Intelligent charging system based on photovoltaic self-adaptation and bluetooth cloud control

Abstract

The present application relates to photovoltaic power generation and intelligent charging control technical field, specifically to intelligent charging system based on photovoltaal adaptive and bluetooth cloud control, including local controller, and photovoltaic acquisition and control module, bluetooth local communication and user terminal module, cloud platform remote operation and forecast optimization module which work cooperatively with the local controller as the core; the local controller realizes second-level photoelectric data acquisition, adaptive MPPT control and power management, the cloud platform completes multi-source meteorological data fusion, photovoltaic power prediction and multi-target global optimization, the bluetooth module realizes local visual interaction and instruction transmission; the present application provides an integrated intelligent charging system integrating photovoltaic adaptive control, bluetooth local interaction and cloud intelligent decision, realizes the improvement of photovoltaic utilization efficiency, the reduction of power cost and the optimization of user experience, and meets the dual needs of local independent operation and cloud global optimization.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic power generation and intelligent charging control technology, specifically to an intelligent charging system based on photovoltaic adaptive and Bluetooth cloud control. Background Technology

[0002] With the popularization of photovoltaic power generation technology, the application scenarios of household, portable energy storage devices and small-scale distributed photovoltaic systems are becoming increasingly diverse. Although traditional photovoltaic charging systems have basic maximum power point tracking (MPPT) functions, they have significant technical deficiencies in terms of intelligent coordination, real-time response capabilities, and regional adaptability. Specifically, these deficiencies are as follows: Functional fragmentation and lack of closed-loop control: Most existing photovoltaic charging devices achieve a single function, some only have MPPT tracking function, some mainly focus on Bluetooth local control, and the very few devices connected to the cloud only use the cloud for data reporting. Real-time policy adjustment between the cloud and the local system has not been realized, and there is a lack of coordination between the various functional modules, making it impossible to form a closed loop of "collection-decision-execution-feedback".

[0003] MPPT adjustment has a lag, resulting in low photovoltaic utilization: In scenarios with rapid fluctuations in light, such as cloud cover or building shading, traditional MPPT control algorithms require feedback of several seconds or even tens of seconds to complete parameter adjustment, which significantly reduces tracking efficiency. This is particularly evident in scenarios with high real-time requirements, such as outdoor mobile and portable energy storage.

[0004] Lack of regional adaptability and rigid charging strategies: The light intensity, sunshine duration and seasonal variation characteristics of different geographical regions vary significantly, and the peak and off-peak electricity prices and price standards of the power grids in different regions are also different. The charging strategies of traditional photovoltaic charging equipment are fixed and cannot automatically learn and adapt to the above regional differences, resulting in low photovoltaic utilization and ineffective control of electricity costs.

[0005] Low interaction and operation and maintenance efficiency, poor user experience: The local interaction methods of existing devices are limited, and there is a lack of visual operation and data display channels; at the same time, there is no cloud-based remote operation and maintenance and predictive scheduling capabilities, and equipment maintenance requires on-site operation, resulting in high operation and maintenance costs, which is especially unsuitable for application scenarios with wide area distribution.

[0006] To address the aforementioned technical challenges, the industry urgently needs an integrated intelligent charging system that combines photovoltaic adaptive control, Bluetooth local interaction, and cloud-based intelligent decision-making. This system would improve photovoltaic utilization efficiency, reduce electricity costs, and optimize user experience, while simultaneously meeting the dual requirements of independent local operation and global cloud-based optimization. Summary of the Invention

[0007] To address the problems of existing photovoltaic charging systems, such as limited control schemes, fixed charging strategies that cannot adapt to various regions, low photovoltaic utilization efficiency, low interaction and maintenance efficiency, and poor user experience, this invention provides an intelligent charging system based on photovoltaic adaptive and Bluetooth cloud control. This system is adaptable to various regional situations and can improve photovoltaic utilization efficiency, reduce electricity costs, and optimize user experience.

[0008] The technical solution of the present invention is as follows: The intelligent charging system based on photovoltaic adaptive and Bluetooth cloud control adopts a hybrid control architecture of cloud collaboration and local priority. Its core includes a local controller, and a photovoltaic data acquisition and control module, a Bluetooth local communication and user terminal module, and a cloud platform remote operation and maintenance and predictive optimization module that work collaboratively with the local controller. The collaborative relationship between each module and the local controller is as follows: The photovoltaic data acquisition and control module is a local core functional module with a local controller at its core. All its functions, including data acquisition, maximum power point tracking control, and power management, are executed by the local controller. The Bluetooth local communication and user terminal module includes a Bluetooth communication unit integrated into the local controller and an external user terminal. The Bluetooth communication unit is a built-in communication interface of the local controller, enabling bidirectional data interaction between the local controller and the user terminal. The cloud platform remote operation and maintenance and predictive optimization module is a remote decision-making terminal. It establishes a wireless data interaction connection with the local controller, sends charging optimization strategies to the local controller, and receives on-site operation data uploaded by the local controller to complete model calibration and rolling optimization. The local controller serves as the control center on the local side of the system. It receives data collected by the photovoltaic acquisition and control module and executes local decisions. It interacts with the user terminal through the Bluetooth communication unit, receives and executes the optimization strategies of the cloud platform's remote operation and maintenance and predictive optimization module, and simultaneously feeds back operating data to the cloud platform.

