A solar MPPT power supply control method and system for an intelligent cabinet

Abstract

The present application belongs to the technical field of solar power supply, and particularly relates to a solar MPPT power supply control method and system for an intelligent cabinet, which comprises the following steps: obtaining real-time voltage sequences and real-time current sequences of a solar cell array through a voltage sensor and a current sensor in a control period; calculating a voltage power decoupling index; obtaining a power output instability factor; calculating a global search urgency index; generating a dynamic adaptive trigger boundary; positioning a high-voltage candidate area based on a comparison result; positioning a global maximum power point voltage in the high-voltage candidate area through an optimization algorithm, and driving the working point to rapidly jump; after the working point rapidly jumps, the system is switched to a steady-state tracking method to perform steady-state tracking and maintain closed-loop monitoring. The present application effectively solves the technical problem that the existing fixed threshold method confuses uniform changes and local shading of the light environment, and causes low sensitivity of the system to local shading events.

Description

Technical Field

[0001] This invention relates to the field of solar power technology. More specifically, this invention relates to a solar MPPT power supply control method and system for intelligent cabinets. Background Technology

[0002] In the intelligent upgrading of urban infrastructure, the deployment of outdoor smart cabinets equipped with solar photovoltaic power generation systems is a key link in achieving energy self-sufficiency. The core objective of maximum power point tracking (MPPT) is not only to track the current power output in real time, but also to stably lock onto the globally unique maximum power point in a complex and ever-changing environment. However, in the complex shading environment of the city, photovoltaic arrays face the coexistence and competition between the normal state of uniform global illumination and the abnormal state of local shading caused by trees and buildings. Any slight fluctuation in ambient light and shadow that causes power changes may cause the photovoltaic system to fall into the multi-peak trap caused by local shading, resulting in the system incorrectly locking onto a local suboptimal peak, causing a decrease in power generation efficiency and energy loss.

[0003] To achieve effective control of the photovoltaic power generation process, existing technologies typically employ a fixed threshold method. This method involves real-time monitoring of power changes or voltage change rates, and using a preset fixed threshold as a trigger line to determine whether to initiate a scan. This significantly improves the efficiency of production line engineers in debugging and deploying MPPT control logic and enables standardized configuration of control parameters.

[0004] However, in the early stages or during dynamic changes of local occlusion, the power drop characteristics caused by local occlusion are extremely similar to the global illumination reduction. Its signal characteristics are often masked by the dominant illumination fluctuations. Existing fixed threshold methods usually rely on single-dimensional amplitude judgment. When dealing with illumination change scenarios with completely different physical causes, this model will seriously confuse uniform changes with local occlusion. This results in extremely low sensitivity for recognizing the key event of local occlusion. It is difficult to effectively identify and initiate global optimization before local occlusion causes the system to fall into a local optimum. It cannot meet the ultra-high sensitivity control requirements of accurately distinguishing illumination attributes and achieving zero misjudgment in complex urban light and shadow environments. Summary of the Invention

[0005] To address the technical problem that existing fixed threshold methods confuse uniform changes in the lighting environment with local occlusion, resulting in low sensitivity of the system to recognize local occlusion events, the present invention provides solutions in the following aspects.

[0006] In a first aspect, the present invention provides a solar MPPT power supply control method for a smart cabinet, comprising:

[0007] During the control cycle, real-time voltage and current sequences of the solar cell array are acquired using voltage and current sensors. A voltage-power decoupling index is calculated based on the difference between the dispersion of power fluctuations and the dispersion of voltage fluctuations. A power output instability factor is obtained based on the amplification relationship of the voltage-power decoupling index to the fluctuations in the illumination environment. A global search urgency index is calculated based on the impact of environmental instability and abnormal power loss in the real-time power sequence on global optimization decisions. A dynamic adaptive trigger boundary is generated based on the statistical characteristics of the global search urgency index over a preset time period and the interaction between the index and environmental stability. The global search urgency index is compared with the dynamic adaptive trigger boundary, and a high-voltage candidate region is located based on the comparison results. Within the high-voltage candidate region, the global maximum power point voltage is located using an optimization algorithm, and the operating point is driven to rapidly transition. After the rapid transition of the operating point is completed, the system switches to a steady-state tracking method for steady-state tracking and maintains closed-loop monitoring.

