A UPS uninterruptible power supply control management method

Abstract

This application discloses a UPS (Uninterruptible Power Supply) control and management method, relating to the field of uninterruptible power supply technology. The method includes: collecting transient response signals from each power supply output terminal to form a time-series dataset, obtaining impedance spectrum characteristics at different frequencies, and generating an impedance spectrum diagram; calculating the difference in impedance magnitude and phase angle between power supplies at the same frequency in the initial power supply sequence based on the impedance spectrum diagram, obtaining a difference sequence; selecting deviating power supplies according to an optimization mechanism, generating a deviation sequence; calculating the impedance matching degree and circulating current loss reduction rate of each power supply in the pre-power supply sequence, re-determining the power supply to obtain the final power supply sequence, and performing power supply. By collecting and analyzing the transient response signals from the power supply output terminals to obtain impedance spectrum characteristics, accurate impedance matching between power supplies is achieved, reducing circulating current and resonance, and improving power supply efficiency; determining the optimal topology and power supply connection method improves the stability and reliability of the power supply system.

Description

Technical Field

[0001] This invention relates to the field of uninterruptible power supply (UPS) technology, and more particularly to a UPS uninterruptible power supply control and management method. Background Technology

[0002] A UPS (Uninterruptible Power Supply) is a power supply that converts the DC output from a battery into continuous AC power when normal AC power is interrupted. It is primarily used to provide uninterrupted power to single computers, computer network systems, or other power electronic equipment.

[0003] In the event of a power outage, when using uninterruptible power supplies (UPS) to provide power, multiple UPS units are pre-configured in parallel to ensure continuous power supply to critical loads, with one unit serving as a redundant backup. When one power supply fails, the redundant power supply immediately takes over the load. During online capacity expansion (such as adding a new server rack) or maintenance of a UPS, the load can be temporarily transferred to other power supplies through parallel operation, achieving uninterrupted maintenance, avoiding overloading of a single power supply, and extending the overall lifespan. This improves power supply reliability and fault tolerance, expands output capacity, and enables load sharing and thermal management. During low-load periods, only a portion of the parallel power supplies are activated to improve overall efficiency.

[0004] However, in existing technologies, due to the limited amount of electricity stored in a single uninterruptible power supply, when multiple power supplies are connected in parallel, there are problems such as circulating current loss, resonance risk, and local overheating caused by impedance mismatch, which can lead to abnormal voltage, efficiency reduction, or even power supply damage, ultimately threatening the stability of power supply. In addition, the power topology after parallel connection is fixed, which leads to the problem of single-point control risk. Summary of the Invention

[0005] This application provides a UPS uninterruptible power supply control and management method, which solves the problems of circulating current loss, resonance risk and local overheating caused by impedance mismatch in the prior art, and achieves the technical effects of impedance optimization matching and topology optimization.

[0006] This application provides a UPS (Uninterruptible Power Supply) control and management method, the method comprising:

[0007] S1: Monitor the mains power supply status. If the power supply is interrupted, generate the initial power supply sequence. Collect the transient response signal of each power supply output terminal in the initial power supply sequence to form a time series dataset. Analyze the time series dataset to obtain the impedance spectrum characteristics at different frequencies, and then generate an impedance spectrum diagram.

[0008] S2: Based on the impedance spectrum, calculate the differences in impedance magnitude and phase angle between power supplies at the same frequency in the initial power supply sequence, and obtain several sets of difference number sequences; compare the difference number in the difference number sequence of each power supply with the difference threshold. If the difference number is greater than the difference threshold, it is marked as unqualified. Calculate the unqualified rate of each power supply and sort them in descending order; select several deviation power supplies according to the optimization mechanism and generate a deviation sequence; adjust the equipment parameters of each power supply one by one according to the deviation sequence, and update the unqualified rate and deviation sequence of each power supply after adjustment until there are no deviation power supplies or the maximum number of iterations is reached, and generate the pre-power supply sequence;

[0009] S3: Calculate the impedance matching degree and circulating current loss reduction rate of each power source in the pre-power supply sequence, redetermine the power supply sequence to obtain the final power supply sequence, and then supply power.

[0010] Furthermore, the time series dataset is analyzed to obtain the impedance spectrum characteristics at different frequencies, including: performing a fast Fourier transform on the time series dataset to convert the time-domain signal into a frequency-domain signal, obtaining the amplitude and phase information of current and voltage at different frequencies, calculating the impedance magnitude for each frequency point according to the complex form of Ohm's law, and calculating the phase angle based on the phase of voltage and current.

[0011] The impedance spectrum characteristics refer to the opposition of the power source to the current at different frequencies, including the impedance magnitude and phase angle.

[0012] Generating an impedance spectrum involves: establishing two coordinate systems with frequency as the abscissa and impedance magnitude and phase angle as the ordinates respectively; plotting the impedance magnitude and phase angle data corresponding to each frequency point in the corresponding coordinate systems; and connecting the data with smooth curves to form curves showing the impedance magnitude changing with frequency and the phase angle changing with frequency, which together constitute the impedance spectrum.

[0013] Furthermore, the optimization mechanism includes: for the initial power supply sequence, setting the proportion of each power supply to the total number of power supplies minus one as a single failure rate, screening out power supplies with a failure rate greater than the single failure rate and arranging them in descending order, and selecting the first 75% of power supplies in sequence to form a deviation sequence.

[0014] Based on the deviation sequence, the equipment parameters are adjusted one by one, including: using the impedance of the reference power supply at the critical frequency point as a benchmark, determining the target impedance of each power supply at the critical frequency point according to the impedance matching requirements; calculating the parameter adjustment amount of the digitally adjustable inductor and capacitor array according to the impedance calculation formula based on the target impedance and the current impedance characteristics of the power supply; adjusting the digitally adjustable inductor through the control circuit according to the calculated inductor parameter adjustment amount; and switching the capacitor combination in the capacitor array according to the calculated capacitor parameter adjustment amount.

[0015] Furthermore, the impedance matching degree and circulating current loss reduction rate of each power source in the pre-power supply sequence are calculated, including:

[0016] Based on the adjusted power supply, arbitrarily select a reference power supply and calculate the relative deviations of the impedance magnitude and phase angle of other power supplies from the reference power supply at key frequency points; calculate the impedance matching degree by weighted average of the relative deviations of the impedance magnitude and phase angle of multiple power supplies at multiple key frequency points; obtain the corresponding circulating current loss according to the circulating current and the resistance of the circulating current path before and after adjustment, and obtain the circulating current loss reduction rate according to the circulating current loss before and after adjustment.

[0017] The impedance matching degree is used to measure the consistency of each power supply impedance at key frequency points; the circulating current loss reduction rate is used to measure the degree of reduction in circulating current loss after impedance matching.

