Optical storage coordination management method, optical storage integration system, storage medium and device

Abstract

The application discloses a kind of light storage collaborative management method, light storage integrated system, storage medium and equipment.The method is applied to the light storage integrated system including at least four not all coplanar light-receiving surfaces of photovoltaic module, and the method comprises: short-circuit current and open-circuit voltage are measured by sampling each photovoltaic sub-string in time;Determine the solar direction vector, direct irradiance and scattered irradiance according to the corresponding relationship of short-circuit current and light-receiving surface normal and solar direction vector, and determine the board surface attitude angle by matching with the solar direction vector determined by astronomical calendar;The photovoltaic sub-string with measured value lower than theoretical value is judged as blocked sub-string and the blocking degree is output, and the above parameters are determined again after being removed;Determine the reduction factor according to the corrected irradiance to determine the weather condition;The photovoltaic power generation capacity is predicted according to the corrected board surface attitude angle, the upper limit of each power supply output branch power is determined combined with the reduction factor and the available energy of energy storage, and the power distribution is controlled.The application can effectively ensure the power supply reliability of system and reduce experience loss.

Description

Technical Field

[0001] This application relates to the field of portable optical storage technology, and in particular to a method for collaborative management of optical storage, an integrated optical storage system, a storage medium and device. Background Technology

[0002] In off-grid power supply scenarios such as camping, field operations, and emergency rescue, portable photovoltaic-storage systems, consisting of foldable photovoltaic modules, energy storage batteries, and power conversion devices, have become mainstream equipment. Currently, these systems have relatively mature technical solutions in terms of structural form and energy management. Regarding structural integration, existing products generally house foldable photovoltaic modules within suitcases, backpacks, or cases, integrating them with battery modules, inverters, control modules, and operation panels to form a portable and quickly deployable integrated device. In terms of photovoltaic module encapsulation, folding areas are typically reserved between multiple cells, and flexible wiring is laid to enable series, parallel, or mixed series-parallel connections between multiple light-receiving panels, balancing the folding lifespan of the modules with electrical output characteristics. Regarding energy distribution and scheduling, existing systems generally collect operating parameters such as remaining battery power, ambient temperature, input and output power, and photovoltaic power generation, receive a user-defined load priority list, and use dynamic allocation... Power allocation strategies are generated using algorithms, fuzzy control, or predictive models based on time-series data. These strategies reduce power or put non-critical loads into standby mode. Some solutions also incorporate geographic location information and real-time weather conditions to estimate the amount of energy that can be retained. In terms of photovoltaic module status sensing, it is known in the industry that the irradiance received by the light-receiving surface can be calculated using the short-circuit current of the photovoltaic module itself, and the module junction temperature can be calculated using the open-circuit voltage. This technology has been applied in module-level operation monitoring and fault diagnosis in fixed-installation photovoltaic power plants. In addition, for low-temperature environments, existing portable energy storage devices are generally equipped with thermal management methods such as electric heating or battery self-heating to raise the cell temperature to the temperature window for normal charging and discharging.

[0003] However, the aforementioned technical solutions still face several unresolved technical problems in the specific scenario of camping. First, the tilt angle and orientation of the modules in a fixed photovoltaic power station are determined and pre-entered during installation, and the system can directly calculate the solar incidence angle at any time and predict the power generation based on the astronomical calendar. However, in a camping scenario, foldable photovoltaic modules are casually laid out, leaned against, or hung by the user, and their spatial orientation and tilt angle are completely random. They may also be moved multiple times during the camping period, blown over by the wind, or touched by people. The system cannot know the current posture of the modules. Although some solutions attempt to combine geographical location and networked weather data to estimate the amount of electricity that can be retained, networked weather data usually only provides the total horizontal irradiance, which cannot be converted into the actual power generation of the modules when the posture of the modules is unknown. In addition, campsites often lack stable network conditions. Therefore, all energy management strategies that rely on power generation prediction are based on unreliable input. Second, existing technologies for inverting irradiance and junction temperature using photovoltaic module parameters implicitly assume a single, known light-receiving surface. The inversion result is merely a scalar measure of the total irradiance falling on that surface, unable to distinguish whether the irradiance originates from strong oblique sunlight or weak direct sunlight. Therefore, it cannot provide effective support for capacity prediction when the panel orientation is unknown. Third, campsites are often located under trees, beside rock walls, or next to tents. Photovoltaic modules are easily affected by the dynamic local shading of tree shadows, tent shadows, and pedestrians. Furthermore, shadows slowly sweep across the panel surface as the sun moves, causing current collapse in the shaded areas. Directly using this for parameter inversion leads to an underestimation of irradiance and misjudgment of trends. Existing shading handling techniques primarily employ binary judgments to avoid false alarms, failing to address the issue of obtaining accurate environmental parameter estimates even with shading present. Fourth, existing energy management systems often employ periodic rolling optimization without a clear end point. However, camping is a finite-term activity with a defined end time. Users are concerned with ensuring continuous power supply before breakup and maintaining sufficient capacity for the return trip, rather than achieving instantaneous optimization at every moment. Therefore, existing management methods are prone to situations where there is ample power supply initially, followed by a sudden power outage later. Fifth, when energy shortages necessitate load reduction, existing solutions typically determine the reduction order based solely on a static priority table. This fails to quantify the impact of each power reduction level on user experience, making it difficult to achieve refined reduction that balances power reliability with user experience degradation. Summary of the Invention

[0004] This application mainly provides a method for coordinated management of photovoltaic and energy storage, an integrated photovoltaic and energy storage system, a storage medium and equipment, aiming to solve the technical problem that existing integrated photovoltaic and energy storage systems are unable to achieve a fine reduction in both power supply reliability and experience loss.

[0005] To solve the above-mentioned technical problems, the technical solution adopted in this application is: to provide a photovoltaic-storage collaborative management method, applied to a photovoltaic-storage integrated system, wherein the photovoltaic-storage integrated system includes photovoltaic modules, energy storage units, sampling circuits, power conversion units, and power output branches, wherein the photovoltaic modules have no less than four light-receiving surfaces, the normals of each light-receiving surface are not all coplanar, and each light-receiving surface corresponds to an independent photovoltaic sub-string; the photovoltaic-storage collaborative management method includes: sampling each photovoltaic sub-string in a time-division manner to measure the short-circuit current and open-circuit voltage of each photovoltaic sub-string; based on the correspondence between the short-circuit current and the normal of the light-receiving surface and the solar direction vector, determining the solar direction vector, direct irradiance, and diffuse irradiance from the short-circuit current of each photovoltaic sub-string and the normal of each light-receiving surface, and determining the panel attitude angle of the photovoltaic module based on the deviation between the solar direction vector and the solar direction vector determined according to the astronomical ephemeris; and determining the solar direction angle from the determined solar direction vector. The theoretical value of the short-circuit current of each photovoltaic sub-string is obtained from the vector, the direct irradiance, and the diffuse irradiance. Photovoltaic sub-strings whose measured short-circuit current is lower than the theoretical value by a preset degree are identified as shading sub-strings and the shading degree is output. After the shading sub-strings are removed, the solar direction vector, the direct irradiance, the diffuse irradiance, and the panel attitude angle are re-determined from the remaining photovoltaic sub-strings. The weather conditions are judged based on the corrected direct irradiance and diffuse irradiance, and the weather condition reduction factor is determined. The incident irradiance of each light-receiving surface in the time domain is predicted based on the corrected panel attitude angle, and the photovoltaic power generation prediction curve is obtained by combining the weather condition reduction factor. The photovoltaic power generation prediction curve and the available energy of the energy storage unit are used as inputs to determine the power upper limit of each power output branch, and the power conversion unit is controlled to distribute power among the photovoltaic modules, the energy storage unit, and each power output branch according to the power upper limit.

[0006] Optionally, the correspondence between the short-circuit current and the normal and solar direction vector of the light-receiving surface is as follows: the short-circuit current of the photovoltaic substring corresponding to any light-receiving surface is equal to the product of the current response coefficient of the photovoltaic substring and the sum of the following two items, multiplied by a temperature correction term: the first item is the product of the direct irradiance and the positive value of the cosine of the angle between the normal of the light-receiving surface and the solar direction vector; the second item is the product of the diffuse irradiance and the sky view factor of the light-receiving surface; the temperature correction term is determined based on the open-circuit voltage; the determination of the solar direction vector, direct irradiance, and diffuse irradiance includes: solving the azimuth and elevation angles of the solar direction vector, the direct irradiance, and the diffuse irradiance by simultaneously solving the correspondences corresponding to each light-receiving surface.

[0007] Optionally, determining a photovoltaic substring whose measured short-circuit current is lower than the theoretical value by a preset degree as a shading substring includes: determining a photovoltaic substring whose difference between the theoretical value and the measured value of the short-circuit current is greater than a preset shading threshold as a shading substring; defining the shading degree as the ratio of the difference between the theoretical value and the measured value of the short-circuit current of the shading substring to the theoretical value; and using the shading degree as an independent output to drive the power optimization control of the shading substring.

[0008] Optionally, the step of determining weather conditions based on the corrected direct irradiance and diffuse irradiance includes: defining the sharpness index as the ratio of direct irradiance to the sum of direct irradiance and diffuse irradiance; determining the weather conditions into at least two types based on the mean and standard deviation of the sharpness index within a sliding time window, and assigning a decreasing weather condition reduction factor to each weather condition type; and maintaining the weather condition reduction factor determined in the previous period when the mean and standard deviation do not fall into any discrimination interval.

[0009] Optionally, after determining the weather conditions based on the corrected direct irradiance and diffuse irradiance, the method further includes: when the clarity index is lower than a preset ill-conditioned threshold, maintaining the panel attitude angle of the previous cycle, determining the total effective irradiance only by the sum of the short-circuit currents of all the photovoltaic sub-strings, and marking the confidence level of the photovoltaic power generation prediction as low.

[0010] Optionally, determining the upper limit of the power output of each of the power supply output branches includes: allocating available energy in time periods based on the constraint that the available energy of the energy storage unit at the end of the power supply task is not lower than the preset reserved energy and the available energy of the energy storage unit throughout the process is not lower than the safety lower limit; when the allocation of a certain scheduling period causes the subsequent period to fail to meet the constraint, backtracking and compressing the upper limit of the power of the earlier period.

[0011] Optionally, after determining the power upper limit of each of the power output branches, the method further includes: adopting a continuous flexible load reduction method for the lighting loads in the power output branches, quantifying the subjective brightness utility loss corresponding to the unit power reduction based on the logarithmic law of human eye subjective brightness; and performing lighting load reduction in order of the utility loss from smallest to largest.

[0012] To solve the above-mentioned technical problems, another technical solution adopted in this application is to provide an integrated photovoltaic and energy storage system. This integrated system includes photovoltaic modules, an energy storage unit, a switch matrix, a sampling circuit, a power conversion unit, a power output branch, and a controller. The photovoltaic modules have at least four light-receiving surfaces, the normals of which are not all coplanar, and each light-receiving surface corresponds to an independent photovoltaic sub-string. The input terminal of the switch matrix is ​​connected to each photovoltaic sub-string, the first output terminal of the switch matrix is ​​connected to the DC bus side of the power conversion unit, and the second output terminal of the switch matrix is ​​connected to the sampling circuit, used to disconnect the sampled photovoltaic sub-string from the power conversion unit and connect it to the sampling circuit during sampling. The output terminal of the circuit is connected to the analog-to-digital conversion interface of the controller to measure the short-circuit current and open-circuit voltage of each photovoltaic sub-string; the DC bus side of the power conversion unit is also connected to the energy storage unit, and the output side of the power conversion unit is connected to the power supply output branch to distribute power among the photovoltaic modules, the energy storage unit and each power supply output branch; the controller includes a real-time clock, and the controller is connected to the control terminal of the switch matrix, the output terminal of the sampling circuit, the communication interface of the energy storage unit and the control terminal of the power conversion unit respectively, to execute the steps of the photovoltaic-energy storage collaborative management method described above.