[0008] This invention effectively solves the problems of existing control methods, such as the inability to accurately distinguish between different environmental changes, low sensitivity to abnormal situations, susceptibility to misjudgment and missed detection, and low search efficiency. This invention achieves differentiation of fluctuations caused by different reasons by accurately collecting equipment operating data, avoiding confusion; it meets the operational needs of different environments by dynamically adjusting judgment criteria, reducing the probability of misjudgment and missed detection; it reduces blind operation and improves optimization efficiency by locating high-voltage candidate areas and conducting precise searches; and it ensures the stability of efficient operation by rapidly adjusting equipment status and continuously tracking it. The entire process forms a complete closed loop from data acquisition, analysis and judgment to optimization execution and continuous monitoring, enabling the system to not only keenly perceive abnormal situations in complex and ever-changing outdoor environments but also efficiently find the optimal operating state, avoiding energy waste and improving the reliability and efficiency of equipment operation.

[0009] Preferably, during the control cycle, the real-time voltage sequence and real-time current sequence of the solar cell array are acquired using voltage and current sensors, including:

[0010] The system uses high-precision voltage and current sensors installed at the input of the DC converter to collect the output voltage and output current values ​​of the solar cell array in real time during each control cycle at a fixed sampling frequency, thereby obtaining real-time voltage and real-time current sequences.

[0011] Preferably, the voltage-power decoupling index is calculated, including:

[0012] The real-time power sequence is obtained by multiplying the real-time voltage sequence and the real-time current sequence.

[0013] ;

[0014] In the formula, Indicates the voltage-power decoupling index; and These represent the standard deviations of the real-time power sequence and the real-time voltage sequence, respectively. and These represent the absolute values ​​of the average values ​​of the real-time power sequence and the real-time voltage sequence, respectively. and For a very small positive number, the denominator must not be 0; This represents the absolute value function.

[0015] This invention, by calculating the voltage-power decoupling index, can distinguish equipment operation fluctuations caused by different environmental changes, rather than just judging based on surface phenomena. Whether it is a natural change in the overall environment or a special situation such as local shading, the voltage-power decoupling index can be used to identify fluctuations caused by different reasons, thus avoiding confusion.

[0016] Preferably, the power output instability factor satisfies the following expression:

[0017] ;

[0018] In the formula, Indicates the power output instability factor; Indicates the voltage-power decoupling index; This represents the standard deviation of the real-time power sequence; Represents the absolute value of the average value of the real-time power sequence; It is a very small positive number, and the denominator is guaranteed to be non-zero.

[0019] This invention uses a voltage-power decoupling index to weight and amplify environmental fluctuations, making key abnormal signals more prominent and effectively preventing them from being masked by normal fluctuations. For special situations that may affect equipment operating efficiency, this method can improve the sensitivity of system identification and provide timely warnings as early as possible when problems occur.

[0020] Preferably, the global search urgency index satisfies the following expression:

[0021] ;

[0022] In the formula, This indicates the global search urgency index; and The power values ​​at adjacent sampling times in the real-time power sequence; This is a power output instability factor; and For a very small positive number, the denominator must not be 0; This represents the maximum value function.

[0023] This invention comprehensively considers environmental stability and energy loss during equipment operation, making a more reasonable judgment on whether to initiate a global search. When the environment is stable, it can also keenly detect even small energy losses and take timely measures. When the environment fluctuates greatly, it can filter out irrelevant interference and avoid frequent search initiation leading to resource waste.

[0024] Preferably, the dynamic adaptive trigger boundary satisfies the following expression:

[0025] ;

[0026] In the formula, Indicates a dynamically adaptive trigger boundary; and These represent the average and standard deviation of the global search urgency index over the past M periods, respectively. This is a power output instability factor; This represents the minimum value function, with a value range of [0, 0.8].

[0027] This invention dynamically adjusts the judgment criteria based on the system's recent operating status, which can meet the environmental changes at different times and avoid the adaptability problems caused by fixed standards. When the environment is stable, the dynamic adaptive trigger boundary is more lenient and can capture subtle anomalies in a timely manner; when the environment fluctuates drastically, the dynamic adaptive trigger boundary is more stringent and reduces misjudgments.