[0018] Furthermore, the method also includes: S4: obtaining impedance matching degree and power health data of the final power supply sequence, forming a power status report, modeling the multi-power supply connection relationship as a graph network, establishing an objective function to maximize power supply stability, and determining the optimal topology and power supply connection method;

[0019] S5: Real-time monitoring of environmental stress distribution data, construction of stress distribution heat map, identification of high stress areas; adjustment of power output curve of power supply in high stress areas based on optimal topology.

[0020] Further, step S3 includes: S31: Real-time monitoring is performed during the power supply cycle. If an abnormal state is detected, the abnormality category is determined. If the abnormality category belongs to the first category, a self-healing mechanism is triggered to generate a repair instruction, the self-healing response time is recorded, and a stability assessment is performed after self-healing. If the abnormality category does not belong to the first category, the faulty power supply is isolated, and the power supply path is reconstructed. The self-healing response time is the time interval from detecting the abnormality to triggering self-healing.

[0021] The first category includes current anomalies caused by current imbalance, local overload, and voltage fluctuations; the self-healing mechanism automatically generates adjustment measures based on the determined anomaly category, combined with the changing trends and interrelationships of voltage, current, and resistance, to achieve dynamic self-healing.

[0022] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0023] By monitoring the mains power supply status and basic power information in real time, qualified power supply equipment is quickly screened to ensure rapid switching to backup power during mains power outages, guaranteeing normal power supply. By collecting and analyzing transient response signals at the power supply output terminals, impedance spectrum characteristics are obtained, and power supply screening and parameter adjustments are performed based on these characteristics, achieving precise impedance matching between power supplies, reducing circulating currents and resonances, and improving power supply efficiency. By establishing an objective function to maximize power supply stability and performing optimization calculations based on graph networks, the optimal topology and power supply connection method are determined, further improving the stability and reliability of the power supply system. By monitoring environmental stress distribution data in real time and constructing a stress distribution heatmap, high-stress areas are identified, and the output power curves of power supplies in these areas are adjusted based on the optimal topology, enhancing the adaptability and stability of the power supply under high-stress environments. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of a UPS uninterruptible power supply control and management method according to an embodiment of the present invention. Detailed Implementation

[0025] To facilitate understanding of the present invention, a more complete description of this application will be given below with reference to the accompanying drawings, which illustrate preferred embodiments of the invention. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to enable a more thorough and complete understanding of the disclosure of the present invention.

[0026] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains; the terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to limit the invention; the term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0027] Example 1: As Figure 1 As shown, a UPS uninterruptible power supply control and management method includes:

[0028] S1: Monitor the mains power supply status. If the power supply is interrupted, generate the initial power supply sequence. Collect the transient response signal of each power output terminal in the initial power supply sequence to form a time series dataset. Analyze the time series dataset to obtain the impedance spectrum characteristics at different frequencies, and then generate an impedance spectrum diagram.

[0029] In some embodiments, the control unit monitors the mains power supply status in real time, acquiring the mains power supply status in real time. When the mains power is supplying power to the load, the mains power supply status is normal; when a mains power supply failure occurs, the mains power supply status is interrupted. During a power interruption, an uninterruptible power supply (UPS) is required to provide power to ensure normal power usage. Mains power refers to industrial frequency alternating current, characterized by the three common quantities of alternating current: voltage, current, and frequency.

[0030] In some embodiments, if there is a power outage, a suitable power source is selected based on the uninterruptible power supply (UPS) equipment. Specifically, basic information of all power sources is obtained, including standard voltage, standard current, and available power. All UPS equipment is traversed, and the basic information is checked to ensure it meets the power supply conditions. Specifically, the output voltage of each device is checked to ensure it is within the standard voltage range. For devices with compliant voltage, their rated output current is further checked to ensure it meets the standard current requirement. Finally, the remaining power of the device is checked to ensure it meets the available power standard. Only power sources that simultaneously meet these three conditions are selected. The identifiers of the compliant power sources are obtained. These identifiers are unique to identify power sources and include their specific location and usage status. An initial power supply order is generated based on the selected power source identifiers.

[0031] In some embodiments, transient response signals are collected from each power supply output terminal in the initial power supply sequence. These transient response signals refer to the current and voltage data output by the power supply. A pre-set acquisition period and period interval (e.g., 20ms) are used. Current and voltage data are collected once every period interval within the acquisition period, forming a time-series dataset. Transient response changes are identified based on this time-series dataset. The time-series dataset includes the transient signals corresponding to each sampling time point, arranged in chronological order. For example, at time point t1, the acquired current value is I1 and the voltage value is V1; at time point t2 (t2=t1+20ms), the acquired current value is I2 and the voltage value is V2, and so on, forming a data set containing multiple time points and their corresponding current and voltage values.

[0032] Time series datasets are digital records of transient responses. A transient response is a continuous physical process, while a time series dataset transforms this continuous process into a series of discrete data points through discrete sampling. By analyzing these time series data, the changing trends and characteristics of the power supply's output transient response can be reconstructed, such as rise time, fall time, and overshoot, thereby providing a deeper understanding of the power supply system's performance during transient processes.

[0033] In some embodiments, transient response refers to the change in output of a power supply system over time when subjected to sudden external changes (such as abrupt changes in input signals or sudden changes in load). For uninterruptible power supply (UPS) systems, when mains power is interrupted, backup power is activated, or the load changes suddenly, the current and voltage output of the power supply will change rapidly within a short period of time. This change process is the transient response of the power supply output, which reflects the dynamic responsiveness of the power supply system to external changes. The change in transient response needs to be quantified based on real-time acquired current and voltage data.

[0034] In some embodiments, time-series datasets are analyzed to obtain impedance spectrum characteristics at different frequencies, thereby generating impedance spectrum diagrams. Specifically, the impedance spectrum characteristics represent the power supply's opposition to current at different frequencies, including impedance magnitude and phase angle. The impedance spectrum characteristics describe the characteristics of the power supply impedance changing with frequency, reflecting the power supply's opposition to current at different frequencies. The impedance magnitude is the magnitude of the impedance, and the phase angle represents the phase relationship. At different frequencies, the impedance of internal components such as inductors and capacitors in the power supply changes, causing the overall impedance of the power supply to change with frequency. This change pattern is the impedance spectrum characteristic.

[0035] A Fast Fourier Transform (FFT) is performed on the time-series dataset to convert the time-domain signal into a frequency-domain signal, obtaining the amplitude and phase information of current and voltage at different frequencies. For each frequency point, the impedance magnitude is calculated using the complex form of Ohm's law. ,in This is the impedance magnitude, in ohms. This refers to the voltage amplitude, measured in volts. This represents the current amplitude, measured in amperes. The phase angle is calculated based on the phase of the voltage and current. ,in, The phase angle, For the phase of the voltage, The phase of the current is expressed in degrees or radians.