[0013] To address the aforementioned technical problems, another technical solution adopted in this application is to provide a storage medium. This storage medium stores program data, which, when executed by a processor, implements the steps of the aforementioned optical-storage collaborative management method.

[0014] To address the aforementioned technical problems, another technical solution adopted in this application is to provide a computer device. This computer device includes a processor and a memory, the memory storing a computer program, and the processor executing the computer program to implement the steps of the aforementioned optical-storage collaborative management method.

[0015] The beneficial effects of this application are as follows: Unlike existing technologies, this application discloses a photovoltaic-storage collaborative management method, an integrated photovoltaic-storage system, a storage medium, and equipment. This application utilizes time-division sampling of short-circuit currents and open-circuit voltages obtained from photovoltaic substrings corresponding to at least four non-fully coplanar light-receiving surfaces of a folded photovoltaic module. It then simultaneously solves for the solar direction vector, direct irradiance, and diffuse irradiance, and registers these with an astronomical calendar to determine the panel attitude angle. Without adding any external attitude sensors, GPS modules, or network communication modules, it achieves autonomous perception of the panel attitude solely using the electrical quantities of the photovoltaic module itself, solving the core technical problem of unreliable capacity prediction due to unknown photovoltaic module attitude in camping scenarios. Furthermore, by removing shading substrings from the set of substrings participating in the inversion and resolving the problem, accurate environmental parameter estimates are still obtained even with dynamic shading. Simultaneously, the shading degree is used as an independent output for driving power optimization control, transforming the shading state from an interference that needs to be eliminated into a useful output. Furthermore, by using a sliding statistic based on the clarity index to determine weather conditions and establish a weather reduction factor, a basis for weather condition correction in production capacity forecasting is provided without relying on external meteorological information. Finally, by allocating available energy in time periods based on the terminal's reserved energy at the end of the power supply task, and by retrospectively compressing the power upper limit of earlier periods when the constraint cannot be met, combined with a flexible load reduction strategy for lighting based on the logarithmic law of subjective human eye brightness, refined energy management is achieved that balances power supply reliability with user experience loss, ensuring continuous power supply during camping and reserving the necessary capacity for the return trip. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort, wherein: Figure 1 This is a flowchart illustrating an embodiment of the photovoltaic-storage collaborative management method provided in this application; Figure 2 This is an overall block diagram of an embodiment of the integrated photovoltaic and energy storage system provided in this application; Figure 3 This is a schematic diagram of the structure of an embodiment of the storage medium provided in this application; Figure 4 This is a schematic diagram of the structure of an embodiment of the computer device provided in this application. Detailed Implementation

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0018] The terms "first," "second," and "third" used in the embodiments of this application are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first," "second," or "third" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0019] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0020] This application provides a method for coordinated management of photovoltaic and energy storage systems, see reference. Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of the photovoltaic-storage collaborative management method provided in this application. The method is applied to an integrated photovoltaic-storage system, which includes photovoltaic modules, an energy storage unit, a sampling circuit, a power conversion unit, and a power output branch. The photovoltaic modules have at least four light-receiving surfaces, and the normals of these surfaces are not all coplanar. Each light-receiving surface corresponds to an independent photovoltaic sub-string. The photovoltaic-storage collaborative management method includes the following steps: Step 101: Sample each photovoltaic substring in a time-division manner to measure the short-circuit current and open-circuit voltage of each photovoltaic substring.

[0021] In this embodiment, the provided photovoltaic-storage collaborative management method is applied to an integrated photovoltaic-storage system, which includes photovoltaic modules, energy storage units, sampling circuits, power conversion units, and power output branches. For the specific composition and connection relationships of this integrated photovoltaic-storage system, please refer to the subsequent sections. Figure 2The described embodiments are only illustrated with respect to the photovoltaic modules and sampling mechanism related to step 101.

[0022] In this embodiment, the photovoltaic module is a portable foldable photovoltaic module with no fewer than four light-receiving surfaces. The normals of each light-receiving surface are not all coplanar, and each light-receiving surface corresponds to an independently extended photovoltaic sub-string. The normal is the direction vector perpendicular to the surface of the light-receiving surface and pointing towards the light-receiving side; the normals are not all coplanar, meaning that at least two of the normal vectors of each light-receiving surface are not in the same plane. The number of light-receiving surfaces of the photovoltaic module is N, where N is an integer greater than or equal to four. The reason for limiting it to no fewer than four light-receiving surfaces is that the unknowns that need to be solved simultaneously in the subsequent step 102 include two degrees of freedom (azimuth and altitude angles of the solar direction vector), and four unknowns (direct irradiance and diffuse irradiance). The short-circuit current corresponding to each light-receiving surface provides an observation equation, so no fewer than four equations are needed to solve for these four unknowns simultaneously. When N equals four, the normals of the four light-receiving surfaces are not all coplanar, forming a perfectly determined system of equations that can be solved simultaneously; when N is greater than four, the number of equations exceeds the number of unknowns, forming an overdetermined system of equations, which has stronger resistance to shading and noise.

[0023] In this embodiment, the photovoltaic module configuration can be implemented in several ways. In one implementation, the photovoltaic module adopts a four-fold W-shaped configuration, including a base light-receiving surface and three side light-receiving surfaces. The base light-receiving surface is placed horizontally, and the three side light-receiving surfaces are folded at different angles along different sides, so that the normals of the four light-receiving surfaces are not all coplanar. In another implementation, the photovoltaic module adopts a five-fold configuration, including a base light-receiving surface and four side light-receiving surfaces. The normals of the five light-receiving surfaces are not all coplanar, forming an overdetermined system of equations. After removing one shading substring, four unshading substrings remain, which can be completely resolved, and the anti-shading capability is better than that of the four-fold configuration. In another implementation, for existing three-fold V-shaped photovoltaic module product lines, without altering the main structure, a small independent reference cell is added as a fourth light-receiving surface to the handle or frame of the three-fold V-shaped main body. Its normal is intentionally designed to be non-coplanar with the normals of the three folded light-receiving surfaces. This independent reference light-receiving surface is also connected to the switching matrix for time-division sampling, but due to its small area, its power generation contribution is negligible. All of the above implementation methods can ensure that the photovoltaic module meets the solvability conditions for simultaneous solutions. The specific choice can be flexibly determined based on product form, cost budget, and anti-shading requirements.

[0024] In this embodiment, to obtain the raw electrical quantities for solar direction vector inversion without adding external attitude sensors, a time-division sampling method is used to sample each photovoltaic sub-string. Time-division sampling involves sampling each photovoltaic sub-string sequentially over time, with only one sub-string in sampling mode at any given time, while the others continue generating electricity normally. Each photovoltaic sub-string participates in normal photovoltaic power generation by connecting to the DC bus through its respective power generation branch switch. When sampling of a particular photovoltaic sub-string is required, the controller controls the switch matrix to disconnect the power generation branch switch of that sub-string, temporarily disconnecting it from the power generation branch. Simultaneously, a short-circuit sampling switch is closed, connecting the sub-string to the shunt resistor branch of the sampling circuit to measure its short-circuit current. Subsequently, the short-circuit sampling switch is opened, and the voltage sampling switch is closed, connecting the sub-string to the high-resistance voltage divider branch of the sampling circuit to measure its open-circuit voltage. After completing one short-circuit current and open-circuit voltage measurement, the power generation branch switch of that sub-string is restored, allowing it to participate in power generation again.

[0025] In this embodiment, within a single sampling window, only the sampled photovoltaic sub-string disconnects from power generation, while the remaining photovoltaic sub-strings continue to generate power as usual, thus ensuring the overall continuity of power generation of the system remains unaffected. Let the sampling window duration be... The interval between two adjacent samples is The number of photovoltaic substrings is The proportion of power generation loss due to sampling is... Divide by , with the number of photovoltaic substrings Irrelevant. Since both the plane attitude and weather conditions are slow variables, high-frequency updates are unnecessary. Larger values ​​can be taken (e.g.) Take 20 milliseconds Taking a time interval of 3 to 10 seconds, the power generation loss ratio is approximately 0.2% to 0.7%, thus keeping the power generation loss at a low level.

[0026] This embodiment uses time-division sampling to sequentially enter the sampling window for each photovoltaic substring. By sacrificing only a minimal amount of power generation time, it obtains the short-circuit current and open-circuit voltage of each photovoltaic substring, providing a data foundation for subsequent solar direction vector inversion. Since only one photovoltaic substring disconnects from power generation during a single sampling, while the remaining substrings continue generating power as usual, the overall power generation continuity of the system remains unaffected.

[0027] Optionally, in some embodiments, the power generation branch switch of the sampled photovoltaic substring is first disconnected within the sampling window, and after a stabilization delay, the short-circuit sampling switch is closed to measure the short-circuit current. Then, the short-circuit sampling switch is disconnected and the open-circuit voltage is measured after a stabilization delay. Subsequently, the power generation branch switch is restored. The duration of the sampling window is not greater than a preset window threshold, and the sampling interval between two adjacent samplings of the same photovoltaic substring is not less than a preset interval threshold.

[0028] In this optional embodiment, a stabilization delay is used to wait for the circuit transition process to complete before measurement, avoiding the impact of transient noise during switching on sampling accuracy. For example, the stabilization delay is 1 to 5 milliseconds. A preset window threshold limits the maximum duration of a single sampling, and a preset interval threshold ensures sufficient recovery time between two samples from the same photovoltaic substring. The analog-to-digital converter of the sampling circuit has an accuracy of no less than 16 bits, a sampling rate of no less than 10,000 times per second, and the shunt resistor uses a low-resistance precision resistor (e.g., 1 milliohm, 1% accuracy, temperature drift no more than 50 ppm per degree Celsius) to ensure the measurement accuracy of the short-circuit current.

[0029] This optional embodiment ensures that the measurement of short-circuit current and open-circuit voltage is performed under steady-state circuit conditions by setting a stabilization delay within the sampling window, thereby improving sampling accuracy.

[0030] Step 102: Based on the correspondence between the short-circuit current and the normal of the light-receiving surface and the solar direction vector, determine the solar direction vector, direct irradiance and diffuse irradiance by the short-circuit current of each photovoltaic sub-string and the normal of each light-receiving surface, and determine the panel attitude angle of the photovoltaic module based on the deviation between the solar direction vector and the solar direction vector determined according to the astronomical calendar.