[0009] Furthermore, the photovoltaic data acquisition and control module is dominated by a local controller and achieves its functions through the following steps: Step 1: Collect the photovoltaic panel voltage V at 1-second intervals. pv Current I pv Temperature T pv Battery voltage V b Current I b Real-time photoelectric data of battery state of charge (SOC); Step 2: Calculate the instantaneous photovoltaic power P based on the collected photoelectric data. pv (t) and the rate of change of power voltage The calculation formula is as follows: P pv (t) = Vpv (t)*I pv (t); ; Step 3: Perform adaptive step size adjustment, as shown in the following formula: ΔD (t+1) = K1*(dPpv / dVpv) + K2*f (Tpv, SOC); Where ∆D(t+1) represents the duty cycle adjustment amount to be applied in the next control cycle, K1 is the MPPT tracking gain coefficient, K2 is the environmental and state compensation coefficient, (dPpv / dVpv) represents the slope of the photovoltaic power curve, f(Tpv,SOC) is the temperature compensation and battery state correction value, and Tpv represents the photovoltaic panel temperature. Step 4: Dynamically adjust the charging mode according to the battery's State of Charge (SOC). When SOC < 0.8, use constant current fast charging; when SOC ≤ 0.8 < 0.95, use constant voltage charging; and when SOC ≥ 0.95, use trickle charging. Step 5: Execute the local power management strategy and allocate photovoltaic power and grid power according to the operating mode and strategy preference parameters.

[0010] Furthermore, the local power management strategy executed by the local controller defines a strategy preference parameter. The formula for the preference parameter is as follows: ; in, This indicates the adjustment of the distribution ratio between photovoltaic power and grid power. The operating mode selected by the user For real-time electricity prices, User priority weight, The time factor; The user-selected operating mode This includes photovoltaic priority mode, economic mode, and off-peak electricity priority mode. The local controller has built-in the corresponding modes for each. The mapping calculation function is dynamically solved based on real-time parameters. .

[0011] Furthermore, the bidirectional data interaction between the Bluetooth local communication and the user terminal module includes: External user terminals send custom setting instructions for charging period, operating mode, and priority weight to the local controller via Bluetooth. The local controller updates the charging strategy immediately after receiving the instructions. The local controller uploads real-time photovoltaic power, charging current, battery state of charge (SOC), and historical charging curve data to an external user terminal via Bluetooth communication unit, enabling data visualization.

[0012] Furthermore, the cloud platform remote operation and maintenance and predictive optimization module achieves its functions in collaboration with the local controller through the following steps: Step 1: Obtain structured forecast data for the next few hours through the meteorological bureau's public API, including ultraviolet radiation data, geographic location and time data, cloud cover, probability of rainfall, relative humidity, visibility, and ambient temperature; Step 2: Perform multi-layer corrections on the original irradiance data based on basic attenuation, humidity, visibility, and solar altitude angle to obtain the corrected irradiance Gpred(t), and substitute it into the formula to calculate the predicted photovoltaic power Ppv,pred(t). Step 3: Generate the target charging power curve Ptarget(t) in two stages: planned generation and rolling correction, to meet the user's minimum charging power requirement Pdemand_min while minimizing the net electricity cost; Step 4: Decompose the target charging power curve into instructions that can be executed by the local controller, send them to the local controller via the MQTT protocol, and receive the running data uploaded by the local controller to complete the model calibration.

[0013] Furthermore, the formula for calculating the corrected irradiance Gpred(t) is as follows: Gpred(t) = G_pred_raw(t) × β_enhanced(t) × K_sunangle(t); β_enhanced (t) = β_base(t) × f_humidity × f_visibility; β_base (t) = 1 - [ω_cloud×C (t) + ω_rain×R (t)]; Wherein, β_enhanced(t) is the enhancement correction factor, β_base(t) represents the base attenuation factor, f_humidity represents the humidity correction factor, f_visibility represents the visibility correction factor, C(t) is the cloud cover rate, R(t) is the probability of rainfall, ω_cloud and ω_rain are weighting coefficients; K_sunangle(t) is the solar altitude angle correction coefficient, which is determined by the current date and time.

[0014] Furthermore, the predicted photovoltaic power The calculation formula is: ; in, For overall system efficiency, the value range is 0.75~0.85; This refers to the total area of ​​the photovoltaic panels; Temperature coefficient; To predict ambient temperature.

[0015] Furthermore, the cloud platform's remote operation and maintenance and predictive optimization module has regional adaptive learning capabilities. It analyzes the lighting patterns and seasonal variation characteristics of different geographical regions through big data analysis, and generates personalized charging plans in conjunction with local power grid prices. These plans are then sent to the local controller for execution, and the personalized charging plans are dynamically updated based on the operating data fed back by the local controller.

[0016] Furthermore, the local controller has built-in operating mode switching logic, which automatically switches between three operating modes based on the power grid status and cloud connection status: Emergency mode: Triggered when the power grid is abnormal or malfunctioning, the control objective is to ensure system safety; Planned Mode: Triggered when the cloud-based optimization strategy is effective, executing the optimized charging plan issued by the cloud platform's remote operation and maintenance and predictive optimization module; Autonomous Mode: Triggered when there is no cloud-based optimization strategy, the local controller independently completes all functions of the photovoltaic acquisition and control module, achieving optimal allocation of local photovoltaic power and grid power.

