Standby power supply intelligent monitoring and switching control system and method based on Internet of Things

Abstract

The invention discloses a standby power supply intelligent monitoring and switching control system and method based on the Internet of Things, and relates to the technical field of standby power supply intelligent switching, and the method comprises the following steps: setting an optimal reference switching time threshold in combination with the historical operation data, load characteristics and technical specifications of standby power supply switching; an initial secure handover period is defined. According to the invention, the key jump from timed switching to state sensing driving switching is realized, so that the system can comprehensively sense the phase synchronization state between the main power supply and the standby power supply before switching, and the phase jump trend is dynamically predicted and intelligently judged in combination with a machine learning model; therefore, impact current, equipment damage and regional tripping risks caused by forced switching due to inconsistent phases are effectively avoided, and the safety, the accuracy and the intelligent level of the switching process of the standby power supply are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent switching of backup power supplies, and in particular to an intelligent monitoring and switching control system and method for backup power supplies based on the Internet of Things. Background Art

[0002] IoT-based intelligent backup power supply monitoring and switching is an intelligent solution that leverages IoT technologies (such as sensors, communication modules, and cloud platforms) to provide real-time monitoring and automatic control of backup power sources (such as UPS, battery packs, and generators) within power systems. This system deploys sensors for voltage, current, and temperature to collect real-time operating status data from both the primary and backup power sources. This data is uploaded to a cloud or local control platform via wireless or wired networks, enabling visual management of power status, anomaly warnings, and fault diagnosis. In the event of a primary power failure or voltage anomaly, the system automatically detects and seamlessly switches to the backup power source, ensuring continuous and stable power supply to critical equipment and systems, significantly improving the reliability, intelligence, and operational efficiency of the power supply system.

[0003] The existing technology has the following deficiencies: In existing intelligent switching technology for backup power supplies, a fixed switching time threshold is typically used to execute the switching operation. However, when the switching time is about to arrive, if there is still a phase jump between the backup power supply and the active power supply, continuing to force the switch to close according to the preset time limit is actually equivalent to instantly connecting two power supplies at different voltage phases in parallel. This instantaneous operation will trigger an inrush current that far exceeds the rated value by several times or even dozens of times, far exceeding the tolerance of the power equipment, and can easily cause damage to the circuit breaker, contactor welding, or cable overheating. More seriously, the huge current will trigger the tripping mechanism through the ground protection system, directly causing a power outage in the entire power supply area, resulting in serious operational interruption and risk of equipment damage.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present invention is to provide an intelligent monitoring and switching control system and method for a backup power supply based on the Internet of Things, which realizes a key leap from "scheduled switching" to "state-aware driven switching", enabling the system to fully perceive the phase synchronization status between the main and backup power supplies before switching, and combine machine learning models to dynamically predict and intelligently judge the phase jump trend, thereby effectively avoiding the impact current, equipment damage and regional tripping risks caused by forced switching due to phase inconsistency, significantly improving the safety, accuracy and intelligence level of the backup power supply switching process, and providing higher power operation guarantee capabilities for critical load scenarios while ensuring power supply continuity, so as to solve the problems in the above-mentioned background technology.

[0006] In order to achieve the above object, the present invention provides the following technical solution: a backup power supply intelligent monitoring and switching control method based on the Internet of Things, comprising the following steps: Combine historical operating data, load characteristics, and technical specifications of backup power supply switching to set an optimal benchmark switching time threshold and clarify the initial safe switching period; During the benchmark waiting period for the backup power supply switching execution, a high-precision synchronous sampling unit is used to perform real-time, high-frequency synchronous sampling of the voltage waveforms of the backup power supply and the active power supply, and the real-time phase angle of the main and backup power supply voltage waveforms is extracted in real time through digital signal processing technology; Based on the collected real-time phase data, key indicators that can reflect the phase jump characteristics between the main and backup power supplies are extracted. In-depth feature engineering processing is performed on the key indicators to construct a set of features that characterize the degree of phase jump. The constructed feature set is converted into a feature vector and input into a pre-trained and operational machine learning model to accurately determine and dynamically predict the real-time phase jump state and its evolution trend during the backup power supply switching period. Based on the output jump prediction results, the risk level of phase jump is evaluated in real time, the safety level of phase synchronization between the current main and standby power supplies is automatically identified, and the benchmark switching time threshold is adaptively delayed according to the risk level of phase jump.