[0031] In this embodiment, the solar direction vector is a unit vector representing the spatial direction of the sun relative to the observation point, determined by two degrees of freedom: solar azimuth and solar altitude. Direct irradiance is the component of sunlight that reaches the illuminated surface directly along the solar direction without being scattered by clouds or other media. Scattered irradiance is the component of sunlight that reaches the illuminated surface from various directions in the sky after being scattered by the atmosphere and clouds. The sky view factor characterizes the proportion of sky-scattered irradiance that the illuminated surface can receive, depending on the tilt angle of the illuminated surface relative to the horizontal plane. The current response coefficient is the ratio of the short-circuit current of the photovoltaic substring under standard test conditions to the received irradiance, reflecting the photovoltaic substring's ability to convert irradiance into short-circuit current. The panel attitude angle is a parameter characterizing the orientation and tilt angle of the photovoltaic module in space, including the tilt angle and azimuth angle of the reference illuminated surface.

[0032] In this embodiment, the correspondence refers to the physical relationship between the short-circuit current of the photovoltaic substring corresponding to any light-receiving surface and the normal, solar direction vector, direct irradiance, and diffuse irradiance of that light-receiving surface. Specifically, the short-circuit current of the photovoltaic substring is approximately linearly related to the irradiance received by the light-receiving surface, while the direct irradiance received by the light-receiving surface varies with the cosine of the angle between the normal of the light-receiving surface and the solar direction. Therefore, light-receiving surfaces with different orientations have different responses to the same solar direction—the closer the normal of the light-receiving surface is to the solar direction, the more direct irradiance it receives and the larger the short-circuit current; the further the normal of the light-receiving surface is from the solar direction, the less direct irradiance it receives and the smaller the short-circuit current. This differentiated response constitutes the physical basis for inverting the solar direction from the short-circuit current of multiple light-receiving surfaces. The temperature correction term is determined based on the junction temperature obtained by inverting the open-circuit voltage of the photovoltaic substring and is used to correct for changes in the short-circuit current when the junction temperature deviates from the standard test conditions.

[0033] In this embodiment, the solar direction vector is described by two parameters: azimuth and altitude, plus direct irradiance and diffuse irradiance, totaling four unknowns. Since the photovoltaic module has at least four light-receiving surfaces, each corresponding to one of the above-mentioned equations, the number of equations is at least four, which is sufficient to solve for these four unknowns simultaneously. The mathematical necessity of solving this simultaneous equation lies in the fact that the solar direction vector has two degrees of freedom (azimuth and altitude), plus direct irradiance and diffuse irradiance, totaling four unknowns. Each light-receiving surface provides one equation, therefore at least four equations are needed for a simultaneous solution. The fact that not all normals are coplanar is a key geometric constraint: if the normals of all light-receiving surfaces are in the same plane, the response of each surface to the solar direction lacks resolution perpendicular to that plane, reducing the rank of the equation set and making it impossible to fully determine the azimuth and altitude angles of the solar direction vector. When the number of light-receiving surfaces exceeds four, the number of equations exceeds the number of unknowns, forming an overdetermined equation set, further enhancing the accuracy of the inversion and its resistance to shading. The solution can be obtained by at least one of the following methods, including but not limited to a combination of grid search and nonlinear iteration, genetic algorithm, and particle swarm optimization.

[0034] In this embodiment, the controller includes a real-time clock that provides the current date and time. An astronomical calendar is a record of the sun's position on the celestial sphere; the sun's direction vector in the horizontal coordinate system can be calculated based on the date, time, and geographical latitude and longitude. Based on the date and time, the sun's direction vector in the horizontal coordinate system can be calculated using the astronomical calendar; this direction vector represents the sun's absolute direction in the Earth's reference frame. The solar direction vector obtained through sampling inversion is registered and compared with the solar direction vector determined by the astronomical calendar. The deviation is the directional difference between the inverted solar direction vector and the solar direction vector determined by the astronomical calendar, determined by solving a rotational transformation that aligns the two direction vectors. The deviation represents the attitude deflection of the photovoltaic module relative to the horizontal plane. Therefore, this application simultaneously obtains both the solar direction vector and the panel attitude angle spatial information without the need for additional sensors.

[0035] This embodiment utilizes the differentiated responses of at least four non-fully coplanar solar-facing surfaces of a photovoltaic module to the solar direction. By linking the short-circuit currents of each photovoltaic sub-string, the solar direction vector, direct irradiance, and diffuse irradiance are determined simultaneously. This is then registered with the solar direction vector determined by the astronomical ephemeris to inversely determine the panel attitude angle, achieving a leap from single scalar inversion to vector-level inversion. Compared to the limitations of a single solar-facing surface, which only inverts the total irradiance (a scalar quantity) falling on that surface and cannot distinguish between strong oblique sunlight and weak orthogonal sunlight, this multi-surface vector inversion can simultaneously separate direct and diffuse irradiance and determine the solar direction. This provides a geometric basis for energy production prediction under unknown attitude conditions that a single solar-facing surface solution cannot offer. Therefore, this application does not require the addition of any dedicated solar sensors, inclinometers, electronic compasses, or positioning units for environmental perception. Instead, it uses each photovoltaic sub-string itself as a sampling unit in a time-division manner, directly obtaining the raw electrical quantities for vector inversion from the same set of power generation units. This overcomes the cognitive barrier of viewing the power generation unit and attitude measurement unit as independent functional modules, and represents an improvement that overcomes technical bias.

[0036] Optionally, in some embodiments, the correspondence between the short-circuit current and the normal and solar direction vector of the light-receiving surface is as follows: the short-circuit current of the photovoltaic substring corresponding to any light-receiving surface is equal to the product of the current response coefficient of the photovoltaic substring and the sum of the following two items, multiplied by a temperature correction term: the first item is the product of the direct irradiance and the positive value of the cosine of the angle between the normal and solar direction vector of the light-receiving surface, and the second item is the product of the diffuse irradiance and the sky view factor of the light-receiving surface; the temperature correction term is determined based on the open-circuit voltage.

[0037] In this optional embodiment, the above correspondence constitutes the observation model. For the first... For each light-receiving surface, the short-circuit current of the corresponding photovoltaic substring satisfies the following formula: ; in, For the first The short-circuit current of each photovoltaic substring, in amperes (A). For the first The current response coefficient of a photovoltaic substring, measured in amperes per watt per square meter (A·m² / W), is related to factors such as cell area and photoelectric conversion efficiency, and can be obtained through standard test conditions. Direct irradiance, measured in watts per square meter (W / m²), with a typical range of 0 to 1000 W / m²; For the first The angle between the normal of the illuminated surface and the solar direction vector is expressed in degrees (°) and ranges from 0° to 180°. This means that when the angle is greater than 90°, it is zero, that is, when the sun is behind the illuminated surface, the illuminated surface no longer receives direct radiation. The diffuse irradiance is expressed in watts per square meter (W / m²), with a typical range of 0 to 300 W / m². For the first The sky view factor of a light-receiving surface is dimensionless and ranges from 0 to 1, representing the proportion of sky-scattered irradiance that the light-receiving surface can receive. This is a temperature correction term, based on the junction temperature. Sure; The junction temperature of the photovoltaic substring is expressed in degrees Celsius (°C) and can be obtained through open-circuit voltage inversion. This formula relates the short-circuit current of the photovoltaic substring to the solar direction, direct irradiance, diffuse irradiance, normal to the light-receiving surface, and junction temperature, thus forming the observation model for vector inversion in this application.

[0038] In this optional embodiment, when solving the simultaneous equations, the four unknowns are the solar azimuth angle. Solar altitude angle Direct irradiance With diffuse irradiance The solar direction vector is determined by... and It is determined that the normals of each light-receiving surface are known in the photovoltaic module's body coordinate system (the body coordinate system is fixed to the photovoltaic module, with the center of the reference light-receiving surface as the origin and the normal of the reference light-receiving surface as the Z-axis). Once the panel's attitude angle is determined, the directions of the normals of each light-receiving surface in the horizontal coordinate system are also determined accordingly. It can be calculated by the dot product of the solar direction vector and the normal to the illuminated surface. Sky perspective factor. Depends on the angle of the light-receiving surface relative to the horizontal plane You can press Calculation. Solve the azimuth and elevation angles of the solar direction vector, as well as the direct and diffuse irradiance, by simultaneously solving the observation models corresponding to each illuminated surface.

[0039] This optional embodiment establishes an observation model by relating the short-circuit current to the solar direction, irradiance, normal of the light-receiving surface, and junction temperature. It transforms the determination of the panel attitude angle into a problem of solving a set of four unknowns. It utilizes the fact that the photovoltaic module itself has at least four light-receiving surfaces with non-fully coplanar normals to provide at least four equations, making the equation set exactly solvable or overdeterminable. This provides a mathematical basis for achieving autonomous sensing of the panel attitude without adding external attitude sensors.

[0040] Optionally, in some embodiments, determining the solar direction vector, direct irradiance, and diffuse irradiance includes: performing a coarse search on a two-dimensional grid composed of solar azimuth and elevation angles with a preset step size; solving for direct irradiance and diffuse irradiance at each grid point using linear least squares; taking the grid point with the smallest sum of squared residuals as the initial value; performing Levenberg-Marquardt iteration with the initial value until convergence; updating the sky view factor of each illuminated surface with the attitude angle obtained from the iteration and resolving; repeating the iteration until the change in the attitude angle of the surface is less than a preset convergence threshold.

[0041] In this optional embodiment, the solution process is divided into two stages: coarse search and fine iteration. The coarse search stage is based on the solar azimuth angle. With solar altitude angle The observation model iterates through a two-dimensional grid with a step size (e.g., 5°). For each grid point, the solar direction vector is fixed to the direction corresponding to that grid point. and The unknown becomes a linear unknown, which is solved using the linear least squares method, and the sum of squared residuals is calculated. The grid point with the smallest sum of squared residuals is taken as the initial value. In the fine iteration phase, Levenberg-Marquardt (LM) iterations are performed with the initial value. , , , Simultaneously, nonlinear least squares iterations are performed as unknowns until the change in the sum of squared residuals between two adjacent iterations is less than the convergence threshold (e.g., The tilt angle of each illuminated surface relative to the horizontal plane is updated using the attitude angle of the plate obtained through iteration. ,according to After updating the sky view factor, the solution is recalculated, and the iterations are repeated. Repeat (e.g., 3 times) or until the attitude angle change is less than 0.5°.

[0042] In this optional embodiment, besides the combination of grid search and LM iteration, other nonlinear optimization methods can also be used to solve the problem. In one implementation, a genetic algorithm is used to perform a global search for the four unknowns, using the sum of squared residuals as the fitness function. Convergence to the global optimum is achieved through selection, crossover, and mutation operations. This method is independent of the initial value quality and is suitable for multimodal objective functions. In another implementation, a particle swarm optimization algorithm is used, representing the four unknowns with particle positions and using the sum of squared residuals as the fitness function. Convergence is guided by the synergistic effect of individual optima and global optima. All of the above solution methods can converge the system of equations to the correct solution. The specific choice can be flexibly determined based on the controller's computational power, real-time requirements, and convergence accuracy.

[0043] This optional embodiment employs a two-stage solution strategy of coarse search and fine iteration, balancing global convergence and solution efficiency. The coarse search stage traverses the entire solution space with a large step size to avoid getting trapped in local optima; the fine iteration stage uses the LM algorithm to quickly converge to a high-precision solution near the initial value. The outer sky view factor iteration further eliminates... right The nonlinear coupling caused by the dependence on it makes the solution process stable and convergent.