[0036] In some embodiments, generating an impedance spectrum includes: establishing two coordinate systems with frequency as the abscissa and impedance magnitude and phase angle as the ordinates respectively; plotting the impedance magnitude and phase angle data corresponding to each frequency point in the corresponding coordinate systems; and connecting the data points with smooth curves to form curves showing the impedance magnitude versus frequency and curves showing the phase angle versus frequency, which together constitute the impedance spectrum. In this embodiment, the impedance spectrum consists of two curves: one is the impedance magnitude-frequency curve, with frequency on the abscissa and impedance magnitude on the ordinate, used to show how the power supply impedance changes with frequency; the other is the phase angle-frequency curve, with frequency on the abscissa and phase angle on the ordinate, reflecting how the power supply impedance phase changes with frequency. These two curves provide a visual understanding of the power supply's impedance characteristics at different frequencies.

[0037] S2: Based on the impedance spectrum, calculate the differences in impedance magnitude and phase angle between power supplies at the same frequency in the initial power supply sequence, obtaining several sets of difference number sequences; compare the difference number in the difference number sequence of each power supply with the difference threshold. If the difference number is greater than the difference threshold, it is marked as unqualified. Calculate the unqualified rate of each power supply and sort them in descending order. For example, if a power supply has 10 difference numbers in its difference number sequence, and 4 of them are greater than the difference threshold, then the unqualified rate of this power supply is 25%. The total number of differences must be one less than the total number of power supplies; select several deviation power supplies according to the optimization mechanism to generate a deviation sequence; adjust the equipment parameters of each power supply one by one according to the deviation sequence, and update the unqualified rate and deviation sequence of each power supply after adjustment until there are no deviation power supplies or the maximum number of iterations is reached, generating the pre-power supply sequence.

[0038] Several sets of difference number sequences are obtained. Specifically, during the comparison process, according to the initial power supply order, one power source is randomly selected and compared with the other power sources to generate a difference number sequence. This process is repeated. After the comparisons are completed, several difference number sequences are obtained. The number of difference number sequences is the same as the number of power sources in the initial power supply order.

[0039] The difference threshold is preset. Specifically, data on impedance magnitude and phase angle differences of power supply equipment under similar power supply environments and load conditions are collected. These data are statistically analyzed to calculate their average and standard deviation. Based on the statistical analysis results, the average plus a certain multiple of the standard deviation is used as the initial difference threshold. The multiple can be determined according to actual needs and experience, generally 1-2 times. For example, if the analysis shows an average impedance magnitude difference of 0.5 and a standard deviation of 0.2, and a standard deviation of 1.5 times is chosen, then the initial impedance magnitude difference threshold is 0.5 + 1.5 × 0.2 = 0.8. Experiments are conducted in actual power supply systems, using the initial difference threshold to screen and adjust power supply equipment. The performance of the adjusted system, such as circulating current loss and power supply stability, is observed. If the system performance improvement after adjustment is not significant or other problems occur, the difference threshold is adjusted appropriately, and the experiment is repeated until satisfactory results are obtained. For example, data on impedance magnitude differences of power supply equipment during 100 previous mains power outages were collected. The calculated average of these difference data is 0.4, and the standard deviation is 0.15. Choosing 1.5 times the standard deviation as the multiplier, the initial impedance magnitude difference threshold is 0.4 + 1.5 × 0.15 = 0.625. In actual experiments, it was found that the system circulating current loss remained high even after adjusting with this threshold. Therefore, the difference threshold was adjusted to 0.55, and the experiment was repeated. A significant improvement in system performance was observed, thus the final difference threshold was determined to be 0.55. The method for setting the phase angle difference threshold is similar, and it can be determined based on historical phase angle difference data.

[0040] The optimization mechanism includes: for the initial power supply sequence, setting the proportion of each power supply to the total number of power supplies minus one as a single defect rate; filtering out power supplies with a defect rate greater than the single defect rate and arranging them in descending order; and selecting the top 75% of power supplies to form a deviation sequence. For example, for five power supplies a, b, c, d, and e, each power supply accounts for 25% of the total number of power supplies minus one. Their defect rates are calculated to be 75%, 67%, 25%, 25%, and 50%, respectively. Power supplies a, b, and e have a defect rate greater than 25%. Arranging them in descending order, the top 75% are selected. Since there are only three in this example, the first two power supplies a and b are selected to form the deviation sequence. That is, a and b are adjusted first, and then the defect rates and deviation sequences of the power supplies are recalculated.

[0041] In a power supply system, different power sources are interconnected and influence each other. A problem with a single power source can have a cascading effect on other power sources and the entire system. Using the proportion of each power source out of the total number of power sources as a benchmark means that each power source must be compared with other power sources. Only when its failure rate exceeds this relative standard is it considered that two or more power sources deviate from this standard and are thus deemed unqualified, avoiding redundant adjustments. Simultaneously adjusting power sources with high failure rates minimizes the number of adjustments and improves efficiency. Selecting the top 75% of power sources for adjustment allows for reasonable control of the scale and resource investment while ensuring the effectiveness of the adjustments. Adjusting all unqualified power sources simultaneously would lead to insufficient resources and excessively long adjustment cycles, hindering the smooth progress of the adjustment work. Selecting the top 75% of power sources allows for concentrated efforts to solve most problems within limited resources and time, improving adjustment efficiency. Considering fault tolerance, even without adjusting all unqualified power sources, prioritizing the top 75% can significantly improve the overall performance of the system. Furthermore, after adjusting the first 75% of the power supply, it is possible to decide whether to adjust the remaining substandard power supply based on the actual operation of the system, thereby allocating resources more flexibly.

[0042] In some embodiments, the selection rule for the reference power supply is as follows: in the pre-power supply sequence, the standard deviation of the impedance magnitude of each power supply at the set of key frequency points is calculated, and the power supply with the smallest standard deviation is selected as the reference power supply. The purpose is to select a power supply with the most stable impedance characteristics as the reference for impedance matching, thereby prioritizing the reliability of the system reference and effectively reducing the overall resonance risk. The impedance characteristics of the reference power supply are considered to be the most ideal or closest to ideal in the current system. Other power supplies are adjusted to align with it to achieve overall impedance consistency. The reference power supply is not fixed; it needs to be re-evaluated and selected after each adjustment.

[0043] Critical frequency points refer to specific frequencies that have a significant impact on the stability, circulating current loss, or resonance risk of the power supply system. The method for selecting critical frequency points is as follows: by analyzing the impedance spectrum, the impedance magnitude frequency curve data is scanned to find all peaks (local maxima) and valleys (local minima) as critical frequency points. Peaks refer to the parallel resonant frequency points, at which the total impedance of the parallel power supply system reaches a local maximum value; valleys refer to the series resonant frequency points, at which the total impedance of the system reaches a local minimum value.

[0044] When adjusting the parameters of other power supplies based on the reference power supply, it is necessary to ensure that the impedance characteristics of all power supplies remain highly consistent at each critical resonant frequency point, so as to effectively suppress the circulating current caused by impedance mismatch near the resonant frequency point.