[0028] Preferably, locating high-voltage candidate regions based on comparison results includes:

[0029] The global search urgency index is compared with the dynamic adaptive trigger boundary. If the global search urgency index is less than or equal to the dynamic adaptive trigger boundary, it is determined that the current situation is a normal environmental fluctuation and there is no need to start global optimization. If the global search urgency index is greater than the dynamic adaptive trigger boundary, it is determined that a high-probability local shading event has occurred, triggering a global optimization command. The system pauses the current MPPT algorithm and controls the DC converter to make the operating point of the solar cell array quickly jump to multiple preset, representative voltage points. It briefly stops at each voltage point and measures the power. By comparing the power values ​​of these voltage points, the voltage range where the point with the highest power is located is determined and defined as a high-voltage candidate region.

[0030] Preferably, the global maximum power point voltage is located within the high-voltage candidate region using an optimization algorithm, and the operating point is driven to transition rapidly, including:

[0031] Within the high-voltage candidate region, a set of voltage values ​​is initialized. By iteratively updating the speed and position of each voltage value, after a finite number of iterations, the voltage value corresponding to the population optimal solution is identified as the global maximum power point voltage. Based on the global maximum power point voltage, the central control unit immediately calculates the corresponding DC converter duty cycle and sends a PWM control signal to drive the DC converter to adjust the operating point of the solar cell array to the global maximum power point voltage in the shortest possible time.

[0032] Preferably, after completing the rapid transition of the operating point, the system switches to a steady-state tracking method for steady-state tracking and to maintain closed-loop monitoring, including:

[0033] After completing the rapid transition of the operating point, the control algorithm immediately uses the global maximum power point voltage as the starting point to perform small-step perturbation and observation methods to ensure that the system operates stably at the power peak. During this period, the system returns to the execution monitoring and judgment process in parallel to continuously monitor the next possible occlusion event, forming a complete closed-loop control.

[0034] Secondly, the present invention provides a solar MPPT power supply control system for a smart cabinet, comprising a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the aforementioned solar MPPT power supply control method for a smart cabinet is implemented.

[0035] By adopting the above technical solution, a computer program is generated from the above-mentioned solar MPPT power supply control method for smart cabinets and stored in the memory so that it can be loaded and executed by the processor. In this way, a terminal device can be made based on the memory and the processor for convenient use.

[0036] The beneficial effects of this invention are as follows: Through scientifically sound control logic, the solar power supply system can better adapt to complex and ever-changing outdoor environments, maintaining optimal operation regardless of whether the environment is stable or experiencing severe fluctuations. This solution improves energy utilization efficiency, reduces unnecessary energy loss, and provides a reliable energy guarantee for the long-term stable operation of the smart cabinet. Simultaneously, the entire control process requires no manual intervention, achieving fully automated control, reducing operation and maintenance costs and operational complexity, making the application of solar power systems in outdoor smart cabinets more practical and valuable for widespread adoption. Furthermore, the stable operation of this solution reduces reliance on traditional energy sources, aligning with the development trend of energy conservation and emission reduction. Attached Figure Description

[0037] Figure 1 This is a flowchart illustrating a solar MPPT power supply control method for a smart cabinet according to the present invention. Detailed Implementation

[0038] This invention discloses a solar MPPT power supply control method for intelligent cabinets, referring to... Figure 1 This includes steps S1-S4:

[0039] S1: During the control cycle, the real-time voltage sequence and real-time current sequence of the solar cell array are obtained through voltage and current sensors.

[0040] It should be noted that in real urban outdoor scenarios, the lighting environment faced by photovoltaic arrays is not static. The dappled shadows cast by swaying leaves and the momentary shadows cast by pedestrians passing by can drastically alter the output characteristics of the solar panels on a millisecond timescale. Low-frequency sampling can only obtain a smoothed average value, masking the true dynamic details of PV curve distortion, making the system unable to detect the instantaneous occurrence of shading. Therefore, this invention establishes a high-frequency data channel to capture those brief voltage drops and current surges, providing data support for subsequently distinguishing between gradual changes in natural lighting and sudden shading.