[0044] Optionally, in some embodiments, the temperature correction term is determined based on the open-circuit voltage, including: obtaining the junction temperature of the photovoltaic substring by inverting the difference between the open-circuit voltage and the open-circuit voltage under standard test conditions, and a preset temperature coefficient; the temperature correction term is determined with the junction temperature as input.

[0045] In this optional embodiment, the junction temperature is obtained by inverting the open-circuit voltage, satisfying the following formula: ; in, For the first The measured open-circuit voltage of each photovoltaic substring, in volts (V). For the first The open-circuit voltage of each photovoltaic sub-string under standard test conditions (irradiance 1000 W / m², temperature 25°C), in volts (V), is determined by the module's factory parameters; This is the irradiance correction factor for open-circuit voltage, dimensionless, with a typical value of approximately 0.055; For falling in the first The effective irradiance on a light-receiving surface, measured in watts per square meter (W / m²), is determined by... Calculated; This is the open-circuit voltage temperature coefficient, expressed in degrees Celsius (°C), with a typical value of approximately... / °C (meaning the open-circuit voltage decreases by approximately 0.30% for every 1°C increase in junction temperature); Let be the junction temperature of the photovoltaic substring, in degrees Celsius (°C), which is the unknown to be inverted. The solution can be obtained from this formula. and then with Determine the temperature correction term for the input. ,in This is the temperature coefficient of short-circuit current, with a typical value of approximately / °C.

[0046] In this optional embodiment, the junction temperature inversion process is illustrated using specific numerical values ​​as an example. Let... V, The measured value is 21.3 V. W / m², , / °C. Substitute into the formula: Simplifying, we get ,Right now Solving for °C. Temperature correction term. That is, the short-circuit current increases by about 0.7% due to the increase in junction temperature.

[0047] This optional embodiment utilizes the physical relationship between open-circuit voltage and junction temperature to invert the junction temperature and determine the temperature correction term accordingly. This eliminates the need for additional temperature sensors to achieve temperature correction of the short-circuit current, further reducing the system's dependence on external sensors.

[0048] Step 103: Obtain the theoretical value of the short-circuit current of each photovoltaic substring from the determined solar direction vector, direct irradiance, and diffuse irradiance. Determine the photovoltaic substring whose measured short-circuit current is lower than the theoretical value to a preset level as a shading substring and output the shading degree. After removing the shading substring from the photovoltaic substrings participating in determining the solar direction vector, redetermine the solar direction vector, direct irradiance, diffuse irradiance, and panel attitude angle from the short-circuit current of the remaining photovoltaic substrings.

[0049] In this embodiment, the necessity of shading correction lies in the fact that determining shading requires first determining parameters such as the solar direction vector, direct irradiance, and diffuse irradiance through simultaneous solutions. Only then can the theoretical short-circuit current of each photovoltaic sub-string under unshaded conditions be calculated, and these theoretical values ​​are used as a benchmark to compare with measured values ​​to identify shaded sub-strings. Therefore, this embodiment adopts a residual robust correction process of first determining, then diagnosing, then eliminating, and then resolving. Specifically, shading determination has a logical dependency: determining shading requires knowing the theoretical short-circuit current under unshaded conditions, and the theoretical short-circuit current depends on the environmental parameters (solar direction vector, direct irradiance, and diffuse irradiance) determined through simultaneous solutions. Therefore, it is necessary to first solve for the approximate parameters using all substrings (including obstructed substrings), then calculate the theoretical short-circuit current of each substring under unobstructed conditions, and use this as a benchmark to compare with measured values ​​to identify obstructed substrings. Finally, after removing obstructed substrings, the remaining unobstructed substrings are solved again to obtain the corrected parameters. This process of "approximate solution with obstruction first, then diagnosis and removal, and finally accurate solution without obstruction" solves the circular dependency problem between obstruction determination and environmental parameter estimation.

[0050] In this embodiment, after determining the solar direction vector, direct irradiance, and diffuse irradiance in step 102, these parameters are substituted into the observation model to calculate the theoretical short-circuit current of each photovoltaic substring under unshaded conditions. The measured short-circuit current of each photovoltaic substring is compared with its corresponding theoretical value. When the measured value of a photovoltaic substring is lower than the theoretical value by a preset level, that photovoltaic substring is determined to be a shaded substring. After identifying the shaded substring, it is removed from the set of photovoltaic substrings involved in determining the solar direction vector. The solar direction vector, direct irradiance, diffuse irradiance, and panel attitude angle are then solved again using the short-circuit currents of the remaining unshaded photovoltaic substrings. Since the photovoltaic module has at least four light-receiving surfaces, when the number of shaded substrings does not exceed the difference between the total number of light-receiving surfaces and four, the number of remaining unshaded photovoltaic substrings is still at least four, and the equations are still solvable. The shading degree is used as an independent output to drive the power optimization control of the shaded substrings. Since shading sources such as tree shadows and tent shadows slowly sweep across the board as the sun moves, the identification and removal of shading substrings are performed dynamically, and are re-executed in each sampling period to track changes in shading.

[0051] This embodiment obtains accurate environmental parameter estimates even with occlusion present by removing the occlusion substring from the set of substrings involved in the inversion and then resolving the problem. At the same time, the occlusion degree is used as an independent output for driving power optimization control, turning the occlusion state from an interference that needs to be eliminated into a useful output.

[0052] Optionally, in some embodiments, photovoltaic substrings whose measured short-circuit current is lower than the theoretical value by a preset degree are identified as shading substrings. This includes: identifying photovoltaic substrings whose difference between the theoretical value and the measured value of the short-circuit current is greater than a preset shading threshold as shading substrings; defining the shading degree as the ratio of the difference between the theoretical value and the measured value of the short-circuit current of the shading substring to the theoretical value; and using the shading degree as an independent output to drive the power optimization control of the shading substrings.

[0053] In this optional embodiment, the preset shading threshold is a determination threshold based on the difference between the theoretical and measured short-circuit current. When the difference between the theoretical and measured short-circuit current of a photovoltaic sub-string exceeds this threshold, the photovoltaic sub-string is determined to be a shading sub-string. The shading degree is defined as the difference between the theoretical and measured values ​​divided by the theoretical value, with a value range of 0 to 1; a larger value indicates more severe shading. For example, if the theoretical short-circuit current of a photovoltaic sub-string is 0.905 amperes and the measured value is 0.362 amperes, then the shading degree is... This means that the substring is 60% shaded. The shading degree is an independent output. When the shading degree of a photovoltaic substring is continuously higher than the reconstruction threshold and the duration exceeds the preset duration, the controller controls the switch matrix to bypass the photovoltaic substring from the series branch. After bypassing, the power conversion unit is triggered to perform a global maximum power point scan to restore the power generation efficiency of the system under partial shading conditions.

[0054] This optional embodiment defines the shading degree as the ratio of the difference between the theoretical and measured values ​​to the theoretical value, quantifying the shading level as a scalar between 0 and 1, which facilitates subsequent threshold determination and control decisions. Using the shading degree as an independent output to drive power optimization control transforms the shading state from an interference that needs to be eliminated into a useful output, improving the system's power generation efficiency under partial shading conditions.

[0055] Optionally, in some embodiments, the method further includes: performing a differential calculation on the panel attitude angles obtained from multiple consecutive sampling periods; when the change in the panel attitude angle exceeds a preset shift threshold, determining that the photovoltaic module has been moved, clearing historical shading records, and re-executing the step of determining the solar direction vector.

[0056] In this optional embodiment, since photovoltaic modules may be moved multiple times by users, blown over by the wind, or touched by people in a camping scenario, the panel attitude angle may change abruptly in a short period of time. The panel attitude angles calculated from multiple consecutive sampling periods are differentially analyzed. When the change in the panel attitude angle exceeds a preset displacement threshold (e.g., a tilt angle change exceeding 10° or an azimuth angle change exceeding 15°), it is determined that the photovoltaic module has been moved. At this time, historical shading records are cleared because the shading source position before the move is no longer applicable, and all steps of the solar direction vector inversion are re-executed to obtain the new attitude after the move.

[0057] This optional embodiment moves the differential detection component for the attitude angle of the board surface. After the component is moved, the historical occlusion records are automatically cleared and the inversion is re-executed. This avoids the contamination of the inversion results under the new attitude by the occlusion records before the movement, and ensures the rapid convergence of the system after the component is moved.

[0058] Optionally, in some embodiments, when the number of remaining unshaded photovoltaic sub-strings after correction is less than four, the panel attitude angle successfully calculated in the previous cycle remains unchanged, the total effective irradiance is determined only by the sum of the short-circuit currents of all photovoltaic sub-strings, and the confidence level of the photovoltaic power generation prediction is marked as low.

[0059] In this optional embodiment, when the number of shaded sub-strings is large, resulting in fewer than four remaining unshaded photovoltaic sub-strings, the equation set becomes underdetermined, making it impossible to solve all four unknowns simultaneously. In this case, the panel attitude angle, successfully calculated in the previous cycle, remains unchanged. The total effective irradiance is then inverted from the sum of the short-circuit currents of all photovoltaic sub-strings to roughly estimate the current power generation capacity. Simultaneously, the confidence level of the photovoltaic power generation prediction is marked as low, prompting subsequent energy management strategies to increase the safety margin.

[0060] In this optional embodiment, when the non-occluded substring is insufficient to support a complete inversion, a degradation processing strategy is adopted to keep the attitude angle unchanged and mark the confidence level as low, thereby avoiding divergence in understanding and providing a conservative but usable capacity estimate for energy management.

[0061] Step 104: Determine the weather conditions based on the corrected direct and diffuse irradiance, and determine the weather condition reduction factor.

[0062] In this embodiment, the sharpness index is defined as the ratio of direct irradiance to the sum of direct and diffuse irradiance, that is, the proportion of direct irradiance in the total irradiance. The sharpness index reflects the proportion of direct irradiance in the total irradiance; a higher value indicates a clearer sky, and a lower value indicates a more overcast sky.

[0063] In this embodiment, the physical basis for weather condition classification is as follows: on clear days, the proportion of direct sunlight is high and stable, characterized by a high mean and low standard deviation; on cloudy days, direct sunlight fluctuates drastically due to cloud cover, resulting in a high standard deviation; and on overcast days, direct sunlight approaches zero, resulting in a low mean and low standard deviation. Therefore, a two-dimensional combination of the mean and standard deviation of the sharpness index can distinguish different weather condition types. The introduction of a sliding time window allows the classification to be based on statistical characteristics over a period of time rather than instantaneous values, avoiding misjudgments caused by a single instance of cloud cover. Based on the two-dimensional classification of the sharpness index within the sliding time window, the weather condition is categorized into at least two types, and a decreasing weather condition reduction factor is assigned to each type.

[0064] In this embodiment, the weather reduction factor is a coefficient determined based on the weather condition assessment results and used to reduce the predicted photovoltaic power generation, reflecting the impact of weather factors such as clouds on future power generation. The weather reduction factor represents the discount ratio of weather conditions on photovoltaic power generation; the closer it is to 1, the smaller the impact of weather conditions on power generation, and the closer it is to 0, the greater the impact of weather conditions on power generation. The determination process of the weather reduction factor does not rely on external network connections, making weather condition assessment still usable in off-grid, network-free camping scenarios.