[0045] In some embodiments, the device parameters are adjusted sequentially according to the deviation sequence, including: using the impedance of a reference power supply at a critical frequency point as a benchmark, determining the target impedance of each power supply at the critical frequency point according to impedance matching requirements; calculating the parameter adjustment amount of the digitally adjustable inductor and capacitor array according to the impedance calculation formula based on the target impedance and the current impedance characteristics of the power supply; adjusting the digitally adjustable inductor through a control circuit according to the calculated inductor parameter adjustment amount; and switching the capacitor combination in the capacitor array according to the calculated capacitor parameter adjustment amount. For example, if the impedance magnitude of a power supply at the critical frequency point is too large, it may be necessary to decrease the value of the digitally adjustable inductor or increase the value of the capacitor array. Specifically, based on known parameters such as the current impedance, target impedance, and frequency, the impedance calculation formula is as follows:

[0046]

[0047] in, Impedance, in ohms; For resistance, Angular frequency, in radians per second; Inductance, measured in Henry; Capacitance, measured in farads;

[0048] The adjustment values ​​for the inductor and capacitor are obtained by solving equations. The current impedance and the target impedance are known. and angular frequency Equations are established to solve for the adjustment amounts of the inductor and capacitor, satisfying the following equations:

[0049]

[0050] in, It is the inductance adjustment amount. It is the capacitance adjustment amount.

[0051] After each adjustment, the corresponding power supply failure rate and deviation sequence are updated, and adjustments are made again until there are no deviation power supplies. The maximum number of iterations is dynamically set based on the total number of power supplies in the initial power supply sequence. The more power supplies there are, the larger the maximum number of iterations should be. The number of iterations should also be determined based on the actual network complexity and data computation volume; for example, it should not exceed 10 iterations to avoid excessive data computation and impact on data efficiency. If deviation power supplies still exist after reaching the maximum number of iterations, the pre-power supply sequence is still output normally, and step S3 is executed.

[0052] Preferably, in addition to adjusting the equipment parameters, adjustments can also be made to the power supply lines. The routing of the power supply lines can be analyzed, and for excessively long lines, the route can be replanned to minimize the length of the line between the critical power source and the load. Longer lines increase line resistance and inductance, leading to increased impedance and signal attenuation. Shortening the line length can reduce these adverse effects. For example, if the line from a power source to the load has a detour, it can be replanned to create a straight connection, reducing unnecessary line detours.

[0053] S3: Calculate the impedance matching degree and circulating current loss reduction rate of each power source in the pre-power supply sequence, redetermine the power supply to obtain the final power supply sequence, generate power supply control commands according to the final power supply sequence to supply power, and reduce circulating current and resonance.

[0054] In some embodiments, calculating the impedance matching degree and circulating current loss reduction rate of each power supply in the pre-power supply sequence includes: arbitrarily selecting a reference power supply based on the adjusted power supply, here randomly selected from several reference power supplies generated during the above adjustment; calculating the relative deviations of the impedance magnitude and phase angle of other power supplies from the reference power supply at key frequency points; calculating the impedance matching degree by weighted averaging the relative deviations of the impedance magnitude and phase angle of multiple power supplies at multiple key frequency points; obtaining the corresponding circulating current loss based on the circulating current and the resistance of the circulating current path before and after adjustment; and obtaining the circulating current loss reduction rate based on the circulating current loss before and after adjustment.

[0055] The impedance matching degree is used to measure the consistency of each power supply impedance at key frequency points; the circulating current loss reduction rate is used to measure the degree of reduction in circulating current loss after impedance matching.

[0056] For the impedance magnitude, the formula for calculating the relative deviation of the impedance magnitude is:

[0057]

[0058] in, It is the relative deviation of the impedance magnitude. It is the impedance module of the reference power supply at the critical frequency point f. It is the impedance magnitude of the i-th power source at the same frequency.

[0059] For the phase angle, the formula for calculating the relative deviation of the phase angle is:

[0060]

[0061] in, It is the relative deviation of the phase angle. It is the phase angle of the reference power supply at the critical frequency point f. It is the phase angle of the i-th power source at the same frequency point; if the range of the phase angle is When performing calculations, the angles need to be normalized.

[0062] The impedance matching degree is calculated by weighted averaging the relative deviations of impedance magnitudes and phase angles at multiple key frequency points and multiple power sources. For example, if there are n key frequency points and m power sources, the impedance matching degree is:

[0063]

[0064] in, It refers to the impedance matching degree. The closer the impedance matching degree is to 1, the better the impedance matching effect. and These are the initial weights for the critical frequency points and the power supplies, used to measure their importance; n is the total number of critical frequency points, and m is the total number of power supplies after adjustment. During calculation, the relative deviations of the phase angle and impedance magnitude need to be normalized beforehand to ensure dimensional consistency.

[0065] Before impedance matching, measure the circulating current loss in the system. By measuring the circulating current and the resistance of the circulating path According to the formula The calculation was performed. After impedance matching, the circulating current loss in the system was measured again. Using the same method described above, the reduction rate of circulating loss is calculated based on the results of the two calculations. :

[0066]

[0067] The reduction rate of circulating current loss reflects the effect of impedance matching on reducing circulating current loss. The higher the reduction rate, the better the impedance matching effect.

[0068] Based on impedance matching and circulating current loss reduction rate, the final power supply sequence is determined, and power supply control commands are generated according to the final power supply sequence to supply power.

[0069] In this embodiment, impedance matching reflects the degree of compatibility between the power supply output and the load. When impedance matching is good, the power supply can efficiently transfer electrical energy to the load, reducing energy loss and reflection, and improving power supply efficiency and stability. If impedance mismatch occurs, it will lead to reduced power transmission efficiency, and may cause problems such as overheating and voltage fluctuations, affecting the performance and lifespan of both the power supply and the load. Therefore, obtaining impedance matching information helps to comprehensively evaluate the actual operating status of the power supply during power supply, providing crucial information for generating accurate power status reports.

[0070] Continue to refer to Figure 1The method further includes: S4: obtaining impedance matching degree and power health data of the final power supply sequence, forming a power status report, modeling the multi-power supply connection relationship as a graph network, establishing an objective function to maximize power supply stability, and determining the optimal topology and power supply connection method.

[0071] In some embodiments, the power supply health data includes voltage stability, current output capability, temperature, operating time, and historical fault records. Impedance matching is pre-classified into four levels: Excellent, Good, Medium, and Poor. Based on the impedance matching range corresponding to each level, the impedance matching after each power supply is combined with the corresponding power supply health data to analyze the power supply's performance and potential risks at different levels, generating a power supply status report. For example, when the impedance matching is poor, if large voltage fluctuations and excessively high temperatures are also observed, the power supply can be determined to be in an unhealthy state, potentially indicating a fault. Finally, these analysis results are integrated into the power supply status report to provide a basis for subsequent power supply management and maintenance.