[0017] Furthermore, the local controller includes a housing and a circuit board mounted inside the housing; The outer casing is made of aluminum alloy and includes a cover, a front panel, a base, and a rear panel. A cooling fan is installed on the rear panel. The circuit board integrates an input reverse polarity protection MOSFET, a boost MOSFET, a freewheeling MOSFET, an output switch control MOSFET, a battery output line, a photovoltaic input line, an electrolytic capacitor, a power inductor, a digital tube display screen, and a control board.

[0018] Compared with the prior art, the present invention has the following beneficial effects: (1) Maintaining an MPPT tracking efficiency of ≥0.99 under this light fluctuation scenario, the photovoltaic utilization rate is greatly improved: The adaptive MPPT control algorithm dominated by the local controller is the core technical feature. Through real-time photoelectric data acquisition with a 1-second cycle and real-time calculation of power-voltage change rate, combined with the dual correction terms of ambient temperature and battery SOC, the duty cycle is adjusted at the second level, which solves the adjustment lag problem of the traditional MPPT algorithm. Even under the rapid light fluctuation scenario such as cloud cover and building shading, it can still maintain a maximum power point tracking efficiency of ≥0.99, which significantly improves the utilization efficiency of photovoltaic energy.

[0019] (2) Realize dual intelligent decision-making of local + cloud, and the system has strong adaptability to different working conditions: The technical feature is the hybrid control architecture of "cloud collaboration and local priority" and the automatic switching logic of the local controller's operating mode. The local controller can independently complete data acquisition, MPPT control and power management, and realize autonomous operation in the offline state; in the network state, it can receive the global optimization strategy of the cloud platform to realize the collaboration of local execution and cloud optimization; when the power grid is abnormal, it can automatically trigger the emergency mode to ensure system safety, solve the limitations of the single control architecture of traditional equipment, and greatly improve the adaptability of the system under different working conditions.

[0020] (3) The first regional adaptive light and electricity price learning mechanism makes the charging strategy intelligent and personalized: The cloud platform’s regional adaptive learning capability and the electricity price adaptation logic of the strategy preference parameter α(t) are technical features. The cloud platform analyzes the light patterns and seasonal changes in different regions through big data analysis, and generates personalized charging solutions in combination with the peak and valley electricity price standards of the local power grid. At the same time, α(t) can dynamically adjust the distribution ratio of photovoltaic and grid power according to the real-time electricity price and the valley electricity period, so that the charging strategy can be adapted to the natural conditions and electricity price policies of different regions, solving the problem of the fixed charging strategy of traditional equipment.

[0021] (4) Bluetooth local visualization interaction + cloud remote operation and maintenance greatly improves user experience and operation and maintenance efficiency: With the Bluetooth communication unit integrated in the local controller and the remote operation and maintenance function of the cloud platform as technical features, users can realize the custom setting of charging parameters, real-time photoelectric data visualization and historical charging curve traceability through mobile APP. The operation is convenient and highly visualized; the cloud platform can realize remote policy distribution, equipment status monitoring and model calibration without on-site operation, which greatly reduces the operation and maintenance cost of equipment, especially suitable for application scenarios with wide distribution in Europe and the United States.

[0022] (5) Multi-objective global optimization minimizes electricity costs and improves photovoltaic self-consumption rate: The multi-objective global optimization model of the cloud platform and the power balance equation of the local controller are the technical features. Under the premise of ensuring the minimum charging needs of users, the optimization model realizes the power allocation strategy of "prioritizing photovoltaic when photovoltaic is sufficient and prioritizing grid when electricity price is low" by minimizing net electricity costs. At the same time, the quadratic penalty function in the rolling correction stage ensures that the actual charging power closely tracks the planned target. Even when photovoltaic prediction is inaccurate or there is a sudden power curtailment, a feasible suboptimal solution can still be found. This not only improves the photovoltaic self-consumption rate and reduces the dependence on the external grid, but also effectively controls the electricity costs of users.

[0023] (6) Integrated hardware design improves system reliability and portability: The integrated aluminum alloy shell of the local controller and the heat dissipation method of "heat sink + forced convection" are technical features. The core power devices are all fixed on the heat sink, and the cooling fan realizes forced convection heat dissipation, keeping the devices in a low temperature range, avoiding derating due to temperature rise, and improving the system's continuous charging capability and hardware reliability. At the same time, the integrated and miniaturized hardware design reduces the size of the device and improves portability, making it suitable for various application scenarios such as home, outdoor, and portable energy storage. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the workflow of the system of the present invention; Figure 2 This is a schematic diagram of the workflow of the cloud platform remote operation and maintenance and predictive optimization module in an embodiment of the present invention; Figure 3 This is a flowchart illustrating the optimization strategy decomposition and distribution in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of the local controller in an embodiment of the present invention; Figure 5 This is a circuit topology diagram of the circuit board in the local controller of an embodiment of the present invention.