[0041] Specifically, during the control cycle, the real-time voltage and current sequences of the solar cell array are acquired using voltage and current sensors, including:

[0042] The system uses high-precision voltage and current sensors installed at the input of the DC converter to collect the output voltage and output current values ​​of the solar cell array in real time during each control cycle at a sampling frequency of not less than 500Hz, thereby obtaining real-time voltage and real-time current sequences.

[0043] Thus, the real-time voltage and current sequences of the solar cell array were obtained.

[0044] S2: Calculate the voltage-power decoupling index based on the difference between the dispersion of power fluctuations and the dispersion of voltage fluctuations; obtain the power output instability factor based on the amplification relationship of the voltage-power decoupling index to the fluctuation of the illumination environment.

[0045] It should be noted that in the actual operation of photovoltaic arrays, power fluctuations are only a superficial feature; the underlying physical causes are key. Cloud cover reducing global illumination causes a synchronous decrease in the output of all battery cells in the array, resulting in a highly linear correlation between voltage and power. Local shading reverses the bias of affected battery cells, triggering bypass diodes and causing a dramatic decoupling and separation of voltage and power response trajectories. Therefore, this invention introduces a voltage-power decoupling index, aiming to capture the unique physical fingerprint of bypass diode operation by utilizing the difference in response pace between the surface power fluctuations and the two, thereby mathematically distinguishing between local shading and global illumination changes.

[0046] Specifically, based on the difference between the dispersion of power fluctuations and the dispersion of voltage fluctuations, the voltage-power decoupling index is calculated, including:

[0047] The real-time power sequence is obtained by multiplying the real-time voltage sequence and the real-time current sequence.

[0048] The voltage-power decoupling index satisfies the following expression:

[0049] ;

[0050] In the formula, Indicates the voltage-power decoupling index; and These represent the standard deviations of the real-time power sequence and the real-time voltage sequence, respectively. and These represent the absolute values ​​of the average values ​​of the real-time power sequence and the real-time voltage sequence, respectively. and It is a very small positive number, and the denominator is guaranteed to be non-zero.

[0051] In the formula, The coefficient of variation represents the rate of change of power in a real-time power sequence and is used to measure the relative dispersion of real-time power fluctuations. The coefficient of variation represents the rate of change of voltage in a real-time voltage sequence and is used to measure the relative dispersion of voltage fluctuations in real time. It is used to calculate the degree of difference between real-time power fluctuation characteristics and real-time voltage fluctuation characteristics. When the illumination changes uniformly, the real-time power and real-time voltage change synchronously, and their coefficients of variation tend to be consistent, with the difference approaching 0. When local shading occurs, the bypass diodes operate, causing a voltage step while the power change is relatively small or asynchronous. The difference in their coefficients of variation increases significantly, thus not only reflecting the existence of fluctuations but also revealing the physical cause of the fluctuations more accurately, i.e., whether it is an asymmetric response caused by shading.

[0052] For example, in case one, the illumination is uniform but gradually changes, and the power variation coefficient is... The voltage variation coefficient is Calculations yielded Scenario 2 exists, with partial tree shadow obstruction, and the power variation coefficient is... The voltage variation coefficient suddenly increased to Calculations yielded The results show that the voltage-power decoupling index increases significantly, which can accurately identify asymmetric shading events.

[0053] It should be noted that in the complex outdoor environment where photovoltaic systems operate continuously, simply measuring power fluctuations is insufficient to identify their underlying physical causes. When the photovoltaic panels are covered by tree shadows, the output power drops rapidly. This, along with the reduction in uniform sunlight caused by passing clouds, both manifest as power fluctuations. However, their natures are fundamentally different. The former implies a multi-peak structure in the PV curve, requiring a global search; the latter is merely an overall decrease in the curve amplitude, which can be further tracked locally. Therefore, this invention introduces a power output instability factor. Its core lies in using the voltage-power decoupling index—a physical fingerprint—to risk-weight the surface power fluctuations. This amplifies fluctuations caused by local shading and inconsistent voltage and power responses, allowing for the precise extraction of truly dangerous signals indicating multi-peak traps and requiring immediate system intervention from a large volume of real-time data streams.