[0072] Specifically, voltage stability refers to the fluctuation range of the power supply's output voltage within a specified time, reflecting the power supply's ability to maintain a stable voltage output. Current output capability refers to the maximum current the power supply can provide and its current output characteristics under different loads. Temperature refers to the operating temperature of the power supply; excessively high or low temperatures can affect its performance and lifespan. Operating time refers to the cumulative operating time of the power supply, used to assess its aging status. Historical fault records show the types of faults the power supply has experienced, their occurrence times, and repair status, helping to predict potential future faults.

[0073] In some embodiments, the multi-power supply connection relationship is modeled as a graph network, including: treating all power supplies as nodes in the graph network; representing the lines connecting two power supplies as edges in the graph network based on the actual connection lines between the power supplies; the attributes of the edges include the impedance and length parameter values ​​of the lines; and establishing a g×g adjacency matrix A based on the relationship between nodes and edges, where g is the number of nodes, and if there is an edge connecting nodes i and j, then... (For undirected graphs), otherwise Construct the impedance matrix Z corresponding to the adjacency matrix, and the elements in the impedance matrix... This represents the impedance value of the line connecting node i and node j.

[0074] Maximizing power supply stability means ensuring that the power supply system maintains stable voltage and current output under various operating conditions, reducing fluctuations and interruptions. An objective function for maximizing power supply stability is established, comprising three indices: minimizing voltage fluctuations, maximizing current stability, and maximizing reliability. Corresponding weighting coefficients are pre-set, and the obtained index values ​​are normalized before being weighted and summed to obtain the objective function for maximizing power supply stability. The function formula is as follows:

[0075]

[0076] in, It is about power supply stability; , , These are the corresponding weighting coefficients, initially set to 0.4, 0.4, and 0.2, used to emphasize the importance of voltage and current changes. The weighting coefficients need to be dynamically adjusted and optimized based on actual applications and experimental results; t is the number of power supplies in the current line. It is about minimizing voltage fluctuations. It is the actual output voltage of the i-th power supply. It is the target output voltage of the i-th power supply; It is to maximize current stability. It is the actual output current of the i-th power supply. is the target output current of the i-th power supply. Variance is a statistic that measures the dispersion of data. The larger the variance value, the wider the range and the greater the fluctuation of the current, and the worse the stability. The smaller the variance value, the more concentrated the current is around the mean, the smaller the fluctuation, and the higher the stability. It is about maximizing reliability. This represents the failure probability of the i-th power supply, obtained through statistical analysis of historical records. When calculating power supply stability, the data for the three indicators—minimizing voltage fluctuations, maximizing current stability, and maximizing reliability—need to be normalized to ensure consistent dimensions. The purpose of the obtained power supply stability data is to provide a comparison and select the most stable power supply device; the unit of measurement does not affect the effectiveness of the comparison.

[0077] In this application, a graph network is constructed to visually represent the interconnections of multiple power sources, where power sources are nodes and connecting lines are edges. The graph network clearly shows the topology and connection methods between power sources. The objective function is established to find an optimal power source configuration and operation mode within the given graph network structure, maximizing power supply stability. The graph network provides the physical model and constraints for the objective function, which is then optimized based on the graph network's structure and parameters. For example, in the graph network, the impedance of the connecting lines between power sources affects voltage and current transmission, thus influencing the voltage fluctuation and current stability indices in the objective function.

[0078] The voltage fluctuation and current stability metrics in the objective function are closely related to the edge properties (such as line impedance) in the graph network. Line impedance affects voltage drop and current distribution, thus influencing the actual output voltage and current of the power supply. System reliability metrics are related to the node properties (such as the health status of the power supply) and topology in the graph network. For example, if a critical power supply fails, it may lead to a decrease in the reliability of the entire system, and the graph network can clearly demonstrate this impact.

[0079] The network topology is optimized based on the objective function of maximizing power supply stability to obtain the optimal topology. Specifically, a set of graph network topologies and power supply operating parameters are randomly generated as initial solutions. Based on the current topology and operating parameters, the objective function is calculated, and the optimization algorithm is used to adjust the topology and operating parameters to generate new objective solutions. This process continues until the termination condition is met (such as reaching the maximum number of iterations or the change in the objective function value is less than the micro-change threshold), at which point the current optimal topology, operating parameters, and power supply connection method are output.

[0080] The micro-change threshold is used to determine whether the optimization process has converged. The value is preset according to the required accuracy or obtained by statistical analysis of historical data, and no specific restrictions are imposed here. During the iterative process of the optimization algorithm, the objective function value after multiple iterations will be calculated continuously, and the absolute change of the objective function value between two adjacent iterations will be calculated. When the absolute change is less than the micro-change threshold, it is considered that the change of the objective function value is small enough, the optimization process has reached a stable state, and the iteration can be terminated.

[0081] This application utilizes a solid-state switch matrix to dynamically switch power supply connection methods. A solid-state switch matrix is ​​a switching network composed of solid-state switches (such as field-effect transistors and insulated-gate bipolar transistors). By controlling the on and off states of these solid-state switches, the connection methods between power supplies can be dynamically changed. For example, in a multi-power supply system, the solid-state switch matrix can connect or disconnect a power supply to a specific load, or change the parallel or series connection relationship between power supplies, according to system requirements. This enables flexible power supply configuration and optimized operation, adjusting the power supply connection methods in real time based on load changes, power supply health status, and environmental conditions, thereby improving the reliability and efficiency of the power supply system. For instance, when a power supply fails, the solid-state switch matrix can quickly isolate it from the system and switch the load to another normally functioning power supply, ensuring continuous power supply to the system.

[0082] The optimal topology is the ideal state of power connection obtained by optimizing the objective function. The role of a solid-state switch matrix is ​​to dynamically adjust the power connection of the system during actual operation to approach or reach the optimal topology. Through the dynamic switching of the solid-state switch matrix, the power connection can be adjusted in real time, ensuring that the system always operates in a state close to the optimal topology, thereby improving power supply stability and efficiency.

[0083] S5: Real-time monitoring of environmental stress distribution data, construction of stress distribution heat map, identification of high stress areas; adjustment of power output curve of power supply in high stress areas based on optimal topology.

[0084] In some embodiments, multiple sensors are used to monitor environmental stress distribution data in real time, such as temperature sensors, humidity sensors, pressure sensors, and vibration sensors. These sensors are distributed in key parts of the power supply system and the surrounding environment. Data from the sensors is collected in real time by a data acquisition system and transmitted to a monitoring center for analysis and processing. Stress distribution data refers to the magnitude and distribution of stress exerted on an object by various environmental factors (such as temperature, humidity, pressure, vibration, etc.) within a specific area. In a power supply system, environmental stress affects the performance and lifespan of the power supply. For example, high temperatures can lead to poor heat dissipation, reducing its output power and reliability; vibration may cause internal components of the power supply to loosen or be damaged. Constructing a stress distribution heatmap includes: dividing the monitoring area into several grid cells; calculating the average stress value within each grid cell based on the collected stress data; using different colors to represent different stress value ranges; and mapping the colors onto the corresponding grid cells to form a stress distribution heatmap. For example, red can represent high-stress areas, blue can represent low-stress areas, and the intensity of the color indicates the magnitude of the stress value.