[0025] In the diagram: 1-Cover; 2-Battery output line; 3-Photovoltaic input line; 4-Aluminum alloy back plate; 5-Cooling fan; 6-Casing; 7-Input electrolytic capacitor; 8-Input reverse polarity protection MOSFET; 9-Boost MOSFET; 10-Power inductor; 11-Freewheeling MOSFET; 12-Output electrolytic capacitor; 13-Output switch control MOSFET; 14-Front panel; 15-Digital display screen and control board. Detailed Implementation

[0026] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0027] See Figure 1 A smart charging system based on photovoltaic adaptive and Bluetooth cloud control is proposed. It adopts a hybrid control architecture of cloud-based collaboration and local priority, with the local controller as the core control hub. This system, along with three main functional modules—a photovoltaic data acquisition and control module, a Bluetooth local communication and user terminal module, and a cloud platform remote operation and maintenance and predictive optimization module—forms a closed-loop control system of "data acquisition—intelligent algorithm—remote interaction." The collaborative relationships and core functions of each module with the local controller are as follows: Photovoltaic data acquisition and control module: This module is the core local-side functional module with the local controller at its core. Its sensing, power, and execution hardware are all integrated into the local controller, and all data acquisition, MPPT control, and power management functions are executed by the local controller. It includes the following steps: Step 1: The local controller collects the photovoltaic panel voltage V at 1-second intervals. pv Current I pv Temperature T pv Battery voltage V b Current I b Real-time photoelectric data of battery state of charge (SOC); Step 2: Calculate the instantaneous photovoltaic power P based on the collected photoelectric data. pv (t) and the rate of change of power voltage The calculation formula is as follows: P pv (t) = V pv (t)*I pv (t); ; Step 3: Perform adaptive step size adjustment, as shown in the following formula: ΔD (t+1) = K1*(dPpv / dVpv) + K2*f (Tpv, SOC); Where ∆D(t+1) represents the duty cycle adjustment amount to be applied in the next control cycle, K1 is the MPPT tracking gain coefficient, K2 is the environmental and state compensation coefficient, (dPpv / dVpv) represents the slope of the photovoltaic power curve, f(Tpv,SOC) is the temperature compensation and battery state correction value, and Tpv represents the photovoltaic panel temperature. when When the maximum power point is approached, the step size D is reduced. when Increasing the step size D speeds up the response.

[0028] Where ϵ1 is the steady-state threshold, which is derived from the characteristics of different materials / combinations of solar panels. The specific value can be obtained from the photovoltaic manufacturer or by engineers through simulation experiments. ϵ2 is a dynamic threshold value, which is derived from the characteristics of different materials / combinations of solar panels. The specific value can be obtained from the photovoltaic manufacturer or through simulation experiments conducted by engineers.

[0029] Step 4: Dynamically adjust the charging mode according to the battery's State of Charge (SOC). When SOC < 0.8, use constant current fast charging; when SOC ≤ 0.8 < 0.95, use constant voltage charging; and when SOC ≥ 0.95, use trickle charging. Step 5: Execute the local power management strategy and allocate photovoltaic power and grid power according to the operating mode (as shown in Table 1 below) and strategy preference parameters; users can set priorities (such as "priority power supply during off-peak hours"). Table 1

[0030] When Bluetooth command parameters change, the local controller updates its policy immediately. ; ; ; The target total charging power of the system at time t. This is the output instruction that the local controller ultimately strives to achieve.

[0031] : The actual output power of the photovoltaic array at time t (obtained by S1 and optimized by S2). Represents "free but fluctuating" clean energy.

[0032] Power drawn from the grid at time t. Represents a "stable but costly" energy source.

[0033] The function is defined as: ensuring that the photovoltaic power used does not exceed the battery's required power demand, nor does it exceed the photovoltaic panel's energy supply, ensuring that it is within the theoretically available range.

[0034] Defined as "strategy preference parameter", The policy preference parameter is dynamically calculated based on local and cloud policies, ranging from 0 to 1. The formula is as follows: ; in, This indicates the adjustment of the distribution ratio between photovoltaic power and grid power. The operating mode selected by the user For real-time electricity prices, User priority weight (range [0, 1]). The time factor ranges from [0, 1]. (See Table 2) An example representing the set of hours during off-peak electricity hours, such as: ={22,23,0,1,2,3,4,5,6,7}; Table 2

[0035] The mapping function f_map is expressed as follows: Photovoltaic priority mode: ; Economic model: ; Off-peak electricity priority mode: ; Example: ; ; ; Example Scenario 1: Photovoltaic Priority Mode ; ; ; ; Example Scenario 2: Off-peak electricity priority mode ; ; ; ; Based on this, a reasonable bias parameter α is derived from the model.

[0036] Given the current grid electricity price, a price normalization factor N(t) is needed during calculation. This normalization factor maps the current electricity price to the range [0,1], reflecting the relative level of the price. Its expression is: ; The user-selected operating mode This includes photovoltaic priority mode, economic mode, and off-peak electricity priority mode. The local controller has built-in the corresponding modes for each. The mapping calculation function is dynamically solved based on real-time parameters. .

[0037] Specifically, the photovoltaic-first mode: α base =1.0 indicates that the default setting fully favors the use of photovoltaics.