[0007] Preferably, the real-time phase angle of the main and standby power supply voltage waveforms is extracted in real time by fast Fourier transform, and the specific steps are as follows: S1. Preprocess the sampled data and perform window function processing to reduce spectrum leakage; S2. Perform fast Fourier transform to convert the time domain signal to the frequency domain, find the amplitude peak near the detected main frequency and read its phase; S3, using the read phase as a real-time reference point to calculate the phase angle between the main power supply and the backup power supply; S4. Continuously repeat S1-S3 and periodically refresh the phase results to achieve real-time monitoring of the phase of the main and standby power supply voltage waveforms.

[0008] Preferably, signal processing technology is used to extract key indicators that can reflect the phase jump characteristics between the main and standby power supplies, wherein the extracted indicators include the cumulative value of the phase angle jump frequency and amplitude per unit time and the degree of skewness of the phase difference change. Within the monitoring window, the extracted indicators are subjected to in-depth feature engineering processing to generate phase mutation reference values and phase fluctuation asymmetry reference values respectively. A set of feature sets that characterize the degree of phase jump is constructed through the phase mutation reference values and phase fluctuation asymmetry reference values.

[0009] Preferably, the feature set constructed by the phase mutation reference value and the phase fluctuation asymmetry reference value is converted into a feature vector and input into a machine learning model that has been pre-trained and put into use. A jump coefficient is generated based on the machine learning model, and the real-time phase jump state and its evolution trend during the backup power supply switching are accurately judged and dynamically predicted through the jump coefficient.

[0010] Preferably, the safety level of the phase synchronization between the current main and standby power supplies is automatically identified, and the reference switching time threshold is adaptively delayed according to the risk level of the phase jump. The specific steps are as follows: During the real-time switching of the backup power supply, the machine learning model outputs the current jump coefficient, which is compared with the set jump coefficient reference threshold to evaluate the phase synchronization risk level between the current main and backup power supplies. A risk level function is introduced to quantify the risk. The expression of the risk level function is: in: is the jump coefficient of the model output, The jump coefficient reference threshold set by the system, The phase jump risk level coefficient reflects the intensity ratio of the current jump degree relative to the preset safety threshold; Based on the phase jump risk level coefficient Calculate an adaptive delay factor to adjust the reference switching time threshold. The calculation expression is: in: is the adaptive delay factor, which determines the adjustment multiple of the actual switching time. is the maximum delay gain coefficient, which controls the upper limit of the delay. It is the jump response sensitivity coefficient, which controls the response steepness of the delay function.

[0011] Preferably, the adaptive delay factor is applied to the original reference switching time threshold , get the adaptive delay switching threshold actually used to control the switching action, the expression is: in: is the delay switching time threshold used to control execution. is the preset reference switching time, The maximum allowed switching delay time.

[0012] Preferably, within the monitoring window, the specific steps of performing in-depth feature engineering processing on the accumulated values of the phase angle jump frequency and amplitude within a unit time to generate a phase mutation reference value are as follows: A transition energy function is constructed to capture the joint strength of frequency and amplitude; The jump energy is normalized and the sequential perturbation expansion rate is introduced to generate the phase jump reference value.

[0013] Preferably, within the monitoring window, the specific steps of performing in-depth feature engineering processing on the skewness of the phase difference change to generate a phase fluctuation asymmetry reference value are as follows: Analyze the phase difference curve of the main and standby power supplies in the monitoring window, identify all amplitude extreme points, record the phase difference value corresponding to each extreme point, and form an amplitude extreme point sequence in sequence; For all extreme points, positive extreme values and negative extreme values are distinguished, and then they are classified into positive amplitude extreme value sets. and negative amplitude extreme value set ; For the positive amplitude extreme value set and negative amplitude extreme value set Normalization processing is performed, and the normalized reference value is used to calculate the relative offset of each extreme point. Then, the phase fluctuation asymmetry reference value is calculated according to the ratio of the relative offsets.