[0085] By analyzing stress distribution heatmaps, high-stress areas can be visually identified. Generally, darker areas indicate higher stress values, i.e., high-stress areas. Furthermore, a stress threshold can be set, defining areas with stress values ​​exceeding this threshold as high-stress areas. This application does not impose specific limitations in this regard.

[0086] Adjusting the power output curve of a power source in a high-stress region based on optimal topology involves: establishing a mathematical model of the environment and power source output power based on the power source's characteristics and historical data; determining the adjustment strategy for the output power curve based on the analysis results; and sending adjustment commands to change the power source's output power curve. For example, when the temperature rises, the power source output power is appropriately reduced to avoid overheating; when vibration intensifies, the power source's operating mode is adjusted to reduce the impact of vibration on the output power.

[0087] The output power curve describes how a power supply's output power changes over time or other variables (such as load and voltage), reflecting the power supply's output capability under different operating conditions. In high-stress areas, environmental factors can adversely affect the power supply's performance, leading to a decrease or instability in its output power. By adjusting the power supply's output power curve, the power supply can maintain a stable output under high-stress environments, improving the reliability of the power supply system. Adjusting the output power curve can prevent the power supply from operating under overload conditions during high stress, reducing the probability of failure; at the same time, it allows for the rational allocation of the power supply's output power according to the load requirements, improving energy utilization efficiency.

[0088] Specifically, establishing a mathematical model of the environment and power supply output power includes: collecting power supply output power data under different environmental stress conditions; and establishing a mathematical model between environmental parameters (such as temperature, humidity, vibration, etc.) and power supply output power based on regression analysis and neural networks. The model includes both linear and nonlinear models. For example, a linear model is as follows:

[0089]

[0090] in, For output power, For the intercept term, For environment variables The coefficients are obtained through data fitting. For error terms, The number of environmental variables is specified. The nonlinear model is built upon a neural network model, specifically as follows: Multi-dimensional environmental data (such as temperature, humidity, input voltage, light intensity, etc.) and corresponding power output values ​​are collected. Outliers are removed, missing values ​​are processed, and normalization is performed. The input features are scaled to [0, 1] to accelerate neural network convergence. Interaction terms (such as temperature × input voltage) are generated to capture nonlinear interaction effects. The training set (70%), validation set (15%), and test set (15%) are divided proportionally. The network structure is designed: input layer: the number of neurons is the same as the number of environmental features; hidden layers: 1-3 layers (the number of layers can be increased for complex problems); the number of neurons per layer: 32-128 (the optimal value is determined experimentally); activation function, loss function, and optimizer are set respectively; input data is processed through network layers to calculate output. The gradient of the loss function with respect to the weights is calculated, the weights are updated to minimize the loss, and a fixed number of rounds or early stopping is set (the validation set stops if the loss does not decrease for 5 consecutive rounds). The trained model is embedded into the power control system to receive environmental data in real time and predict the output power. The power parameters are adjusted based on the deviation between the predicted value and the target value.

[0091] Based on the power supply's performance specifications and safety requirements, threshold values ​​for environmental parameters are set. When environmental parameters exceed these thresholds, the power supply's output power is adjusted according to pre-defined adjustment rules. Sensors are used to monitor environmental parameters in high-stress areas in real time, and the data is transmitted to the control system. The control system dynamically adjusts the power supply's output power curve based on the environmental parameters and adjustment rules. Simultaneously, data during the adjustment process is recorded for subsequent analysis and optimization of the adjustment strategy.

[0092] In this embodiment, by monitoring the mains power supply status and basic power information in real time, qualified power supply devices are quickly screened to ensure rapid switching to backup power in the event of a mains power outage, guaranteeing normal power usage. By collecting and analyzing the transient response signals at the power supply output, impedance spectrum characteristics are obtained, and power supply screening and parameter adjustment are performed based on these characteristics, achieving precise impedance matching between power supplies, reducing circulating current and resonance, and improving power supply efficiency. By establishing an objective function to maximize power supply stability and performing optimization calculations based on graph networks, the optimal topology and power supply connection method are determined, further improving the stability and reliability of the power supply system. By monitoring environmental stress distribution data in real time and constructing a stress distribution heatmap, high-stress areas are identified, and the output power curves of the power supplies in these areas are adjusted based on the optimal topology, enhancing the adaptability and stability of the power supply under high-stress environments.

[0093] It solves the problem of how to quickly and accurately select a backup power source when the mains power is interrupted; by collecting and analyzing the transient response signal at the power supply output, the impedance spectrum characteristics are obtained, and the circulating current and resonance problems caused by impedance mismatch between power supplies are solved; the optimal topology and power supply connection method are determined, and the problem of power supply system stability optimization is solved.

[0094] By monitoring environmental stress distribution data in real time and constructing a stress distribution heatmap, high-stress areas are identified, and the output power curves of the power supply in these areas are adjusted based on the optimal topology, thus solving the problem of the adverse effects of environmental stress on power supply performance.

[0095] Example 2: The above example improved power supply stability by monitoring environmental stress and impedance spectrum characteristics, but it failed to classify abnormal states, which made it impossible to effectively distinguish the types of abnormalities and reduced the efficiency of abnormality handling. This example makes further improvements based on the above.

[0096] Step S3 includes: S31: Real-time monitoring is performed during the power supply cycle. If an abnormal state is detected, the abnormality category is determined. If the abnormality category belongs to the first category, the self-healing mechanism is triggered to generate a repair command, the self-healing response time is recorded, and the stability is evaluated after self-healing. If the abnormality category does not belong to the first category, the faulty power supply is isolated and the power supply path is reconstructed. The self-healing response time is the time interval from detecting the abnormality to triggering self-healing.

[0097] In power supply systems, real-time monitoring is achieved by installing various types of sensors at key nodes (such as power output terminals and important load input terminals). For example, voltage sensors measure the supply voltage in real time, current sensors measure the supply current, temperature sensors monitor equipment temperature, and frequency sensors measure the supply frequency. These sensors convert the collected analog signals into digital signals and transmit the data in real time to a central monitoring system via data transmission lines (such as wired communication lines or wireless communication modules). The central monitoring system analyzes and processes the received data in real time to achieve real-time monitoring of the power supply system's status.

