infrared beacon receiver, a payment verification module, and an electrically operated lifting protective cover. Foreign object detection is based on changes in the transmitter coil's quality factor: the ratio of the difference between the currently measured quality factor and the no-load quality factor to the no-load quality factor is calculated. If this ratio is less than negative 15% and persists for three consecutive measurement cycles, a metallic foreign object is determined to be present, and the output is cut off. The charging power is dynamically adjusted according to external device requests, taking the minimum value among the device's requested power, the maximum output power, and the safety margin power to ensure the total household power does not exceed the circuit breaker capacity.

7. The photovoltaic energy storage integrated system for balconies according to claim 1, characterized in that, The safety interface and protection unit includes: a DC fast shutdown module, a residual current operated protector, a mechanical disconnect switch, and an insulation resistance detection circuit. The DC fast shutdown module is integrated into the photovoltaic junction box. When the arc current change rate is detected to be greater than 5 amperes per microsecond or the insulation resistance is less than 1 megohm, the series-connected silicon carbide field-effect transistor is disconnected within 100 nanoseconds, and the positive and negative terminals of the photovoltaic string are shorted to the discharge resistor. The insulation resistance detection adopts the unbalanced bridge method: two sets of known resistors are connected respectively, the voltage divider is measured twice, and the insulation resistance value is obtained through a rational formula containing the known resistance, the standard resistance, and the two sampled voltages. When the calculated insulation resistance is less than 1 megohm, the protection shutdown is executed.

8. The photovoltaic energy storage integrated system for balconies according to claim 1, characterized in that, The intelligent energy control unit executes self-diagnosis and health management algorithms, automatically running a detection sequence every morning. The battery's internal resistance is calculated as: terminal voltage change divided by 0.05 times the rated capacity. If the calculated internal resistance exceeds 20% of the initial value, the battery is marked as degraded. The photovoltaic module's fill factor is calculated as: the product of the maximum power point voltage and current divided by the product of the open-circuit voltage and short-circuit current. If the fill factor is less than 0.65, an alarm is triggered. All relay contacts are closed sequentially, and the contact voltage drop is detected. If the voltage drop is greater than 0.5 volts, the contacts are considered stuck. A 500-volt DC test voltage is applied between the DC bus and ground, and the insulation resistance is measured. If it is less than 2 megohms, a protective shutdown is initiated. All detection data is stored locally and uploaded to the cloud to generate a health score. : ； in To score the system's health, This represents the total number of fault items. For fault item index, For the first The weight of each fault item These are measured values. For the corresponding threshold; when Maintenance suggestions will be pushed when the value is less than 0.

8.

9. A photovoltaic energy storage integrated system for balconies according to claim 1, characterized in that, The intelligent energy control unit also performs load forecasting and photovoltaic power forecasting based on a long short-term memory neural network. The forecasting model structure is a three-layer LSTM network with 64 units per layer. Input features include the photovoltaic power time series of the past 24 hours, cloud map infrared data, irradiance, temperature, humidity, and wind speed. The output is the power curve for the next 4 hours, with a target average absolute percentage error of less than 10%. The forecast results are input to the model predictive control optimizer, where the optimization variables are the battery charging and discharging power and the grid exchange power at the next moment. The battery state of charge is updated by subtracting the battery power at the next moment from the current state of charge, multiplying by the control period, and then dividing by the total battery energy, where battery power is positive for discharging and negative for charging. Constraints include: battery discharging power between zero and maximum discharging power, battery charging power between maximum charging power and zero, state of charge between the lower limit of 10% and the upper limit of 90%, and grid exchange power not less than the negative anti-reverse current upper limit. The model predictive control uses a quadratic programming algorithm, which is optimized every 5 minutes and the system operating mode is switched based on the optimization results.

10. A photovoltaic energy storage integrated system for balconies according to claim 1, characterized in that, The intelligent energy control unit, bidirectional DC-DC converter, bidirectional DC-AC inverter, battery management system, clothes rack motor driver, wireless charging controller, and electrolytic cell controller are interconnected via a bus, using a standard communication protocol with a data refresh rate of no less than 10 Hz. All critical status variables are broadcast to the bus, and any node that detects an anomaly immediately triggers a global safety shutdown. The logic for global safety shutdown is as follows: take a logical OR of the fault flags of all nodes in the entire system. If the result is true, then simultaneously disconnect all relays, short-circuit the photovoltaic input terminal, shut down the inverter output, and turn on the emergency lights.

Images