[0065] This embodiment uses a sliding statistic based on the clarity index to determine weather conditions and a weather reduction factor, providing a basis for weather condition correction for production capacity forecasting without relying on external meteorological information, making weather condition determination still usable in offline or no-network camping scenarios.

[0066] Optionally, in some embodiments, the discrimination interval for determining weather conditions and the weather condition reduction factor are set as follows.

[0067] In this optional embodiment, the sharpness index Sliding time window For example, for a 30-minute period, the mean of the sharpness index is calculated within the sliding time window. with standard deviation .when and The weather was determined to be sunny, so a weather condition reduction factor was applied. ;when The weather forecast was cloudy. ;when and The time was judged to be Yin, and the result was taken. ;in .For example , , , , , When the mean and standard deviation do not fall into any discrimination interval, the weather reduction factor determined in the previous period is maintained to avoid frequent switching.

[0068] This optional embodiment uses the mean and standard deviation of the clarity index to determine the weather conditions in two dimensions, and assigns decreasing weather condition reduction factors to each, so that the weather condition determination is based on statistical characteristics rather than instantaneous values, avoiding misjudgments caused by one-time cloud obstruction, while maintaining stability in the weather condition transition range.

[0069] Optionally, in some embodiments, after determining the weather conditions based on the corrected direct irradiance and diffuse irradiance, the method further includes: when the clarity index is lower than a preset ill-conditioned threshold, maintaining the panel attitude angle of the previous cycle, determining the total effective irradiance only by the sum of the short-circuit currents of all photovoltaic sub-strings, and marking the confidence level of the photovoltaic power generation prediction as low.

[0070] In this optional embodiment, when the sharpness index is below a preset ill-conditioned threshold (e.g., 0.25), it indicates that the direct irradiance is close to zero, the sky is dominated by scattering, and the equations are ill-conditioned with respect to the solar direction vector. At this point, the attitude does not need to be re-estimated (the panel has not been moved), and the old value can be maintained to avoid solution divergence. The total effective irradiance is determined solely by the sum of the short-circuit currents of all photovoltaic sub-strings, used to roughly estimate the current power generation capacity, and the confidence level of the photovoltaic power generation prediction is marked as low.

[0071] This optional embodiment employs a degradation strategy when the sharpness index is below the ill-conditioned threshold, keeping the attitude angle unchanged and marking the confidence level as low to avoid divergence in understanding, while providing a conservative but usable capacity estimate for energy management.

[0072] Step 105: Based on the corrected panel attitude angle, predict the incident irradiance of each light-receiving surface in the time domain, and combine it with the weather reduction factor to obtain the photovoltaic power generation prediction curve. Using the photovoltaic power generation prediction curve and the available energy of the energy storage unit as input, determine the power upper limit of each power output branch, and control the power conversion unit to distribute power among the photovoltaic module, energy storage unit and each power output branch according to the power upper limit.

[0073] In this embodiment, the photovoltaic power generation prediction curve is a curve that varies with time, which is composed of the prediction results of the photovoltaic module's output power at each future time within the prediction time domain; the available energy is the energy that the energy storage unit can actually release at the current time, which is reported by the energy storage unit's battery management system.

[0074] In this embodiment, the complete process of capacity prediction is as follows: First, the solar direction vector at each moment in the prediction time domain is calculated according to the astronomical calendar. Then, the solar direction vector is transformed from the horizontal coordinate system to the photovoltaic module body coordinate system using the corrected panel attitude angle. The cosine of the angle between the normal of each light-receiving surface and the solar direction vector is calculated. Then, the contributions of the direct and scattered components of each light-receiving surface are calculated by combining the corrected direct irradiance and diffuse irradiance. The summation is then used to obtain the incident irradiance time domain curve for each light-receiving surface. Finally, the curve is multiplied by the weather reduction factor to obtain the photovoltaic power generation prediction curve. This prediction curve reflects the electrical power that the photovoltaic module can output at each future moment and is the capacity input for energy management.

[0075] In this embodiment, the power upper limit of each power output branch is determined using the photovoltaic power generation prediction curve and the available energy of the energy storage unit as inputs. The power conversion unit then distributes power among the photovoltaic modules, energy storage unit, and each power output branch according to this upper limit. The power output branches include, but are not limited to, at least one of the following: AC socket branch, DC interface branch, USB interface branch, and lighting load drive branch.

[0076] This embodiment constructs a photovoltaic power generation prediction curve by using the corrected panel attitude angle and weather reduction factor, and uses the prediction curve and the available energy of the energy storage unit as input to determine the power upper limit of each power output branch. This realizes a complete closed loop from environmental perception to power generation prediction and then to power allocation, ensuring continuous power supply during camping and reserving the margin required for the return trip.

[0077] Optionally, in some embodiments, determining the power upper limit of each power output branch includes: allocating available energy in time periods based on the constraint that the available energy of the energy storage unit at the end of the power supply task is not lower than the preset reserved energy and the available energy of the energy storage unit throughout the process is not lower than the safety lower limit; when the allocation of a certain scheduling period causes subsequent periods to fail to meet the constraints, backtracking and compressing the power upper limit of the earlier period.

[0078] In this optional embodiment, the user sets the end time of the power supply task through a human-machine interface. Energy retention at the terminal Available energy of the energy storage unit At the present moment to Within the interval, two constraints must be satisfied: the entire process Not lower than the safety lower limit ,as well as time Not less than Using the photovoltaic power generation forecast curve as the capacity input, available energy is allocated segment by segment according to the scheduling period. The execution logic of backtracking compression is as follows: available energy is allocated from the current time to the terminal time segment by segment. After allocating each segment, it is checked whether the remaining energy can still meet the constraints of the subsequent segments. If not, it backtracks to the segment with the highest power limit in the previous segments for compression, and the compression amount is equal to the gap of the current segment. After compression, it checks whether the constraints are met again. If they are still not met, it continues to backtrack to earlier segments until the constraints are met or all previous segments have been compressed to the minimum limit.

[0079] This optional embodiment ensures continuous power supply before the breakup time and reserves the necessary reserve for the return trip through terminal constraints and backtracking compression mechanisms, avoiding the problem of insufficient power supply in the early stage but sudden power outage in the later stage due to the lack of terminal rolling optimization.

[0080] Optionally, in some embodiments, after determining the power upper limit of each power output branch, the method further includes: adopting a continuous flexible load reduction method for the lighting load in the power output branch, quantifying the subjective brightness utility loss corresponding to the unit power reduction based on the logarithmic law of human eye subjective brightness, and performing lighting load reduction in order of utility loss from smallest to largest.

[0081] In this optional embodiment, according to the Weber-Fechner law, the subjective luminance of the human eye is logarithmically related to illuminance. The physical meaning of the Weber-Fechner law is that the human eye's perception of changes in luminance exhibits logarithmic compression; that is, when illuminance increases proportionally, the perceived subjective luminance increases only arithmetically. Therefore, when the lighting power decreases by a certain percentage, the illuminance decreases proportionally, but the perceived subjective luminance only decreases by a small percentage. For example, when the lighting power decreases by 30%, the illuminance decreases by 30%, but the perceived subjective luminance only decreases by about 10%. The subjective luminance of the lighting load is logarithmically related to the illuminance, satisfying the following formula: ; in, The subjective luminance effect of the lighting load is dimensionless; , is a luminance coefficient, dimensionless, obtained from human eye luminance perception experiments; Illuminance produced by the lighting load at the camp reference location, in lux (lx). is a brightness constant, dimensionless, obtained from human eye brightness perception experiments.

[0082] In this optional embodiment, the subjective brightness utility loss corresponding to the unit power reduction of the lighting load is calculated accordingly. Since human eye brightness perception follows a logarithmic law, when the lighting power decreases by a certain percentage, the subjective brightness perception only decreases by a small percentage. Therefore, the unit power utility loss of the lighting load is significantly less than that of electric heating and cooling loads. The utility of electric heating loads has a linear relationship with power, satisfying… ,in For the effect of electric heating loads, For the utility coefficient of electrothermal loads, The power rating for electric heating loads is expressed in watts (W). Let be the utility constant for electrothermal loads, where the unit power utility loss is constant. The efficiency of cooling loads remains constant regardless of power consumption. Within the comfortable temperature range, the efficiency of cooling loads exhibits a gradually increasing linear relationship, while outside this range, it becomes a sharply changing linear relationship. The efficiency loss per unit power is relatively small within the comfortable range but increases dramatically in the uncomfortable range. Load throttling is implemented in ascending order of efficiency loss; that is, lighting loads are prioritized for continuous and flexible throttling by adjusting the PWM duty cycle of the lighting loads. Only when the lighting load is reduced to its minimum and still does not meet the constraints are throttling implemented for heating and cooling loads.

[0083] This optional embodiment quantifies the unit power utility loss of each load based on the logarithmic law of subjective brightness of the human eye and performs load reduction according to the ranking of utility loss, minimizing the loss of user experience with the same amount of energy saving. Since the utility loss of lighting loads increases logarithmically with decreasing power, while the utility loss of electric heating and cooling loads increases linearly with decreasing power, prioritizing the reduction of lighting load power can maximize the preservation of user experience while meeting energy constraints.

[0084] Optionally, in some embodiments, the method further includes: obtaining the camp presence probability by fusing the trigger frequency of the human body pyroelectric infrared sensing unit with the load current characteristics of each power output branch; using the camp presence probability as a load prior input to determine the power upper limit of each power output branch; when the camp presence probability is lower than the departure threshold and the photovoltaic power generation prediction curve is higher than the surplus threshold, controlling the power conversion unit to perform a timed energy consumption task, the timed energy consumption task including at least one of cell balancing and low-temperature preheating of the energy storage unit.

[0085] In this optional embodiment, the triggering frequency of the human body pyroelectric infrared sensing unit reflects the personnel activity status of the camp. The triggering frequency is fused with the load current characteristics of each power output branch to obtain the camp presence probability. The camp presence probability serves as a priori input for the load demand in each scheduling period. When the camp presence probability is low, the load demand prediction for subsequent periods is correspondingly reduced. When the camp presence probability is lower than the departure threshold and the photovoltaic power generation prediction curve is higher than the surplus threshold, it indicates that personnel have left the site and photovoltaic capacity is surplus. At this time, the control power conversion unit performs timed energy consumption tasks, including at least one of energy storage unit cell balancing and low-temperature preheating, using surplus power to complete non-urgent but beneficial maintenance tasks.

[0086] This optional embodiment obtains the camp presence probability by fusing the trigger frequency of the human body pyroelectric infrared sensing unit with the load current characteristics, and uses the camp presence probability as the load prior input for energy management. When personnel leave the site and photovoltaic capacity is surplus, the system automatically executes the timed energy consumption task, thereby improving energy utilization efficiency.

[0087] Optionally, in some embodiments, the method further includes: starting from the corrected plate attitude angle, searching for the target attitude angle that maximizes the cumulative power generation in the prediction time domain within the reachable attitude range, calculating the adjustment amount of the target attitude angle relative to the current attitude angle and the corresponding power generation gain percentage, and outputting the adjustment amount and power generation gain percentage through the human-machine interface.