[0038] = max(0.7, min(1.0, α) base- 0.3·P_user·(1-N(t)) - 0.2·(1-T(t))); Economic model: α base =0.5 indicates initial neutrality, which can be dynamically adjusted according to electricity prices and photovoltaic power.

[0039] = max(0.3, min(0.8, α) base + 0.2·(N(t)-0.5) + 0.1·P_user·(1-N(t))- 0.15·(1-T(t))); Off-peak electricity priority mode: α base =0.1 (off-peak electricity period) / α base =0.6 (non-off-peak hours) indicates a default preference for using the power grid, but only during off-peak hours when electricity prices are low.

[0040] If T(t) < 0.5 (off-peak electricity period): α base = 0.0; α(t) = min(0.95, max(0.05, α base + 0.3·N(t) + 0.15·P_user + 0.05·(1-T(t))); If T(t) ≥ 0.5 (non-valley electricity period): α base = 0.6; α(t) = min(0.95, max(0.05, α base + 0.4·N(t) + 0.2·P_user - 0.1·(1-T(t))); α base This indicates the base value of the biased parameter.

[0041] Bluetooth local communication and user terminal module: This module includes a Bluetooth communication unit integrated into the local controller and an external user terminal (mobile APP). The Bluetooth communication unit is a built-in communication interface of the local controller, enabling bidirectional data interaction between the local controller and the user terminal. Users can customize parameters such as charging period, operating mode, and priority weight through a mobile app. After the command is transmitted to the local controller via Bluetooth, the local controller updates the charging strategy in real time. At the same time, the local controller uploads real-time photovoltaic power, charging current, battery SOC, historical charging curves and other operating data to the mobile app through the Bluetooth communication unit to realize the visualization and recording of data.

[0042] Cloud platform remote operation and maintenance and predictive optimization module This module is a remote intelligent decision-making terminal that establishes a wireless data interaction connection with the local controller. It is the "remote brain" of the local controller and mainly realizes the functions of multi-source data fusion, photovoltaic power prediction, multi-objective global optimization and optimization strategy distribution. At the same time, it receives the field operation data uploaded by the local controller to complete model calibration and rolling optimization. See Figure 2 The cloud platform remote operation and maintenance and predictive optimization module achieves its functions in collaboration with the local controller through the following steps: Multi-source meteorological data fusion and acquisition: The cloud platform acquires structured meteorological data for the next 24-48 hours through the meteorological bureau's public API, including ultraviolet radiation data, cloud coverage, rainfall probability, relative humidity, visibility, ambient temperature, etc. Combined with the user's local sunrise and sunset times, it is used to accurately locate the power generation range; Cloud cover and rainfall probability are used to correct the baseline illumination forecast; Photovoltaic power prediction modeling: The original irradiance data is corrected in multiple layers for basic attenuation, humidity, visibility and solar altitude angle to obtain the corrected irradiance Gpred(t), and then substituted into the formula to calculate the predicted photovoltaic power Ppv,pred(t). The formula for calculating the corrected irradiance Gpred(t) is as follows: Gpred(t) = G_pred_raw(t) × β_enhanced(t) × K_sunangle(t); β_enhanced (t) = β_base(t) × f_humidity × f_visibility; β_base (t) = 1 - [ω_cloud×C (t) + ω_rain×R (t)]; Wherein, β_enhanced(t) is the enhancement correction factor, β_base(t) represents the base attenuation factor, f_humidity represents the humidity correction factor, f_visibility represents the visibility correction factor, C(t) is the cloud cover rate, R(t) is the probability of rainfall, ω_cloud and ω_rain are weighting coefficients; K_sunangle(t) is the solar altitude angle correction coefficient, which is determined by the current date and time.

[0043] Time window: Based on sunrise and sunset times, the corrected Gpred(t) curve is set to zero during periods of no sunlight, retaining only data from the effective power generation period; Power calculation refers to inputting the corrected illumination data into the historically trained system model.

[0044] The predicted photovoltaic power The calculation formula is: ; in, For overall system efficiency, the value range is 0.75~0.85; This refers to the total area of ​​the photovoltaic panels; Temperature coefficient; To predict the ambient temperature, a value of -0.004℃ is used.

[0045] The following implementation case will further illustrate this step: The next-day forecast data (for some time periods) obtained from the meteorological API are shown in Table 3 below: Table 3

[0046] Step 1: Calculate the basic attenuation factor β_base(t); β_base(t) = 1 - [ω_cloud × C(t) + ω_rain × R(t)]; C(10:00) = 0.8, R(10:00) = 0.4; β_base(10:00) = 1 - (0.4×0.8 + 0.3×0.4) = 0.56; This means that due to the cloudy (80% cloud cover) and 40% probability of rainfall, the original predicted irradiance needs to be multiplied by a base attenuation factor of 0.56.