[0054] Preferably, based on the amplification relationship between the voltage-power decoupling index and fluctuations in the illumination environment, a power output instability factor is obtained, including:

[0055] The power output instability factor satisfies the following expression:

[0056] ;

[0057] In the formula, Indicates the power output instability factor; Indicates the voltage-power decoupling index; This represents the standard deviation of the real-time power sequence; Represents the absolute value of the average value of the real-time power sequence; It is a very small positive number, and the denominator is guaranteed to be non-zero.

[0058] In the formula, The coefficient of variation, representing the real-time power sequence, is a statistical indicator that measures the fundamental volatility of the lighting environment. As a weighting coefficient, the voltage-power decoupling index is used to nonlinearly amplify the fundamental volatility; when When the light intensity is low and the illumination is uniform, the weighting coefficients approach 1. It mainly reflects simple fluctuations in light intensity; when When the value is large, local occlusion occurs, and the weighting coefficient is greater than 1, forcibly increasing the value. The final value; It achieves multiplicative coupling between the amplitude and the nature of the fluctuation, making the power output instability factor highly sensitive to instability caused by local occlusion; the power output instability factor takes values ​​in the range of [0,5], and when the calculation result exceeds 5, 5 is taken as the upper limit.

[0059] For example, in scenario one, the overall light intensity fluctuates weakly on cloudy days. , Calculations yielded Scenario two exists: partial occlusion causes severe fluctuations. , Calculations yielded As can be seen from the comparison, although the basic volatility only increased... However, due to the influence of the voltage-power decoupling index, the final power output instability factor is amplified. This effectively highlights the risk of occlusion. The above calculation results... Round to two decimal places.

[0060] It should be noted that in dynamically changing natural environments, the amount of information contained in power drops of the same magnitude is completely different. In moments of extremely stable illumination, even a small power loss highly likely indicates the appearance of new obstructions and should be considered a high-risk event. However, in cloudy conditions with swaying tree shadows, a large power jump may simply be part of the environmental background noise and not worthy of initiating a global search. Therefore, this invention constructs a global search urgency index, weighting the absolute magnitude of the power drop within the context of current environmental stability. The aim is to endow the system with an environmental awareness capability, enabling it to keenly detect abnormal disturbances in a stable environment.

[0061] S3: Calculate the global search urgency index based on the impact of environmental instability and power anomaly loss in the real-time power sequence on the global optimization decision; generate a dynamic adaptive trigger boundary based on the interaction between the statistical characteristics of the global search urgency index and environmental stability during a preset time period; compare the global search urgency index with the dynamic adaptive trigger boundary, and locate high-voltage candidate regions based on the comparison results.

[0062] It should be noted that simply setting a fixed threshold when using the global search urgency index as the basis for triggering decisions often leads to either oversensitivity or insensitivity. The environment in which photovoltaic arrays operate is unstable, ranging from clear skies to complex conditions with dense clouds, and the system's background fluctuation level is constantly changing. An urgency value sufficient to trigger a search in a stable environment may be merely a normal value in a drastically fluctuating environment. Therefore, this invention introduces a global search urgency index, which no longer relies on a fixed value set empirically, but dynamically generates a statistical boundary that reflects the upper limit of normal fluctuations in the current environment by learning the system's recent historical performance online. This allows the system to intelligently distinguish between normal environmental noise and truly abnormal events that require intervention, thereby avoiding frequent false triggers.

[0063] Specifically, based on the impact of environmental instability and power anomaly losses in the real-time power sequence on global optimization decisions, a global search urgency index is calculated, including:

[0064] The global search urgency index satisfies the following expression:

[0065] ;

[0066] In the formula, This indicates the global search urgency index; and The power values ​​at adjacent sampling times in the real-time power sequence; This is a power output instability factor; and For a very small positive number, the denominator must not be 0; This represents the maximum value function.

[0067] In the formula, It represents the relative rate of decrease of power values ​​in a real-time power sequence, and the negative sign is used to convert power drops into positive values ​​as a global search urgency index. Used to filter out power increases and focus on power loss events; It is the reciprocal of the power output instability factor, and in this invention it is used as a weighting coefficient for environmental stability. The larger the value of this item, the more stable the environment. This means that in a more stable environment, the smaller the denominator, the greater the weight, and the higher the global search urgency index caused by the same power drop; conversely, in a volatile environment, the weight is greatly reduced, thereby suppressing misjudgments caused by environmental noise.