[0014] The IoT-based intelligent monitoring and switching control system for backup power supplies includes a benchmark switching strategy configuration module, a real-time phase acquisition and analysis module, a phase jump feature construction module, a phase jump prediction and recognition module, and an adaptive switching duration control module. The benchmark switching strategy configuration module combines historical operating data, load characteristics, and technical specifications of backup power supply switching to set an optimal benchmark switching time threshold and clarify the initial safe switching period; The real-time phase acquisition and analysis module uses a high-precision synchronous sampling unit to perform real-time, high-frequency synchronous sampling of the voltage waveforms of the backup power supply and the active power supply during the benchmark waiting period for the backup power supply switching execution. It uses digital signal processing technology to extract the real-time phase angle of the main and backup power supply voltage waveforms in real time; The phase jump feature construction module extracts key indicators that reflect the phase jump characteristics between the main and backup power supplies based on the collected real-time phase data. It then performs in-depth feature engineering on these key indicators to construct a set of features that characterize the degree of phase jump. The phase jump prediction and recognition module converts the constructed feature set into a feature vector and inputs it into a pre-trained and operational machine learning model to accurately determine and dynamically predict the real-time phase jump status and evolution trend during backup power supply switching. The adaptive switching duration control module evaluates the risk level of phase jump in real time based on the output jump prediction results, automatically identifies the current safety level of phase synchronization between the main and standby power supplies, and adaptively delays the benchmark switching time threshold according to the phase jump risk level.

[0015] In the above technical solution, the technical effects and advantages provided by the present invention are: The present invention achieves a key leap from "scheduled switching" to "state-aware driven switching", enabling the system to fully perceive the phase synchronization status between the main and standby power supplies before switching, and combine machine learning models to dynamically predict and intelligently judge the phase jump trend, thereby effectively avoiding the impact current, equipment damage and regional tripping risks caused by forced switching due to phase inconsistency, significantly improving the safety, accuracy and intelligence level of the backup power supply switching process, and providing higher power operation guarantee capabilities for critical load scenarios while ensuring power supply continuity. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction to the drawings required for use in the embodiments will be given below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0017] Figure 1 This is a flow chart of the method for intelligent monitoring and switching control of backup power supply based on the Internet of Things of the present invention.

[0018] Figure 2 This is a module diagram of the backup power supply intelligent monitoring and switching control system based on the Internet of Things of the present invention.

[0019] Figure 3 It is a framework diagram of the present invention. DETAILED DESCRIPTION

[0020] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art.

[0021] The present invention provides Figure 1 The IoT-based intelligent monitoring and switching control method for backup power supply shown includes the following steps: Combine historical operating data, load characteristics, and technical specifications of backup power supply switching to set an optimal benchmark switching time threshold and clarify the initial safe switching period; Historical operating data refers to a series of key parameters and status information recorded and accumulated during the backup power system's past operation. This data typically includes the frequency and duration of primary power failures, the backup power supply startup time, the time of successful switchover, voltage and frequency trends, phase synchronization status, failed switchover cases, and device response delays. By analyzing this data, typical time points and risk factors for backup power switchover problems can be identified, providing a data basis for appropriately setting baseline switchover time thresholds, avoiding premature or delayed switching and power supply delays.

[0022] Load characteristics refer to the operating characteristics of the electrical equipment or system that the backup power supply is required to support. These include load type (e.g., resistive, inductive, capacitive), load starting current, power fluctuation range, sensitivity to voltage and frequency, and tolerance for power continuity. For example, lighting systems have relatively loose switching time requirements, while data center servers, medical equipment, or motor drive systems are extremely sensitive to power supply fluctuations and may even be unable to tolerate millisecond-level power outages. Therefore, when setting the appropriate switching time threshold, it is necessary to fully consider the load type currently connected to the system and its tolerance for switching timing to ensure stable and uninterrupted system power supply.