[0047] Step 2: Calculate the enhancement correction factor β_enhanced(t); The values ​​for f_humidity(RH) are determined as follows: When RH < 0.5: f_humidity(RH) = 1.0; When 0.5 ≤ RH ≤ 0.8: f_humidity(RH) = 1 - 0.2×(RH-0.5); When RH>0.8:f_humidity(RH) = 0.94 - 0.3×(RH-0.8); The values ​​for f_visibility(VIS) are determined as follows: When VIS ≥ 10km: f_visibility(VIS) = 1.0; When 5 ≤ VIS<10km: f_visibility(VIS) = 0.95; When 2 ≤ VIS<5km: f_visibility(VIS) = 0.85; When VIS<2km: f_visibility(VIS) = 0.7; 1. Humidity correction: RH(10:00) = 0.65 ∈ [0.5, 0.8] f_humidity = 1 - 0.2×(0.65-0.5); = 1 - 0.2 × 0.15; = 1 - 0.03; = 0.97; 2. Visibility correction: VIS(10:00) = 15km ≥ 10km; f_visibility = 1.0; 3. Enhance the correction factor: β_enhanced(10:00) = β_base × f_humidity × f_visibility = 0.56 ×0.97 × 1.0 = 0.5432; Step 3: Input the current date into a specific table (Table 4) to obtain the impact value of the solar altitude angle on the actual illumination in the region; Table 4 date time Solar altitude angle (degrees) K_sunangle Remark spring equinox 06:00 0.00 0.100 morning spring equinox 08:00 15.35 0.246 spring equinox 10:00 37.89 0.614 spring equinox 12:00 55.07 0.820 noon spring equinox 14:00 50.25 0.769 spring equinox 16:00 33.21 0.547 spring equinox 18:00 10.95 0.319 evening Summer Solstice 06:00 4.12 0.182 morning Summer Solstice 08:00 25.68 0.433 Summer Solstice 10:00 47.91 0.741 Summer Solstice 12:00 64.94 0.906 noon Summer Solstice 14:00 60.98 0.874 Summer Solstice 16:00 43.80 0.692 Summer Solstice 18:00 20.93 0.519 evening Autumn Equinox 06:00 0.00 0.100 morning Autumn Equinox 08:00 15.35 0.264 Autumn Equinox 10:00 37.89 0.614 Autumn Equinox 12:00 55.07 0.820 noon Autumn Equinox 14:00 50.25 0.769 Autumn Equinox 16:00 33.21 0.547 Autumn Equinox 18:00 10.95 0.319 evening winter solstice 06:00 0.00 0.100 morning winter solstice 08:00 5.23 0.205 winter solstice 10:00 26.78 0.449 winter solstice 12:00 43.08 0.681 noon winter solstice 14:00 38.28 0.628 winter solstice 16:00 21.29 0.526 winter solstice 18:00 0.00 0.100 evening The above scenarios are illustrated in Table 5 below, which contains meteorological and photovoltaic parameter data. Table 5 parameter symbol value unit illustrate Raw meteorological parameters - Raw irradiance G_raw(t) 980.00 W / m Weather station forecast Raw meteorological parameters - cloud cover C(t) 90.0% 0 - 1 Severely cloudy (90%) Raw meteorological parameters - probability of precipitation R(t) 50.0% 0 - 1 Moderate chance of rainfall (50%) Raw meteorological parameters - relative humidity RH(t) 60.0% 0 - 1 Moderate humidity (60%) Raw meteorological parameters - visibility VIS 25.00 km Good visibility Raw meteorological parameters - ambient temperature <![CDATA[T αmb ]]> 35.00 ℃ high temperature Original meteorological parameters - solar altitude angle correction K_sunangle 82.0% - The sun's altitude is moderate. Correction calculation - Basic attenuation factor β_base 49.0% - Cloud cover + rainfall reduction Correction Calculation - Humidity Correction Factor f_humidity 98.0% - Humidity Correction Calculation - Visibility Correction Factor f_visibility 100.0% - Visibility impact Correction calculation - Enhanced correction factor β_enhanced 48.0% - Overall attenuation Corrected calculation - Corrected irradiance <![CDATA[G pred (t)]]> 385.89 W / m Practical available irradiance Correction calculation - attenuation rate - 60.6% - Total attenuation ratio Photovoltaic power calculation - photovoltaic panel area <![CDATA[A pv ]]> 20.00 m System Scale Photovoltaic power calculation - temperature coefficient γ -0.4% / ℃ Typical values ​​of silicon photovoltaic panels Photovoltaic power calculation - temperature correction factor T_factor 104.0% - Attenuation caused by high temperature Photovoltaic power calculation - system efficiency <![CDATA[η sys (t)]]> 78.3% - Considering the influence of irradiance Photovoltaic power calculation - Predicting photovoltaic power <![CDATA[p pv,pred (t)]]> 6.29 kW Final output power Multi-objective global optimization solution: The target charging power curve Ptarget(t) is generated in two stages: planned generation and rolling correction, which satisfies the user's minimum charging power requirement Pdemand_min while minimizing the net electricity cost; Specifically as follows: Phase 1, Plan Generation (Based on Forecast Data): P target (t)≥P demand_min ; Phase 2, Rolling Correction (executed every hour): Used to adjust the schedule at time t and subsequent times; ; The goal is to maximize the benefits of photovoltaic self-consumption and participation in grid dispatch while minimizing net electricity costs, all while meeting user experience constraints. The relevant parameter information is shown in Table 6 below: Table 6

[0048] See Figure 3 The optimization strategy decomposition and distribution: Finally, the target charging power curve is decomposed into instructions that can be executed by the local controller, which are then distributed to the local controller via the MQTT protocol, and the running data uploaded by the local controller is received to complete the model calibration.