[0098] Abnormal conditions include voltage abnormalities, current abnormalities, temperature abnormalities, and frequency abnormalities. The first category includes, but is not limited to, current abnormalities caused by current imbalance, local overload, and voltage fluctuations. Specifically, the current imbalance detection involves real-time monitoring of the current in each branch of the parallel circuit using a current sensor, calculating the floating value, and pre-setting a reference threshold and a current imbalance threshold. If the floating value is greater than the reference threshold, it indicates current imbalance. If the floating value is greater than the reference threshold but less than the current imbalance threshold, it is determined to be a first-category abnormality. The reference threshold is less than the current imbalance threshold.

[0099] The local overload detection is based on the rated current of the load. A current overload threshold is preset (it can also be set as the percentage of the rated current, such as 10%-20%). If the real-time current exceeds the rated current but is less than the current overload threshold, it is judged as a first-class abnormality.

[0100] The current anomaly detection caused by voltage fluctuations involves simultaneously monitoring the voltage and the current in each branch. When the voltage fluctuates within a pre-set short period and the fluctuation value is less than the baseline fluctuation threshold, it is judged as a Type I anomaly, indicating that the voltage fluctuation-induced current anomaly causes a corresponding change in current but does not pose a risk of equipment damage. In this application, the baseline threshold, current imbalance threshold, current overload threshold, and baseline fluctuation threshold are all pre-set based on historical experimental data. The short-term threshold needs to be set based on historical experimental data, specifically the critical time value that causes a corresponding change in current without posing a risk of equipment damage.

[0101] The self-healing mechanism automatically generates adjustment measures based on the determined anomaly category and the changing trends and interrelationships of voltage, current, and resistance, achieving dynamic self-healing. For self-healing of minor current imbalances, the total resistance is increased, changing the total current. Based on the current distribution principle of parallel circuits, the current in each branch is redistributed, utilizing the data correlation between voltage, current, and resistance to restore the current to a relatively balanced state. For self-healing of localized overloads, the total resistance is increased, reducing the total current. Based on the characteristics of parallel circuits, the current in each branch changes accordingly, utilizing the data relationship between voltage, current, and resistance in the circuit to restore the overloaded branch current to the normal range. For self-healing of current anomalies caused by voltage fluctuations, the total resistance is adjusted, changing the total current. Utilizing the quantitative relationship between voltage, current, and resistance, the system strives to maintain the current required by the load under voltage fluctuation conditions, achieving self-healing.

[0102] In some embodiments, the method for accurately determining self-healing includes: collecting data such as voltage and current, and analyzing their changing trends and interrelationships. The interrelationships refer to multi-level dynamic correlations, including at least physical law constraints, behavioral coupling relationships, and control logic relationships; its core is based on fundamental circuit laws, using real-time data monitoring and model prediction to quantify the causal changes and system-level impacts between circuit parameters, thereby providing a precise basis for self-healing decisions. For example, if it is found that the current in a certain branch increases while the voltage slightly decreases, and other branch currents change accordingly, calculation and simulation can determine whether adjusting the resistance value can balance the current and achieve self-healing. Alternatively, a mathematical model of the circuit can be established, real-time monitored voltage and current data can be input, and the model can be used to predict the current changes under different resistance adjustments. If the prediction results show that the current can return to the normal range after adjusting the resistance, it is determined that the self-healing method can be adopted.

[0103] Adjusting the variable resistor involves connecting it in series in a parallel circuit and adjusting its value manually or automatically. When an excessive current is detected in a branch, the variable resistor value is increased, increasing the total resistance and decreasing the total current. This redistributes the current among the branches, thus reducing the current in the overloaded branch. Switching the load is another method. If the circuit has multiple switchable loads, when an overload is detected in a branch, some of the load is switched to other branches with lighter loads. This can be achieved using switching elements such as relays, changing the connection method of the loads to redistribute the current.

[0104] Voltage, current, and resistance in a circuit are interconnected and work together to maintain circuit balance. When a local anomaly occurs, adjusting the resistance value or other operations can change the data relationships in the circuit, allowing the circuit to return to a balanced state. For example, in the case of local overload, increasing the total resistance can reduce the total current, easing the burden on the overloaded branch. At the same time, the current in other branches will also change accordingly, ultimately achieving current balance in the entire circuit.

[0105] The power supply system's status is continuously monitored over a certain period (e.g., 30 minutes to 1 hour) to observe whether various parameters are stable. Stability can be assessed by calculating the standard deviation of the parameters. For example, for voltage parameters, the standard deviation of the voltage values ​​over the monitoring period is calculated. A smaller standard deviation indicates less voltage fluctuation and better system stability.

[0106] Record all relevant information about the self-healing process, including the self-healing trigger time, self-healing response time, specific operations for reconfiguring the power supply path, and changes in system parameters before and after self-healing. Generate a detailed self-healing report for subsequent analysis and reference.

[0107] Once the power supply system is confirmed to be stable and all parameters are normal, the system will be switched back from self-healing mode to normal operation mode. For example, equipment such as reactive power compensation devices and load switching switches that were previously adjusted for self-healing will be restored to normal operation to ensure that the power supply system supplies power to the load in the optimal way.

[0108] In this embodiment, multiple sensors and a central monitoring system are used to acquire timely information on parameters such as voltage, current, temperature, and frequency of the power supply system. The system accurately identifies anomaly categories, clearly distinguishing between Category I anomalies and other anomalies based on set thresholds and detection methods. For different Category I anomalies, adjustment measures are automatically generated to effectively adjust circuit parameters and restore the circuit to a balanced state. By collecting and analyzing data and building predictive models, the feasibility of a self-healing method is determined. By adjusting variable resistors and switching loads, the data relationships in the circuit are altered, achieving current redistribution. The stability of the power supply system after self-healing is objectively evaluated by calculating the parameter standard deviation. Information related to the self-healing process is recorded to provide a basis for subsequent analysis and reference.