[0023] Technical specifications refer to technical guidance documents provided by national standards, industry standards, or equipment manufacturers regarding the configuration, operation, safety, and switching control of backup power systems. Typical specifications include ATS response time ranges, power switching safety thresholds, generator set startup time requirements, phase synchronization standards, voltage and frequency tolerances, and protection mechanism setting requirements. For example, a certain type of UPS system requires switching within 10ms, while a diesel generator may take 15 seconds to start and stabilize. Therefore, the setting of the benchmark switching time should comply with relevant technical specifications to ensure that the system design conditions and equipment performance boundaries are met, achieving optimal switching efficiency while ensuring safety.

[0024] This baseline switching time threshold can be determined based on the backup power source type (such as a diesel generator, UPS, or energy storage system) and the type of load it serves. The switching time threshold specifies how long the backup power source must wait before switching after a primary power source failure or anomaly. This ensures the backup power source enters a stable output state, providing a basic time window to ensure the backup power source achieves stable frequency, voltage, and phase. A clear initial "safe switching period" ensures stable operation of the backup power source and avoids secondary failures caused by premature startup and premature switching.

[0025] During the benchmark waiting period for the backup power supply switching execution, a high-precision synchronous sampling unit is used to perform real-time, high-frequency (2kHz to 10kHz is recommended) synchronous sampling of the voltage waveforms of the backup power supply and the active power supply. The real-time phase angle (phase information) of the active and standby power supply voltage waveforms is extracted in real time through digital signal processing technology (such as fast Fourier transform (FFT), phase-sensitive detection, or phase-locked loop (PLL) algorithm). The real-time phase angle of the main and standby power supply voltage waveforms is extracted in real time through Fast Fourier Transform (FFT). The specific steps are as follows: Preprocess the sampled data with window functions (such as Hanning window) to reduce spectrum leakage; Perform a fast Fourier transform to convert the time domain signal to the frequency domain, find the amplitude peak near the detected main frequency and read its phase; Using this phase as a real-time reference point, the phase angle between the main power supply and the backup power supply is calculated; Continuously repeat the above process and periodically refresh the phase results to achieve real-time monitoring of the phase of the main and standby power supply voltage waveforms; Through these steps, the phase relationship between the main and standby power supplies can be accurately obtained, providing key data support for evaluating phase jumps and implementing synchronous switching.

[0026] Based on the collected real-time phase data, key indicators that reflect the phase jump characteristics between the main and backup power sources are extracted. In-depth feature engineering processing is performed on these key indicators to construct a set of features that characterize the degree of phase jump. This provides accurate data support for subsequent model prediction and intelligent judgment. Signal processing technology is used to extract key indicators that can reflect the phase jump characteristics between the main and standby power supplies. The extracted indicators include the cumulative value of the phase angle jump frequency and amplitude per unit time and the skew degree of the phase difference change. Within the monitoring window, the extracted indicators are subjected to in-depth feature engineering processing to generate phase mutation reference values and phase fluctuation asymmetry reference values respectively. A set of feature sets that characterize the degree of phase jump are constructed based on the phase mutation reference values and phase fluctuation asymmetry reference values.

[0027] During the baseline waiting period for backup power supply switching, if the frequency of phase angle jumps increases per unit time and the cumulative value of the jump amplitude is large, it can usually be determined that there is a significant phase jump between the primary and backup power supplies. This indicates that the voltage waveforms of the two power supplies are frequently and drastically inconsistent within a short period of time. This may be caused by frequency instability, generator output not yet reaching steady state, control system delays, or external interference. Continuous phase jumps not only indicate that the current primary and backup power supplies are not synchronized, but may also cause inrush current and electrical risks during switching. Therefore, monitoring the frequency and amplitude of phase angle jumps per unit time can serve as a key basis for determining whether the "safe switching window" has been entered.