[0088] In this optional embodiment, starting from the currently corrected panel attitude angle, the target attitude angle that maximizes the cumulative power generation in the prediction time domain is searched within the achievable attitude range of the photovoltaic module. The achievable attitude range is determined by the mechanical structure constraints of the photovoltaic module itself, with a tilt angle range of 0° to 90° (from horizontal placement to vertical orientation towards the sky) and an azimuth angle range of -180° to 180° (the full circumference of rotation around the vertical axis). During the search, the search can be discretized according to the actual locking position of the module's hinge structure, effectively reducing the computational load.

[0089] In this optional embodiment, the search can adopt a two-stage strategy combining coarse scanning and fine searching: first, the entire attitude space is traversed with a coarse step size (e.g., 5°) to quickly locate the local optimum interval with the largest cumulative power generation; then, a fine search is performed within this interval with a fine step size (e.g., 1°) to ensure rapid convergence to the globally optimal target attitude angle under limited computing resources. In calculating the adjustment amount of the target attitude angle relative to the current attitude angle and the corresponding percentage of power generation gain, the adjustment amount can be intuitively output in the form of "tilt adjustment Δβ°, azimuth adjustment Δα°", with positive and negative signs indicating tilt increase / decrease and azimuth rotation clockwise / counterclockwise, respectively; the percentage of power generation gain is calculated using the formula (cumulative power generation in the predicted time domain under the target attitude - cumulative power generation in the predicted time domain under the current attitude) / cumulative power generation in the predicted time domain under the current attitude × 100%. The adjustment amount and gain percentage are only output to the user through the human-machine interface when the gain is greater than a preset threshold (e.g., 1%), avoiding meaningless frequent adjustment prompts that interfere with user operation. After seeing the adjustment amount and gain percentage, users can manually adjust the placement angle of the photovoltaic modules to align the light-receiving surface of the modules in a more favorable direction, thereby improving power generation efficiency without the need for an electric tracking mechanism. The above is merely an exemplary implementation; in practical applications, other search and calculation methods can be used. As long as the core logic of searching for the optimal target attitude angle within the achievable attitude range using the corrected panel attitude angle can achieve the effect of increasing the cumulative power generation in the predicted time domain, it is understandable that it falls within the scope of protection of this application. By searching for the optimal attitude angle and outputting the adjustment amount and gain percentage through the human-machine interface, intuitive attitude adjustment guidance is provided to users, improving power generation efficiency.

[0090] In this optional embodiment, for application scenarios requiring automatic adjustment, in addition to the manual adjustment method described above, an automatic attitude adjustment extension scheme can also be introduced. For example, a lightweight drive module (such as a micro motor) can be added to receive adjustment commands output by the controller, automatically completing the calibration of the photovoltaic module's tilt angle and azimuth angle. This scheme requires no manual intervention throughout the process. After adjustment, the attitude sensing process is automatically re-executed to verify the adjustment effect. If the power generation gain after adjustment does not meet expectations, a secondary fine-tuning is triggered. This extension scheme is only an optional implementation path and does not change the core attitude sensing and angle calculation framework. It can be flexibly selected according to the actual application scenario and is suitable for unattended, long-term deployed off-grid power supply scenarios.

[0091] Optionally, in some embodiments, backtracking and compressing the power limit of earlier time periods includes: sequentially compressing the power limit of the lighting load in ascending order of the subjective brightness utility loss corresponding to the unit power reduction of the lighting load in each earlier time period, wherein the subjective brightness utility loss is quantified according to the logarithmic law of human eye subjective brightness.

[0092] In this optional embodiment, during retrospective compression, the compression is performed sequentially in ascending order of the unit power utility loss of the lighting load in each prior time period, i.e., compression is prioritized for the lighting load. Since the utility loss of the lighting load increases slowly with decreasing power in a logarithmic manner, the impact on user experience is minimal in the initial stages of compression. Compression is then applied to other types of loads only when the lighting load is reduced to its minimum and still does not meet the constraints.

[0093] This optional embodiment combines the logarithmic utility law of lighting load with a backtracking compression mechanism to prioritize reducing lighting loads with the least utility loss during the backtracking compression process, thereby minimizing the overall experience loss.

[0094] See Figure 2 , Figure 2 This is an overall block diagram of an embodiment of the photovoltaic-energy storage integrated system provided in this application. The photovoltaic-energy storage integrated system 200 includes a photovoltaic module 210, an energy storage unit 220, a switch matrix 230, a sampling circuit 240, a power conversion unit 250, a power output branch 260, and a controller 270. The photovoltaic module 210 is connected to the DC bus side of the power conversion unit 250 via the switch matrix 230. The switch matrix 230 is also connected to the sampling circuit 240. The output terminal of the sampling circuit 240 is connected to the analog-to-digital conversion interface of the controller 270. The DC bus side of the power conversion unit 250 is also connected to the energy storage unit 220. The output side of the power conversion unit 250 is connected to the power output branch 260. The controller 270 is connected to the control terminal of the switch matrix 230, the output terminal of the sampling circuit 240, the communication interface of the energy storage unit 220, and the control terminal of the power conversion unit 250.

[0095] In this embodiment, the photovoltaic module 210 is a portable foldable photovoltaic module with at least four light-receiving surfaces. The normals of each light-receiving surface are not fully coplanar, and each light-receiving surface corresponds to an independently extended photovoltaic sub-string. The normal is the direction vector perpendicular to the surface of the light-receiving surface and pointing towards the light-receiving side; the normals are not fully coplanar, meaning that at least two of the normal vectors of each light-receiving surface are not in the same plane. The photovoltaic module 210 includes a base light-receiving surface and at least three side light-receiving surfaces, each side light-receiving surface being connected to the base light-receiving surface via a hinge. The hinge allows the unfolding angle of each side light-receiving surface relative to the base light-receiving surface to be locked at several discrete levels, and in at least one unfolding configuration, the normals of the base light-receiving surface and at least three side light-receiving surfaces are not fully coplanar. Taking a four-fold W-shaped configuration as an example, the base light-receiving surface is placed horizontally, and the three side light-receiving surfaces are folded at different angles along different sides, making the normals of the four light-receiving surfaces not fully coplanar. In this configuration, the normal of the reference light-receiving surface is (0,0,1), and the normals of the three side light-receiving surfaces are respectively ( 0.574,0,0.819), (0.342,0,0.940)), (0, 0.574, 0.819), the four elements are not entirely coplanar, satisfying the solvability condition for simultaneous solution. Each photovoltaic sub-string of the photovoltaic module 210 outputs DC power in power generation mode, which is then fed to the DC bus via the switch matrix 230, providing photovoltaic power generation input to the power conversion unit 250.

[0096] In this embodiment, the input terminal of the switch matrix 230 is connected to each photovoltaic sub-string of the photovoltaic module 210, the first output terminal of the switch matrix 230 is connected to the DC bus side of the power conversion unit 250, and the second output terminal of the switch matrix 230 is connected to the sampling circuit 240. The switch matrix 230 is used to disconnect the sampled photovoltaic sub-string from the power conversion unit 250 and connect it to the sampling circuit 240 during sampling, realizing time-division multiplexing of power generation and sampling. Each photovoltaic sub-string in the switch matrix 230 corresponds to a set of switches, which includes a power generation branch switch connected in series between the photovoltaic sub-string and the DC bus, a short-circuit sampling switch connected in series between the photovoltaic sub-string and the shunt resistor branch of the sampling circuit 240, and a voltage sampling switch connected in series between the photovoltaic sub-string and the high-resistance voltage divider branch of the sampling circuit 240. When the power generation branch switch is closed, the photovoltaic sub-string participates in normal power generation; when it is open, the photovoltaic sub-string is disconnected from the power generation branch. When the short-circuit sampling switch is closed, the photovoltaic sub-string is connected to the shunt resistor branch, and the short-circuit current is measured. When the voltage sampling switch is closed, the photovoltaic sub-string is connected to the high-resistivity voltage divider branch, and the open-circuit voltage is measured. All three switches use at least one of the following: metal-oxide-semiconductor field-effect transistors, insulated-gate bipolar transistors, and relays. The controller 270 controls their on / off states according to a preset timing sequence, with a stable delay (e.g., 1 to 5 milliseconds) between switches to allow the circuit transition process to complete before measurement.

[0097] In this embodiment, the input terminal of the sampling circuit 240 is connected to the second output terminal of the switch matrix 230, and the output terminal of the sampling circuit 240 is connected to the analog-to-digital converter interface of the controller 270. The sampling circuit 240 is used to measure the short-circuit current and open-circuit voltage of the photovoltaic sub-string after the photovoltaic sub-string to be sampled is connected to the switch matrix 230. The sampling circuit 240 includes a shunt resistor branch and a high-resistance voltage divider branch. The shunt resistor branch uses a low-resistance shunt resistor, such as a shunt resistor with a resistance of 1 milliohm, an accuracy of 1%, and a temperature drift of no more than 50 ppm per degree Celsius. When the short-circuit sampling switch is closed, the short-circuit current of the photovoltaic sub-string flows through the shunt resistor. The voltage across the shunt resistor is amplified by the differential amplifier and then sent to the analog-to-digital converter. The analog-to-digital converter has a resolution of no less than 16 bits and a sampling rate of no less than 10,000 times per second to measure the short-circuit current. The high-resistance voltage divider branch uses a high-resistance voltage divider network. When the voltage sampling switch is closed, the open-circuit voltage of the photovoltaic substring is divided and sent to the analog-to-digital converter to measure the open-circuit voltage. The sampling circuit 240 outputs the measured short-circuit current and open-circuit voltage to the controller 270 through the analog-to-digital conversion interface, as the raw electrical quantities for subsequent solar direction vector inversion.

[0098] In this embodiment, the DC bus side of the power conversion unit 250 is connected to the first output terminal of the switch matrix 230 and the energy storage unit 220, respectively, and the output side of the power conversion unit 250 is connected to the power supply output branch 260. The power conversion unit 250 is used to distribute power among the photovoltaic module 210, the energy storage unit 220, and each power supply output branch 260, realizing power scheduling of photovoltaic power generation, energy storage charging and discharging, and load power supply. The power conversion unit 250 includes a DC-DC conversion circuit and a DC-AC inverter circuit. The DC-DC conversion circuit converts the output voltage of the photovoltaic module 210 into a DC bus voltage to charge the energy storage unit 220 and supply power to each DC power supply output branch 260. The DC-AC inverter circuit converts the DC bus voltage into an AC voltage to supply power to the AC power supply output branch 260. The DC-DC conversion circuit also integrates a maximum power point tracking function, dynamically adjusting the scanning strategy according to the shading degree signal output by the controller 270 to maintain the power generation efficiency of the photovoltaic module 210 under partial shading conditions. The power conversion unit 250 controls the output power of each power supply output branch 260 according to the power limit determined by the controller 270.

[0099] In this embodiment, the power output branch 260 is the output channel for the power conversion unit 250 to supply power to external loads, including but not limited to at least one of the following: AC socket branch, DC interface branch, USB interface branch, and lighting load drive branch. The AC socket branch provides AC power to AC loads (e.g., camping lights, fans, small refrigerators). The DC interface branch provides DC power to DC loads (e.g., mobile phone chargers, tablets). The USB interface branch provides low-voltage DC power to USB devices. The lighting load drive branch provides drive power to camping lighting fixtures and supports continuous flexible load reduction by adjusting the PWM duty cycle. The output power of each power output branch 260 is controlled by the controller 270 according to the power upper limit determined by the photovoltaic power generation prediction curve and the available energy of the energy storage unit 220.