[0068] For example, the power drop is 10, and there is a scenario where the environment is extremely stable. , The value is 100, and the calculation yields... Scenario two exists: drastic environmental fluctuations. , The value is 5, and the calculation yields... The results show that, under stable conditions, the same power drop triggers a global search urgency index that is 20 times greater than under turbulent conditions.

[0069] It should be noted that the present invention designs a dynamic adaptive trigger boundary. By learning the recent historical state of the online learning system, a dynamic baseline that closely follows environmental changes is constructed in real time. This ensures that the system always uses the statistical characteristics of the current environment as a benchmark to determine whether the current urgency is an abnormal change, thereby achieving adaptive decision-making in all weather and all working conditions.

[0070] Preferably, a dynamic adaptive trigger boundary is generated based on the interaction between the statistical characteristics of the global search urgency index over a preset time period and environmental stability, including:

[0071] The dynamic adaptive trigger boundary satisfies the following expression:

[0072] ;

[0073] In the formula, Indicates a dynamically adaptive trigger boundary; and These represent the average and standard deviation of the global search urgency index over the past M periods, respectively. This is a power output instability factor; This represents the minimum value function, with a value range of [0, 0.8].

[0074] In the formula, This represents the average level of the system's recent global search urgency index, reflecting the background noise baseline for the current triggering decision; This represents the fluctuation range of the system's recent global search urgency index, reflecting the range of uncertainty in the past environment; It is a dynamic weighting coefficient based on environmental stability, through dynamic instability factors. Real-time status of the associated environment The settings are based on a large amount of experimental data, when At this point, environmental fluctuations are already in an extreme state, and further increasing the weight would lead to an excessively high trigger threshold; therefore, the upper limit is set at [value missing]. ; This indicates that the fluctuation tolerance range is dynamically adjusted through environmental stability; that is, the more stable the environment, the greater the tolerance. The smaller the value, the closer the weight is to 1, and the more sensitive it is to minor anomalies; the more volatile the environment... The larger the value, the greater the weight, the higher the boundary, and the higher the tolerance to environmental noise, thus achieving adaptive judgment of stable environment sensitivity and fluctuating environment robustness; the preset time period is the past W control cycles, where W is a preset positive integer, and it is recommended to take a value of 5-10 control cycles.

[0075] For example, , Scenario 1 exists: the environment is extremely stable. , =1.02, calculated as follows Scenario two exists: drastic environmental fluctuations. , 1.13, calculated as follows The results show that the dynamic adaptive triggering boundary is higher in fluctuating environments than in stable environments. The above results show improved noise immunity without sacrificing anomaly detection sensitivity in stable environments. Round to two decimal places.

[0076] It should be noted that after partial occlusion occurs, Characteristic curves exhibit complex multi-peak shapes. Blindly performing detailed scans across the entire voltage range is not only time-consuming but also causes significant energy interruptions. However, the global maximum power point often appears near a specific percentage of the open-circuit voltage. Therefore, this invention employs a segmented fast scanning strategy. By probing several key voltage nodes, it quickly eliminates obviously inefficient low-potential regions, thereby precisely focusing valuable computing resources and search time on the most promising areas.

[0077] Preferably, the global search urgency index is compared with the dynamic adaptive trigger boundary, and the high-voltage candidate region is located based on the comparison result, including:

[0078] The global search urgency index is compared with the dynamic adaptive trigger boundary. If the global search urgency index is less than or equal to the dynamic adaptive trigger boundary, the current situation is considered a normal environmental fluctuation, and global optimization is not required. If the global search urgency index is greater than the dynamic adaptive trigger boundary, a high-probability local shading event is considered to have occurred, triggering a global optimization command. The system pauses the current MPPT algorithm and controls the DC converter to quickly jump the operating point of the solar array to multiple preset, representative voltage points. The system briefly pauses at each voltage point and measures the power. By comparing the power values ​​at these voltage points, the voltage range containing the highest power point is determined and defined as a high-voltage candidate region. It should be noted that each voltage point is paused for 3-5 sampling cycles. If the sampling frequency is 500Hz, the dwell time is 6-10ms to ensure stable power measurement.