[0028] Within the monitoring window, the steps for performing in-depth feature engineering on the accumulated values of the phase angle jump frequency and amplitude per unit time to generate the phase mutation reference value are as follows: A transition energy function is constructed to capture the joint strength of frequency and amplitude. The expression of the transition energy function is: in: For the i The amplitude of the instantaneous change in the phase angle between the secondary main and backup power supplies is calculated by the difference between consecutive phase sampling values. is the jump amplitude enhancement coefficient (it is recommended to be set between 1.5 and 2.5), which is used to amplify the influence weight of large-amplitude phase angle jumps. is the marker factor for the transition event, defined as If the value exceeds a set threshold (such as 0.01 radians), it is 1, otherwise it is 0. N is the total number of phase change events within the monitoring window, represents the transition energy function, which is used to measure the comprehensive intensity of the phase transition between the primary and backup power supplies within the monitoring window; This step is achieved by constructing a nonlinear weighted cumulative jump function , the hopping frequency (by ) and the jump amplitude (determined by It especially amplifies the influence of large jumps and effectively avoids the interference of noise or slight disturbance on the overall characteristics.

[0029] Normalize the jump energy and introduce the sequential perturbation expansion rate to generate the phase mutation reference value. The generation expression of the phase mutation reference value is: in: is the phase mutation reference value, and They are the maximum and minimum values of the phase angle within the monitoring window, which are used to normalize the overall phase fluctuation amplitude to prevent the overall system offset from affecting the judgment result. For very small values (such as ), to prevent the denominator from being zero, is the maximum length of the continuous jump segment (that is, the maximum sequence length of the continuous jump events), is the number of transition segments (i.e., how many consecutive transition segments there are); This step will be done by switching the energy The data is normalized and further multiplied by an expansion factor that takes into account the complexity of the transition structure. This expansion factor measures the continuity and concentration of the transitions. If the phase transitions are continuous and highly concentrated for a long time, the phase mutation reference value will increase significantly. This design allows the phase mutation reference value to not only reflect the strength of the transition, but also to detect whether it poses a risk of systemic fluctuations.

[0030] The phase jump reference value indicates that within the monitoring window, the greater the performance value of the phase jump reference value generated through in-depth feature engineering of the accumulated values of the phase angle jump frequency and amplitude per unit time. As a comprehensive quantitative indicator of the phase jump behavior between the primary and backup power supplies, the phase jump reference value fuses the jump frequency and amplitude per unit time through a nonlinear weighted approach and introduces an extended analysis of jump continuity and disturbance structure. A larger phase jump reference value indicates more frequent and drastic phase angle changes between the primary and backup power supplies within the monitoring window, possibly even exhibiting a trend of continuous disturbances, indicating a severe phase mismatch in the system and a high handover risk. Conversely, a smaller phase jump reference value indicates fewer phase changes or smaller amplitudes, indicating good synchronization between the primary and backup power supplies, minimal phase jumps, and the conditions for safe handover. Therefore, the phase jump reference value serves as an important basis for intelligently determining handover timing and synchronization reliability.

[0031] When the phase difference between the primary and backup power sources shows significant skew, particularly when the phase difference curve exhibits asymmetric shifts in both positive and negative directions and increasingly deviates from the center, this typically indicates deteriorating phase synchronization between the two power sources, indicating the presence of a phase jump. For example, if the primary power source's phase consistently leads or lags behind the backup power source, and the skew increases, this indicates that the primary and backup power sources are not balanced and are experiencing significant phase jumps. This increased skew can indicate potential instability in the power system.

[0032] Within the monitoring window, the specific steps for performing in-depth feature engineering on the skewness of the phase difference change to generate a phase fluctuation asymmetry reference value are as follows: First, analyze the phase difference curve of the main and backup power supplies within the monitoring window and identify all amplitude extreme points (i.e., local maxima and minima). These extreme points reflect the peak fluctuation state of the phase difference. The phase difference value corresponding to each extreme point is recorded, forming a sequence of amplitude extreme points. Among these extreme points, positive extreme values (i.e., peak values) and negative extreme values (i.e., trough values) are distinguished, and then they are classified as positive amplitude extreme value sets. and negative amplitude extreme value set ; This step aims to identify the primary fluctuations in phase difference by extracting extreme points. The separation of positive and negative extreme points provides a clear reference for subsequent asymmetry calculations, ensuring that the analysis focuses on the actual peak behavior of the transition fluctuations.