[0049] In this invention, the local controller includes a housing and a circuit board mounted inside the housing; The outer shell is made of aluminum alloy and includes a cover 1, a front plate 14, a base 6 and an aluminum alloy rear plate 4. A cooling fan 5 is installed on the aluminum alloy rear plate 4. The circuit board integrates an input reverse polarity protection MOSFET 8, a boost MOSFET 9, a freewheeling MOSFET 11, an output switch control MOSFET 13, a battery output line 2, a photovoltaic input line 3, an electrolytic capacitor, a power inductor 10, a digital tube display screen, and a control board 15. The electrolytic capacitor includes an input electrolytic capacitor 7 and an output electrolytic capacitor 12, achieving integrated hardware design and miniaturization.

[0050] Specifically, the local controller of this invention is an integrated hardware device, namely... Figure 4 The complete unit shown in the exhibition adopts an aluminum alloy shell structure 6, which is composed of an aluminum alloy cover 1, an aluminum alloy front plate 14, an aluminum alloy shell 6, and an aluminum alloy rear plate 4. The aluminum alloy rear plate 4 is equipped with a cooling fan 5, which adopts a "heat sink + forced convection" heat dissipation method. Among them, the input reverse polarity protection MOSFET 8, boost MOSFET 9, freewheeling MOSFET 11, and output switch control MOSFET 13 are all fixed on heat sinks coated with thermally conductive silicone grease, and the cooling fan 5 blows air inward to achieve efficient heat dissipation of the core components; see also Figure 5 This allows you to visually see the connection relationships between the various components on the circuit board; The battery output line has two OT2.5-8 terminals; The photovoltaic input line has an MC4 male and female connector.

[0051] The cloud platform remote operation and maintenance and prediction optimization module of this invention adopts a rolling decision-making closed loop of "prediction-optimization-execution-verification". Every hour, it uses the measured data (photovoltaic output, battery SOC, actual charging power) uploaded by the local controller to calibrate the prediction model and re-optimize the charging strategy for the next 24 hours. At the same time, the cloud platform has regional adaptive learning capabilities. Through big data analysis of the sunlight patterns and seasonal variation characteristics of different geographical areas, and combined with the peak and valley electricity price standards of the local power grid, it automatically generates personalized charging plans and sends them to the local controller. The plans can be dynamically updated according to the on-site feedback data.

[0052] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A smart charging system based on photovoltaic adaptive and Bluetooth cloud control, characterized in that, The system adopts a hybrid control architecture that prioritizes local operation while collaborating with cloud computing. The core includes a local controller, and modules that work collaboratively with the local controller, such as a photovoltaic data acquisition and control module, a Bluetooth local communication and user terminal module, and a cloud platform remote operation and maintenance and predictive optimization module. The collaborative relationship between each module and the local controller is as follows: The photovoltaic data acquisition and control module is a local core functional module with a local controller at its core. All its functions, including data acquisition, maximum power point tracking control, and power management, are executed by the local controller. The Bluetooth local communication and user terminal module includes a Bluetooth communication unit integrated into the local controller and an external user terminal. The Bluetooth communication unit is a built-in communication interface of the local controller, enabling bidirectional data interaction between the local controller and the user terminal. The cloud platform remote operation and maintenance and predictive optimization module is a remote decision-making terminal. It establishes a wireless data interaction connection with the local controller, sends charging optimization strategies to the local controller, and receives on-site operation data uploaded by the local controller to complete model calibration and rolling optimization. The local controller serves as the control center on the local side of the system. It receives data collected by the photovoltaic acquisition and control module and executes local decisions. It interacts with the user terminal through the Bluetooth communication unit, receives and executes the optimization strategies of the cloud platform's remote operation and maintenance and predictive optimization module, and simultaneously feeds back operating data to the cloud platform.

2. The intelligent charging system based on photovoltaic adaptive and Bluetooth cloud control according to claim 1, characterized in that, The photovoltaic data acquisition and control module is controlled by a local controller and achieves its functions through the following steps: Step 1: The local controller collects the photovoltaic panel voltage V at 1-second intervals. pv Current I pv Temperature T pv Battery voltage V b Current I b Real-time photoelectric data of battery state of charge (SOC); Step 2: Calculate the instantaneous photovoltaic power P based on the collected photoelectric data. pv (t) and the rate of change of power voltage The calculation formula is as follows: P pv (t) = V pv (t)*I pv (t); ； Step 3: Perform adaptive step size adjustment, as shown in the following formula: ΔD (t+1) = K1*(dPpv / dVpv) + K2*f (Tpv, SOC); Where ∆D(t+1) represents the duty cycle adjustment amount to be applied in the next control cycle, K1 is the MPPT tracking gain coefficient, K2 is the environmental and state compensation coefficient, (dPpv / dVpv) represents the slope of the photovoltaic power curve, f(Tpv,SOC) is the temperature compensation and battery state correction value, and Tpv represents the photovoltaic panel temperature; Step 4: Dynamically adjust the charging mode according to the battery's State of Charge (SOC). When SOC < 0.8, use constant current fast charging; when SOC ≤ 0.8 < 0.95, use constant voltage charging; and when SOC ≥ 0.95, use trickle charging. Step 5: Execute the local power management strategy and allocate photovoltaic power and grid power according to the operating mode and strategy preference parameters.