[0079] S4: Within the high-voltage candidate region, the global maximum power point voltage is located using an optimization algorithm, and the operating point is driven to transition rapidly. After the rapid transition of the operating point is completed, the system switches to a steady-state tracking method to perform steady-state tracking and maintain closed-loop monitoring.

[0080] It should be noted that although coarse localization locks in the approximate range, local extrema may still exist within this region. Existing hill-climbing methods are prone to getting stuck on local hills and missing the true highest peak. While particle swarm optimization (PSO) has the ability to overcome local extrema, its computational cost is too high if it searches the entire region. Therefore, this invention executes PSO within the identified high-voltage candidate region, leveraging the collaborative advantages of swarm intelligence to perform high-density iterative optimization within a very small search space.

[0081] It should be noted that the global maximum power point voltage calculated at the algorithm level is merely a numerical target. Energy gain can only be achieved by instantly converting the global maximum power point voltage into the physical action of power electronic devices. Any delay from the search state to the steady state means a loss of power generation. Therefore, the operation of the execution command system's rapid transition of the operating point aims to eliminate the time lag between control commands and hardware response, driving the DC converter to cross intermediate states in the fastest way possible, directly transitioning the photovoltaic array to the newly locked optimal operating point, ensuring that the optimization results take effect immediately.

[0082] Specifically, within the high-voltage candidate region, an optimization algorithm is used to locate the global maximum power point voltage and drive a rapid transition of the operating point, including:

[0083] Within the high-voltage candidate region, a set of voltage values ​​is initialized. By iteratively updating the speed and position of each voltage value, after a finite number of iterations, the voltage value corresponding to the population optimal solution is identified as the global maximum power point voltage. Based on the global maximum power point voltage, the central control unit immediately calculates the corresponding DC converter duty cycle and sends a PWM control signal to drive the DC converter to adjust the operating point of the solar cell array to the global maximum power point voltage in the shortest possible time.

[0084] It should be noted that although the intelligent algorithm of this invention has successfully found the global maximum power point voltage, its random search mechanism near the peak can cause unnecessary small fluctuations in the output power, making it difficult to maintain a perfect steady state. While the existing perturbation-observation method has a limited field of view, it exhibits extremely high efficiency and stability when tracking slow changes in natural illumination. Therefore, this invention utilizes the low-jitter and high-response characteristics of the perturbation-observation method to closely track the locked global maximum power point voltage, in order to cope with the small but continuous natural drift of illumination or temperature over time.

[0085] Preferably, after completing the rapid transition of the operating point, the system switches to a steady-state tracking method for steady-state tracking and to maintain closed-loop monitoring, including:

[0086] After completing the rapid transition of the operating point, the control algorithm immediately uses the global maximum power point voltage as the starting point to perform small-step perturbation and observation methods to ensure that the system operates stably at the power peak. During this period, the system returns to the execution monitoring and judgment process in parallel to continuously monitor the next possible occlusion event, forming a complete closed-loop control.

[0087] This invention also discloses a solar MPPT power supply control system for a smart cabinet, including a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement a solar MPPT power supply control method for a smart cabinet according to the present invention.

[0088] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.

[0089] While this specification has shown and described numerous embodiments of the invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in the practice of this invention.