[0100] In this embodiment, the energy storage unit 220 is connected to the DC bus side of the power conversion unit 250. It stores excess energy generated by the photovoltaic module 210 and releases it when photovoltaic power generation is insufficient, thus buffering the energy between photovoltaic power generation and load power consumption. The energy storage unit 220 uses at least one of the following: lithium-ion batteries, lithium iron phosphate batteries, and lithium polymer batteries. Taking a lithium iron phosphate battery as an example, its nominal voltage is 51.2 volts, its capacity is 20 amp-hours, and its usable energy is approximately 1024 watt-hours. The energy storage unit 220 has a built-in battery management system that monitors the battery's voltage, current, temperature, and state of charge, calculates the usable energy, and reports it to the controller 270 via a communication interface. It also implements safety strategies such as overcharge protection, over-discharge protection, overcurrent protection, and temperature protection.

[0101] In this embodiment, the controller 270 includes a real-time clock. The controller 270 is connected to the control terminal of the switch matrix 230, the output terminal of the sampling circuit 240, the communication interface of the energy storage unit 220, and the control terminal of the power conversion unit 250, respectively, for performing actions such as... Figure 1 The described steps of the photovoltaic-storage collaborative management method are as follows: Controller 270 is a microcontroller unit or digital signal processor with a floating-point arithmetic unit to support nonlinear iterative calculations and a main frequency of not less than 120 MHz. A real-time clock provides the current date and time with an accuracy of no more than ±5 ppm and has a backup battery to ensure that time is not lost after power failure, providing a time reference for calculating the solar direction vector in astronomical calendars. Controller 270 controls the switching sequence of each switch in switch matrix 230, reads the short-circuit current and open-circuit voltage measured by sampling circuit 240, performs solar direction vector inversion, shading determination and correction, weather condition assessment, power generation prediction and power allocation, and controls the power output of power conversion unit 250. Controller 270 also connects to a human-machine interface to output attitude adjustment and power generation gain percentage for user reference.

[0102] In this embodiment, the integrated photovoltaic and energy storage system 200 is functionally divided into three layers: a sensing layer, a computing layer, and an execution layer. The sensing layer includes a switch matrix 230 and a sampling circuit 240, responsible for time-division sampling of the short-circuit current and open-circuit voltage of each photovoltaic substring. The computing layer includes a controller 270, responsible for all calculation tasks such as vector inversion, shading correction, weather condition determination, and power allocation. The execution layer includes a power conversion unit 250 and a power output branch 260, responsible for executing power allocation according to the power limit determined by the controller 270. The three layers are coupled through electrical signals and data interfaces, forming a complete closed loop from sensing to decision-making to execution.

[0103] In this embodiment, the switching matrix 230 enables each photovoltaic sub-string to be connected to the sampling circuit 240 in a time-division manner. Without adding an external attitude sensor, the solar direction vector inversion and panel attitude angle determination are achieved by using the short-circuit current and open-circuit voltage of the photovoltaic module 210 itself. The inversion results are then directly used for power distribution control by the controller 270, forming a complete closed loop from sensing to control.

[0104] Optionally, in some embodiments, the photovoltaic module 210 includes a reference light-receiving surface and at least three side light-receiving surfaces. Each side light-receiving surface is connected to the reference light-receiving surface through a hinge. The hinge allows the unfolding angle of each side light-receiving surface relative to the reference light-receiving surface to be locked at several discrete levels. In at least one unfolding configuration, the normals of the reference light-receiving surface and the at least three side light-receiving surfaces are not fully coplanar.

[0105] In this optional embodiment, the hinge portion has several discrete locking positions, such as 0° (flat storage), 20°, and 35°. In the unfolded state, the light-receiving surfaces of each side wing are locked at different positions, so that the normals of the reference light-receiving surface and the light-receiving surfaces of each side wing are not fully coplanar. In one implementation, a four-fold W-shaped configuration is used, with the reference plane horizontal and the three side wing surfaces folded at 35°, 20°, and 35° along different edges, respectively. In another implementation, a five-fold configuration is used, with the reference plane horizontal and the four side wing surfaces folded at different angles along different edges, so that the normals of the five light-receiving surfaces are not fully coplanar, forming an overdetermined system of equations. After removing one occlusion substring, four non-occlusion substrings remain, which can be completely resolved, and the anti-occlusion capability is better than that of the four-fold configuration.

[0106] This optional embodiment locks the discrete positions of the hinge, so that the normals of the light-receiving surfaces of the photovoltaic module 210 are not all coplanar in the unfolded state, providing a geometric basis for solar direction vector inversion.

[0107] Optionally, in some embodiments, the photovoltaic module 210 includes three folded light-receiving surfaces and one independent reference light-receiving surface. The three folded light-receiving surfaces are connected to each other through hinges to form a three-fold V-shaped body. The independent reference light-receiving surface is fixed to the handle or frame of the three-fold V-shaped body. The normal of the independent reference light-receiving surface is not fully coplanar with the normal of the three folded light-receiving surfaces. The photovoltaic sub-string corresponding to the independent reference light-receiving surface is connected to the switch matrix 230 to participate in time-division sampling, and the power generation of the independent reference light-receiving surface is less than the power generation of the folded light-receiving surface.

[0108] In this optional embodiment, for the existing three-fold V-shaped photovoltaic module 210 product line, without modifying the main structure, a small independent reference cell is added as a fourth light-receiving surface on the handle or frame of the three-fold V-shaped main body. Its normal is intentionally designed so that it is not coplanar with the normals of the three folded light-receiving surfaces. The independent reference light-receiving surface is also connected to the switch matrix 230 for time-division sampling, but due to its small area, its power generation contribution is negligible. This solution falls within the protection scope of claim 1, which states "not less than four light-receiving surfaces, and the normals of each light-receiving surface are not all coplanar."

[0109] This optional embodiment adds an independent reference light-receiving surface to the three-fold V-shaped main body, enabling the existing three-fold product line to meet the solvability conditions of simultaneous solutions without changing the main body structure, thereby reducing product modification costs.

[0110] Optionally, in some embodiments, each photovoltaic sub-string in the switch matrix 230 corresponds to a set of switches. The set of switches includes a power generation branch switch connected in series between the photovoltaic sub-string and the DC bus, a short-circuit sampling switch connected in series between the photovoltaic sub-string and the shunt resistor branch of the sampling circuit 240, and a voltage sampling switch connected in series between the photovoltaic sub-string and the high-resistance voltage divider branch of the sampling circuit 240.

[0111] In this optional embodiment, three switches corresponding to each photovoltaic substring control the on / off state of the power generation branch, short-circuit sampling branch, and voltage sampling branch, respectively. All three switches employ at least one of the following: metal-oxide-semiconductor field-effect transistors (MOSFETs), insulated-gate bipolar transistors (IGBTs), and relays. Under normal power generation conditions, the power generation branch switch is closed, while the short-circuit sampling switch and voltage sampling switch are open. When sampling short-circuit current, the power generation branch switch is open, the short-circuit sampling switch is closed, and the voltage sampling switch is open. When sampling open-circuit voltage, both the power generation branch switch and the short-circuit sampling switch are open, and the voltage sampling switch is closed. The switching of the three switches is controlled by the controller 270 according to a preset timing sequence, with a stable delay (e.g., 1 to 5 milliseconds) between switching operations.

[0112] This optional embodiment achieves time-division multiplexing of power generation and sampling by designing three switches for each photovoltaic substring, allowing the photovoltaic substring to switch between the two states of power generation and sampling without interference.

[0113] See Figure 3 , Figure 3 This is a schematic diagram of an embodiment of the storage medium provided in this application. The storage medium 300 stores program data 310, which, when executed by a processor, implements the steps of the aforementioned optical-storage collaborative management method.

[0114] In this embodiment, the storage medium 300 is a non-transitory computer-readable storage medium, including but not limited to at least one of read-only memory, random access memory, magnetic disk, optical disk, solid-state drive, or flash memory. The program data 310 includes instructions organized into multiple program modules, which include a sampling control module, a vector inversion module, an occlusion correction module, a weather condition discrimination module, and a power allocation module. The sampling control module controls the switching timing of the switch matrix and reads the measurement results of the sampling circuit. The vector inversion module solves for the solar direction vector, direct irradiance, and diffuse irradiance based on the short-circuit current and the normal of the receiving surface, and determines the panel attitude angle by registering with the astronomical ephemeris. The occlusion correction module identifies occlusion substrings, outputs the occlusion degree, and resolves after removing occlusion substrings. The weather condition discrimination module determines the weather condition and determines the weather condition reduction factor based on the sliding statistics of the sharpness index. The power allocation module determines the power upper limit of each power output branch and controls the power conversion unit based on the photovoltaic power generation prediction curve and the available energy of the energy storage unit.

[0115] In this embodiment, the program data 310 also includes an astronomical ephemeris data module and an attitude history module. The astronomical ephemeris data module stores a solar direction vector lookup table or calculation formula indexed by date and time, which is called by the vector inversion module for plate attitude angle registration. The attitude history module stores plate attitude angles, weather reduction factors, and occlusion records for historical sampling periods, which are called by the occlusion correction module for differential shift detection and by the weather discrimination module for sliding statistics calculation.

[0116] This embodiment stores each step of the optical-storage collaborative management method in the form of program data 310 in the storage medium 300, enabling the processor to implement all steps of the above method when executing the program data 310, thus achieving computerized implementation of the method. The program data 310 can be downloaded via the network or copied to the storage medium 300 via a removable storage medium, facilitating system firmware upgrades and function updates.

[0117] See Figure 4 , Figure 4 This is a schematic diagram of an embodiment of the computer device provided in this application. The computer device 400 includes a processor 410 and a memory 420. The memory 420 stores a computer program, and the processor 410 executes the computer program to implement, for example... Figure 1 The steps of the described photovoltaic-storage collaborative management method.

[0118] In this embodiment, the processor 410 is a microcontroller unit or a digital signal processor, equipped with a floating-point arithmetic unit to support nonlinear iterative calculations. The memory 420 includes a program memory and a data memory. The program memory stores the computer program, and the data memory stores sampling data, astronomical ephemeris data, historical data of the plate attitude angle, and historical data of the weather reduction factor. The computer device 400 also includes a switch drive interface connected to the switch matrix, an analog-to-digital converter interface connected to the sampling circuit, a clock interface connected to the real-time clock, a power control interface connected to the power conversion unit, and a display interface connected to the human-machine interface. When the processor 410 executes the computer program, it controls the switching timing of the switch matrix through the switch drive interface, reads the short-circuit current and open-circuit voltage measured by the sampling circuit through the analog-to-digital converter interface, obtains the current date and time through the clock interface, performs solar direction vector inversion, shading determination and correction, weather condition discrimination, power generation prediction and power allocation, controls the power output of the power conversion unit through the power control interface, and outputs the attitude adjustment amount and power generation gain percentage through the display interface.