3. The intelligent charging system based on photovoltaic adaptive and Bluetooth cloud control according to claim 2, characterized in that, In the local power management policy executed by the local controller, the policy preference parameter The formula is as follows: ； in, This indicates the adjustment of the distribution ratio between photovoltaic power and grid power. The operating mode selected by the user For real-time electricity prices, User priority weight, The time factor; The user-selected operating mode This includes photovoltaic priority mode, economic mode, and off-peak electricity priority mode. The local controller has built-in the corresponding modes for each. The mapping calculation function is dynamically solved based on real-time parameters. .

4. The intelligent charging system based on photovoltaic adaptive and Bluetooth cloud control according to claim 1, characterized in that, The bidirectional data interaction between the Bluetooth local communication and the user terminal module includes: External user terminals send custom setting instructions for charging period, operating mode, and priority weight to the local controller via Bluetooth. The local controller updates the charging strategy immediately after receiving the instructions. The local controller uploads real-time photovoltaic power, charging current, battery state of charge (SOC), and historical charging curve data to an external user terminal via Bluetooth communication unit, enabling data visualization.

5. The intelligent charging system based on photovoltaic adaptive and Bluetooth cloud control according to claim 1, characterized in that, The cloud platform remote operation and maintenance and predictive optimization module works in conjunction with the local controller to achieve its functions through the following steps: Step 1: Obtain structured forecast data for the next few hours through the meteorological bureau's public API, including ultraviolet radiation data, geographic location and time data, cloud cover, probability of rainfall, relative humidity, visibility, and ambient temperature; Step 2: Perform multi-layer corrections on the original irradiance data based on basic attenuation, humidity, visibility, and solar altitude angle to obtain the corrected irradiance Gpred(t), and substitute it into the formula to calculate the predicted photovoltaic power Ppv,pred(t). Step 3: Generate the target charging power curve Ptarget(t) in two stages: planned generation and rolling correction, to meet the user's minimum charging power requirement Pdemand_min while minimizing the net electricity cost; Step 4: Decompose the target charging power curve into instructions that can be executed by the local controller, send them to the local controller via the MQTT protocol, and receive the running data uploaded by the local controller to complete the model calibration.

6. The intelligent charging system based on photovoltaic adaptive and Bluetooth cloud control according to claim 5, characterized in that, The formula for calculating the corrected irradiance Gpred(t) is as follows: Gpred(t) = G_pred_raw(t) × β_enhanced(t) × K_sunangle(t); β_enhanced (t) = β_base(t) × f_humidity × f_visibility; β_base (t) = 1 - [ω_cloud×C (t) + ω_rain×R (t)]; Wherein, β_enhanced(t) is the enhancement correction factor, β_base(t) represents the base attenuation factor, f_humidity represents the humidity correction factor, f_visibility represents the visibility correction factor, C(t) is the cloud cover rate, R(t) is the probability of rainfall, ω_cloud and ω_rain are weighting coefficients; K_sunangle(t) is the solar altitude angle correction coefficient, which is determined by the current date and time.

7. The intelligent charging system based on photovoltaic adaptive and Bluetooth cloud control according to claim 5, characterized in that, The predicted photovoltaic power The calculation formula is: ； in, For overall system efficiency, the value range is 0.75~0.85; This refers to the total area of ​​the photovoltaic panels; Temperature coefficient; To predict ambient temperature.

8. The intelligent charging system based on photovoltaic adaptive and Bluetooth cloud control according to claim 1, characterized in that, The cloud platform's remote operation and maintenance and predictive optimization module has regional adaptive learning capabilities. It analyzes the sunlight patterns and seasonal variation characteristics of different geographical regions through big data analysis, and generates personalized charging plans in conjunction with local power grid prices. These plans are then sent to the local controller for execution, and the personalized charging plans are dynamically updated based on the operating data fed back by the local controller.

9. The intelligent charging system based on photovoltaic adaptive and Bluetooth cloud control according to claim 1, characterized in that, The local controller has built-in operating mode switching logic, which automatically switches between three operating modes based on the power grid status and cloud connection status: Emergency mode: Triggered when the power grid is abnormal or malfunctioning, the control objective is to ensure system safety; Planned Mode: Triggered when the cloud-based optimization strategy is effective, executing the optimized charging plan issued by the cloud platform's remote operation and maintenance and predictive optimization module; Autonomous Mode: Triggered when there is no cloud-based optimization strategy, the local controller independently completes all functions of the photovoltaic acquisition and control module, achieving optimal allocation of local photovoltaic power and grid power.

10. The intelligent charging system based on photovoltaic adaptive and Bluetooth cloud control according to claim 1, characterized in that, The local controller includes a housing and a circuit board mounted inside the housing; The outer casing is made of aluminum alloy and includes a cover, a front panel, a base, and a rear panel. A cooling fan is installed on the rear panel. The circuit board integrates an input reverse polarity protection MOSFET, a boost MOSFET, a freewheeling MOSFET, an output switch control MOSFET, a battery output line, a photovoltaic input line, an electrolytic capacitor, a power inductor, a digital tube display screen, and a control board.

Images