Claims

1. A solar MPPT power supply control method for intelligent cabinets, characterized in that, include: During the control cycle, the real-time voltage and current sequences of the solar cell array are acquired using voltage and current sensors. Based on the difference between the dispersion of power fluctuations and the dispersion of voltage fluctuations, the voltage-power decoupling index is calculated; based on the amplification relationship of the voltage-power decoupling index with the fluctuation of the illumination environment, the power output instability factor is obtained. , In the formula, Indicates the voltage-power decoupling index; This represents the standard deviation of the real-time power sequence; Represents the absolute value of the average value of the real-time power sequence; For a very small positive number, the denominator must not be 0; Based on the impact of environmental instability and abnormal power loss in the real-time power sequence on global optimization decision-making, a global search urgency index is calculated. , In the formula, and The power values ​​at adjacent sampling times in the real-time power sequence; and For a very small positive number, the denominator must not be 0; Represents the maximum value function; Based on the interaction between the statistical characteristics of the global search urgency index and environmental stability over a preset time period, a dynamic adaptive trigger boundary is generated. , In the formula, and These represent the average and standard deviation of the global search urgency index over the past M periods, respectively. This represents the minimum value function, with a range of values ​​in the range [0, 0.8]. The global search urgency index is compared with the dynamic adaptive trigger boundary, and the high-voltage candidate region is located based on the comparison results; Within the high-voltage candidate region, the global maximum power point voltage is located using an optimization algorithm, and the operating point is driven to transition rapidly. After the rapid transition of the operating point is completed, the system switches to a steady-state tracking method to perform steady-state tracking and maintain closed-loop monitoring.

2. The solar MPPT power supply control method for a smart cabinet according to claim 1, characterized in that, The process of acquiring the real-time voltage and current sequences of the solar cell array using voltage and current sensors during the control cycle includes: The system uses high-precision voltage and current sensors installed at the input of the DC converter to collect the output voltage and output current values ​​of the solar cell array in real time during each control cycle at a fixed sampling frequency, thereby obtaining real-time voltage and real-time current sequences.

3. The solar MPPT power supply control method for a smart cabinet according to claim 1, characterized in that, The calculation of the voltage-power decoupling index includes: The real-time power sequence is obtained by multiplying the real-time voltage sequence and the real-time current sequence. ； In the formula, Indicates the voltage-power decoupling index; and These represent the standard deviations of the real-time power sequence and the real-time voltage sequence, respectively. and These represent the absolute values ​​of the average values ​​of the real-time power sequence and the real-time voltage sequence, respectively. and For a very small positive number, the denominator must not be 0; This represents the absolute value function.

4. The solar MPPT power supply control method for a smart cabinet according to claim 1, characterized in that, The process of locating high-voltage candidate regions based on comparison results includes: The global search urgency index is compared with the dynamic adaptive trigger boundary. If the global search urgency index is less than or equal to the dynamic adaptive trigger boundary, it is determined that the current situation is a normal environmental fluctuation and there is no need to start global optimization. If the global search urgency index is greater than the dynamic adaptive trigger boundary, it is determined that a high-probability local shading event has occurred, triggering a global optimization command. The system pauses the current MPPT algorithm and controls the DC converter to make the operating point of the solar cell array quickly jump to multiple preset, representative voltage points. It briefly stops at each voltage point and measures the power. By comparing the power values ​​of these voltage points, the voltage range where the point with the highest power is located is determined and defined as a high-voltage candidate region.

5. The solar MPPT power supply control method for a smart cabinet according to claim 1, characterized in that, The step of locating the global maximum power point voltage within the high-voltage candidate region using an optimization algorithm and driving a rapid transition of the operating point includes: Within the high-voltage candidate region, a set of voltage values ​​is initialized. By iteratively updating the speed and position of each voltage value, after a finite number of iterations, the voltage value corresponding to the population optimal solution is identified as the global maximum power point voltage. Based on the global maximum power point voltage, the central control unit immediately calculates the corresponding DC converter duty cycle and sends a PWM control signal to drive the DC converter to adjust the operating point of the solar cell array to the global maximum power point voltage in the shortest possible time.

6. The solar MPPT power supply control method for a smart cabinet according to claim 1, characterized in that, After completing the rapid transition of the operating point, the system switches to a steady-state tracking method for steady-state tracking and to maintain closed-loop monitoring, including: After completing the rapid transition of the operating point, the control algorithm immediately uses the global maximum power point voltage as the starting point to perform small-step perturbation and observation methods to ensure that the system operates stably at the power peak. During this period, the system returns to the execution monitoring and judgment process in parallel to continuously monitor the next possible occlusion event, forming a complete closed-loop control.

7. A solar-powered MPPT power supply control system for intelligent cabinets, characterized in that, include: A processor and a memory, the memory storing computer program instructions that, when executed by the processor, implement a solar MPPT power supply control method for a smart cabinet according to any one of claims 1-6.

Images