[0119] In this embodiment, the computer device operates as follows: After startup, the processor 410 reads the current date and time from the real-time clock, loads astronomical ephemeris data and historical attitude data from the memory 420, and enters a periodic sampling cycle. In each sampling cycle, the processor 410 first sequentially switches the switch matrix corresponding to each photovoltaic sub-string through the switch drive interface, controlling each photovoltaic sub-string to enter the short-circuit sampling state and voltage sampling state in a time-sharing manner, and reads the short-circuit current and open-circuit voltage of each photovoltaic sub-string through the analog-to-digital conversion interface. Subsequently, the processor 410 calls the vector inversion module to solve for the solar direction vector, direct irradiance, and diffuse irradiance based on each short-circuit current and the normal of each light-receiving surface, and registers them with the astronomical ephemeris data to determine the plate attitude angle. Next, the processor 410 calls the shading correction module to compare the measured values ​​of each short-circuit current with the theoretical values, determines the shading sub-strings and outputs the shading degree, removes the shading sub-strings, and re-solves the problem. Subsequently, the processor 410 calls the weather condition discrimination module to determine the weather condition and the weather condition reduction factor based on the sliding statistics of the sharpness index. Finally, the processor 410 calls the power distribution module, predicts the incident irradiance of each light-receiving surface in the time domain based on the corrected plate attitude angle, and obtains the photovoltaic power generation prediction curve by combining the weather reduction factor. Using the photovoltaic power generation prediction curve and the available energy of the energy storage unit as input, the power upper limit of each power output branch is determined, and the power output of the power conversion unit is controlled through the power control interface.

[0120] In this embodiment, the processor 410 executes a computer program to implement all the steps of the above-mentioned photovoltaic-energy storage collaborative management method, enabling the computer device 400 to complete the attitude perception and energy collaborative management of portable photovoltaic modules without adding external attitude sensors. The computer device 400 can be integrated into the control box of the portable photovoltaic-energy storage system, forming an integrated device with the photovoltaic modules, energy storage unit, and power conversion unit, facilitating portability and rapid deployment in camping scenarios.

[0121] Unlike existing technologies, this application discloses a photovoltaic-storage collaborative management method, an integrated photovoltaic-storage system, a storage medium, and equipment. By utilizing the short-circuit current and open-circuit voltage obtained from time-division sampling of photovoltaic substrings corresponding to at least four non-fully coplanar light-receiving surfaces of a folded photovoltaic module, the solar direction vector, direct irradiance, and diffuse irradiance are simultaneously solved. This data is then registered with an astronomical calendar to determine the panel attitude angle. Without adding any external attitude sensors, GPS modules, or network communication modules, autonomous perception of the panel attitude is achieved solely using the electrical quantities of the photovoltaic module itself. This solves the core technical problem of unreliable capacity prediction caused by unknown photovoltaic module attitude in camping scenarios. Furthermore, by removing shading substrings from the set of substrings participating in the inversion and resolving the problem, accurate environmental parameter estimates are still obtained even with dynamic shading. Simultaneously, the shading degree is used as an independent output for driving power optimization control, transforming the shading state from an interference to a useful output. Furthermore, by using a sliding statistic based on the clarity index to determine weather conditions and establish a weather reduction factor, a basis for weather condition correction in production capacity forecasting is provided without relying on external meteorological information. Finally, by allocating available energy in time periods based on the terminal's reserved energy at the end of the power supply task, and by retrospectively compressing the power upper limit of earlier periods when the constraint cannot be met, combined with a flexible load reduction strategy for lighting based on the logarithmic law of subjective human eye brightness, refined energy management is achieved that balances power supply reliability with user experience loss, ensuring continuous power supply during camping and reserving the necessary capacity for the return trip.

[0122] It should be noted that the division of modules and steps in the above embodiments is only a logical functional division, and other division methods may be used in actual implementation. The optional implementation methods in the above embodiments can be combined arbitrarily. Under the premise of no contradiction, those skilled in the art can choose appropriate combination methods according to actual needs, and these combination methods are all within the protection scope of this application. In addition, the numerical parameters involved in the embodiments of this application (including but not limited to the number of light-receiving surfaces, the unfolding angle, the duration of the sampling window, the time interval between adjacent sampling, the stabilization delay, the current response coefficient, the sky view factor, the short-circuit current temperature coefficient, the open-circuit voltage temperature coefficient, the occlusion judgment threshold, the occlusion degree reconstruction threshold, the sharpness index discrimination threshold, the sky condition reduction factor, the power supply task end time, the preset reserved energy, the safety lower limit, the utility model fitting coefficient, etc.) are all exemplary descriptions. In the specific product implementation, they can be determined and adjusted according to standard test conditions, measured data and product hardware characteristics, and do not constitute a specific limitation on the protection scope of this application.

[0123] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments, storage medium embodiments, and computer device embodiments are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0124] The above description is merely an embodiment of this application and does not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A photovoltaic-storage collaborative management method, applied to an integrated photovoltaic-storage system, characterized in that, The integrated photovoltaic and energy storage system includes photovoltaic modules, energy storage units, sampling circuits, power conversion units, and power output branches. Each photovoltaic module has at least four light-receiving surfaces, and the normals of these surfaces are not all coplanar. Each light-receiving surface corresponds to an independent photovoltaic sub-string. The photovoltaic-energy storage collaborative management method includes: The short-circuit current and open-circuit voltage of each photovoltaic substring are measured by time-division sampling. Based on the correspondence between the short-circuit current and the normal and solar direction vector of the light-receiving surface, the solar direction vector, direct irradiance and diffuse irradiance are determined by the short-circuit current of each photovoltaic sub-string and the normal of each light-receiving surface. The panel attitude angle of the photovoltaic module is determined based on the deviation between the solar direction vector and the solar direction vector determined according to the astronomical calendar. The theoretical value of the short-circuit current of each photovoltaic sub-string is obtained from the determined solar direction vector, direct irradiance, and diffuse irradiance. Photovoltaic sub-strings whose measured short-circuit current is lower than the theoretical value by a preset degree are identified as shading sub-strings and the shading degree is output. After the shading sub-strings are removed, the solar direction vector, direct irradiance, diffuse irradiance, and panel attitude angle are re-determined from the remaining photovoltaic sub-strings. The weather conditions are determined based on the corrected direct and diffuse irradiance, and the weather condition reduction factor is determined. Based on the corrected panel attitude angle, the incident irradiance of each of the light-receiving surfaces in the time domain is predicted, and the photovoltaic power generation prediction curve is obtained by combining the weather reduction factor. Using the photovoltaic power generation prediction curve and the available energy of the energy storage unit as input, the power upper limit of each of the power output branches is determined, and the power conversion unit is controlled to distribute power among the photovoltaic module, the energy storage unit and each of the power output branches according to the power upper limit.

2. The photovoltaic-storage collaborative management method according to claim 1, characterized in that, The correspondence between the short-circuit current and the normal and solar direction vector of the light-receiving surface is as follows: the short-circuit current of the photovoltaic substring corresponding to any light-receiving surface is equal to the product of the current response coefficient of the photovoltaic substring and the sum of the following two items, multiplied by a temperature correction term: the first item is the product of the direct irradiance and the positive value of the cosine of the angle between the normal and solar direction vector of the light-receiving surface; the second item is the product of the diffuse irradiance and the sky view factor of the light-receiving surface. The temperature correction term is determined based on the open-circuit voltage; The determination of the solar direction vector, direct irradiance, and diffuse irradiance includes: solving the azimuth and elevation angles of the solar direction vector, the direct irradiance, and the diffuse irradiance by simultaneously solving the corresponding relationships between each of the light-receiving surfaces.

3. The photovoltaic-storage collaborative management method according to claim 1, characterized in that, The step of determining a photovoltaic sub-string whose measured short-circuit current is lower than the theoretical value by a preset level as a shading sub-string includes: A photovoltaic substring whose difference between the theoretical value and the measured value of the short-circuit current is greater than a preset shading threshold is identified as a shading substring, and the shading degree is defined as the ratio of the difference between the theoretical value and the measured value of the short-circuit current of the shading substring to the theoretical value. The occlusion degree is used as an independent output to drive the power optimization control of the occlusion substring.

4. The photovoltaic-storage collaborative management method according to claim 1, characterized in that, The method of determining weather conditions based on corrected direct and diffuse irradiance includes: The sharpness index is defined as the ratio of direct irradiance to the sum of direct irradiance and diffuse irradiance; Based on the mean and standard deviation of the clarity index within the sliding time window, the weather conditions are determined to be of at least two types, and a decreasing weather reduction factor is assigned to each type of weather condition. When the mean and the standard deviation do not fall into either discrimination interval, the weather reduction factor determined in the previous period is maintained.

5. The photovoltaic-storage collaborative management method according to claim 4, characterized in that, After determining the weather conditions based on the corrected direct irradiance and diffuse irradiance, the method further includes: when the clarity index is lower than a preset ill-conditioned threshold, maintaining the panel attitude angle of the previous cycle, determining the total effective irradiance only by the sum of the short-circuit currents of all the photovoltaic sub-strings, and marking the confidence level of the photovoltaic power generation prediction as low.

6. The photovoltaic-storage collaborative management method according to claim 1, characterized in that, Determining the upper limit of the power output of each of the aforementioned power supply output branches includes: Available energy is allocated periodically based on the constraint that the available energy of the energy storage unit at the end of the power supply task is not lower than the preset reserved energy, and the available energy of the energy storage unit throughout the entire process is not lower than the safety lower limit. When the allocation of a certain scheduling period results in subsequent periods failing to meet the constraints, the power limit of the earlier period is backtracked and compressed.

7. The photovoltaic-storage collaborative management method according to claim 1, characterized in that, After determining the power upper limit of each of the power output branches, the method further includes: The lighting load in the power output branch adopts a continuous flexible load reduction method, and the subjective brightness effect loss corresponding to the unit power reduction is quantified according to the logarithmic law of subjective brightness of the human eye. Lighting load reduction is performed in ascending order of utility loss.

8. A photovoltaic-storage integrated system, characterized in that, Includes photovoltaic modules, energy storage units, switch matrices, sampling circuits, power conversion units, power output branches, and controllers; The photovoltaic module has no less than four light-receiving surfaces, the normals of each light-receiving surface are not all coplanar, and each light-receiving surface corresponds to an independent photovoltaic sub-string; The input terminal of the switch matrix is ​​connected to each of the photovoltaic sub-strings, the first output terminal of the switch matrix is ​​connected to the DC bus side of the power conversion unit, and the second output terminal of the switch matrix is ​​connected to the sampling circuit, which is used to disconnect the sampled photovoltaic sub-string from the power conversion unit and connect it to the sampling circuit during sampling. The output of the sampling circuit is connected to the analog-to-digital conversion interface of the controller, and is used to measure the short-circuit current and open-circuit voltage of each photovoltaic sub-string; The DC bus side of the power conversion unit is also connected to the energy storage unit, and the output side of the power conversion unit is connected to the power supply output branch, which is used to distribute power among the photovoltaic module, the energy storage unit and each of the power supply output branches; The controller includes a real-time clock and is connected to the control terminal of the switch matrix, the output terminal of the sampling circuit, the communication interface of the energy storage unit, and the control terminal of the power conversion unit, respectively, for executing the steps of the photovoltaic-energy storage collaborative management method as described in any one of claims 1 to 7.

9. A storage medium storing program data thereon, characterized in that, When the program data is executed by the processor, it implements the steps of the optical-storage collaborative management method as described in any one of claims 1 to 7.

10. A computer device, characterized in that, It includes a processor and a memory, the memory storing a computer program, and the processor executing the computer program to implement the steps of the optical-storage collaborative management method as described in any one of claims 1 to 7.

Images