[0109] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. For those skilled in the art, the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A UPS (Uninterruptible Power Supply) control and management method, characterized in that, include: S1: Monitor the mains power supply status. If the power supply is interrupted, generate the initial power supply sequence. The transient response signals of each power supply output terminal in the initial power supply sequence are collected to form a time series dataset; the time series dataset is analyzed to obtain the impedance spectrum characteristics at different frequencies and generate an impedance spectrum map. S2: Based on the impedance spectrum, calculate the differences in impedance magnitude and phase angle between power supplies at the same frequency in the initial power supply sequence, and obtain several sets of difference number sequences; compare the difference number in the difference number sequence of each power supply with the difference threshold. If the difference number is greater than the difference threshold, it is marked as unqualified. Calculate the unqualified rate of each power supply and sort them in descending order; select several deviation power supplies according to the optimization mechanism and generate a deviation sequence; adjust the equipment parameters of each power supply one by one according to the deviation sequence, and update the unqualified rate and deviation sequence of each power supply after adjustment until there are no deviation power supplies or the maximum number of iterations is reached, and generate the pre-power supply sequence; The preferred mechanism includes: for the initial power supply sequence, setting the proportion of each power supply to the total number of power supplies minus one as a single failure rate, screening out power supplies with a failure rate greater than the single failure rate and arranging them in descending order, and selecting the first 75% of power supplies in sequence to form a deviation sequence. Based on the deviation sequence, the equipment parameters are adjusted one by one, including: using the impedance of the reference power supply at the critical frequency point as a benchmark, determining the target impedance of each power supply at the critical frequency point according to the impedance matching requirements; calculating the parameter adjustment amount of the digitally adjustable inductor and capacitor array according to the impedance calculation formula based on the target impedance and the current impedance characteristics of the power supply; adjusting the digitally adjustable inductor through the control circuit according to the calculated inductor parameter adjustment amount; and switching the capacitor combination in the capacitor array according to the calculated capacitor parameter adjustment amount. S3: Calculate the impedance matching degree and circulating current loss reduction rate of each power source in the pre-power supply sequence, redetermine the power supply sequence to obtain the final power supply sequence, and then supply power.

2. The UPS uninterruptible power supply control and management method as described in claim 1, characterized in that, Analyze the time series dataset to obtain the impedance spectrum characteristics at different frequencies, including: performing a fast Fourier transform on the time series dataset to convert the time-domain signal into a frequency-domain signal, obtaining the amplitude and phase information of current and voltage at different frequencies, calculating the impedance magnitude for each frequency point according to the complex form of Ohm's law, and calculating the phase angle based on the phase of voltage and current; The impedance spectrum characteristics refer to the opposition of the power source to the current at different frequencies, including the impedance magnitude and phase angle. Generating an impedance spectrum involves: establishing two coordinate systems with frequency as the abscissa and impedance magnitude and phase angle as the ordinates respectively; plotting the impedance magnitude and phase angle data corresponding to each frequency point in the corresponding coordinate systems; and connecting the data with smooth curves to form curves showing the impedance magnitude changing with frequency and the phase angle changing with frequency, which together constitute the impedance spectrum.

3. The UPS uninterruptible power supply control and management method as described in claim 1, characterized in that, Calculate the impedance matching degree and circulating current loss reduction rate of each power source in the pre-power supply sequence, including: Based on the adjusted power supply, arbitrarily select a reference power supply and calculate the relative deviations of the impedance magnitude and phase angle of other power supplies from the reference power supply at key frequency points; calculate the impedance matching degree by weighted average of the relative deviations of the impedance magnitude and phase angle of multiple power supplies at multiple key frequency points; obtain the corresponding circulating current loss according to the circulating current and the resistance of the circulating current path before and after adjustment, and obtain the circulating current loss reduction rate according to the circulating current loss before and after adjustment. The impedance matching degree is used to measure the consistency of each power supply impedance at key frequency points; the circulating current loss reduction rate is used to measure the degree of reduction in circulating current loss after impedance matching.

4. The UPS uninterruptible power supply control and management method as described in claim 1, characterized in that, The time series dataset includes transient signals corresponding to each sampling time point and arranged in chronological order. The transient response signals refer to the current and voltage data output by the power supply. The initial power supply sequence is generated by: obtaining basic information of all power sources, traversing all uninterruptible power supply devices, and checking whether the basic information meets the power supply conditions; obtaining the identifiers of the power supply devices that meet the conditions, and generating the initial power supply sequence based on the selected power source identifiers. The basic information includes the standard voltage, standard current, and available power of the power supply. The power supply identifier is a unique identifier used to identify and distinguish power supply devices. The power supply identifier includes the specific location information, usage status information, and basic device information of the power supply device.

5. The UPS uninterruptible power supply control and management method as described in claim 1, characterized in that, The method further includes: S4: obtaining impedance matching degree and power health data of the final power supply sequence, forming a power status report, modeling the multi-power supply connection relationship as a graph network, establishing an objective function to maximize power supply stability, and determining the optimal topology and power supply connection method. S5: Real-time monitoring of environmental stress distribution data, construction of stress distribution heat map, identification of high stress areas; adjustment of power output curve of power supply in high stress areas based on optimal topology.

6. The UPS uninterruptible power supply control and management method as described in claim 5, characterized in that, Modeling the multi-power supply connection relationship as a graph network includes: treating all power supplies as nodes in the graph network; representing the lines connecting two power supplies as edges in the graph network based on the actual connection lines between the power supplies; the attributes of the edges include the impedance and length parameter values ​​of the lines; and establishing an adjacency matrix and an impedance matrix based on the relationship between nodes and edges, where the elements in the impedance matrix represent the impedance values ​​of the connection lines between nodes. Establish an objective function to maximize power supply stability, which includes three indicators: minimizing voltage fluctuations, maximizing current stability, and maximizing reliability. Pre-set corresponding weight coefficients, normalize the obtained indicator values, and then use weighted summation to calculate the objective function to maximize power supply stability.

7. The UPS uninterruptible power supply control and management method as described in claim 5, characterized in that, Determining the optimal topology and power supply connection method includes: randomly generating a set of graph network topologies and power supply operating parameters as initial solutions; calculating the objective function value based on the current topology and operating parameters; using optimization algorithms to adjust the topology and operating parameters to generate new objective solutions; and outputting the current optimal topology, operating parameters, and power supply connection method until the termination condition is met.

8. The UPS uninterruptible power supply control and management method as described in claim 5, characterized in that, Stress distribution data refers to the magnitude and distribution of stress exerted on an object by various environmental factors within a specific area; Constructing a stress distribution heatmap includes: dividing the monitoring area into several grid cells, calculating the average stress value in each grid cell based on the collected stress data, using different colors to represent different stress value ranges, mapping the colors onto the corresponding grid cells, and forming a stress distribution heatmap. Adjusting the power output curve of a power source in a high-stress region based on the optimal topology includes: establishing a mathematical model of the environment and power source output power based on the characteristics and historical data of the power source; determining the adjustment strategy for the output power curve based on the analysis results; sending adjustment commands; and changing the power source output power curve.

9. The UPS uninterruptible power supply control and management method as described in claim 1, characterized in that, Step S3 includes: S31: Real-time monitoring is performed during the power supply cycle. If an abnormal state is detected, the abnormality category is determined. If the abnormality category belongs to the first category, the self-healing mechanism is triggered to generate a repair command, the self-healing response time is recorded, and the stability is evaluated after self-healing. If the abnormality category does not belong to the first category, the faulty power supply is isolated and the power supply path is reconstructed. The self-healing response time is the time interval from detecting the abnormality to triggering self-healing. The first category includes current anomalies caused by current imbalance, local overload, and voltage fluctuations; the self-healing mechanism automatically generates adjustment measures based on the determined anomaly category, combined with the changing trends and relationships of voltage, current, and resistance, to achieve dynamic self-healing.

Images