[0033] For the positive amplitude extreme value set and negative amplitude extreme value set Normalization is performed and the normalized reference value is used to calculate the relative offset of each extreme point. Then, the phase fluctuation asymmetry reference value is calculated according to the ratio of the relative offsets. The calculation expression of the phase fluctuation asymmetry reference value is: in: Indicates the phase fluctuation asymmetry reference value, It represents the average intensity of the phase difference shifting in the positive direction, Indicates the average intensity of the phase difference shifting in the negative direction, It represents the normalized reference value, which is used to standardize the amplitude of the phase difference extreme value so that the eigenvalues extracted in different time periods, different power systems, or different load conditions are comparable and stable.

[0034] The phase fluctuation asymmetry reference value directly characterizes the relative weight of the distribution of positive and negative amplitude extremes. When the phase jump is severe, the distribution of positive and negative amplitude extremes often exhibits significant asymmetry. By calculating the phase fluctuation asymmetry reference value, we can quantify the degree of directional deviation of the fluctuation, avoid the smoothing effect of traditional statistics, and more sensitively capture the sudden change characteristics of the phase jump.

[0035] The phase fluctuation asymmetry reference value indicates that, within the monitoring window, the larger the phase fluctuation asymmetry reference value, generated through in-depth feature engineering of the phase difference skew, the more severe the phase jump between the primary and backup power supplies. The magnitude of the phase fluctuation asymmetry reference value clearly reflects the severity of the phase jump between the primary and backup power supplies. A large phase fluctuation asymmetry reference value indicates that within the monitoring window, the phase difference between the primary and backup power supplies fluctuates significantly in a specific direction (e.g., continuously in the positive or negative direction). This indicates poor synchronization between the primary and backup power supplies, severe phase jumps, and a potential risk of synchronization instability. Conversely, when the phase fluctuation asymmetry reference value is close to 1, meaning that the positive and negative fluctuations are relatively symmetrical, this indicates minimal phase fluctuation, relatively stable synchronization, and minimal phase jump. Therefore, the magnitude of the phase fluctuation asymmetry reference value provides a direct assessment of the risk of phase jump between the primary and backup power supplies.

[0036] The constructed feature set is converted into a feature vector and input into a pre-trained and operational machine learning model to accurately determine and dynamically predict the real-time phase transition state and its evolution trend during backup power supply switching, providing a reliable basis for intelligent switching decisions. The feature set constructed by the phase mutation reference value and the phase fluctuation asymmetry reference value is converted into a feature vector and input into a pre-trained and put into use machine learning model. The jump coefficient is generated based on the machine learning model, and the real-time phase jump status and its evolution trend during the backup power supply switching are accurately judged and dynamically predicted through the jump coefficient.

[0037] A machine learning model that has been pre-trained and put into use refers to a high-quality, high-precision, robust and reliable machine learning model that has undergone rigorous data collection, data cleaning, feature extraction, feature engineering, model training, verification testing and parameter tuning before it is actually deployed in the backup power switching system for real-time operation. First of all, the "pre-training" of this model means that before actual application, the historical operating data or simulation data has been fully utilized to conduct systematic and comprehensive supervised or semi-supervised training on the machine learning model, which can effectively capture and identify the various characteristics and trends of phase jumps between the main and backup power supplies. In order to obtain such an efficient model, a series of rigorous steps are required: for example, the collection of a large amount of historical data in the initial stage, including phase stability data of normal operation and data of severe phase jumps under abnormal conditions; secondly, the collected data needs to be carefully cleaned and preprocessed to remove outliers, fill in missing values, and reasonably adjust the data distribution to ensure the overall quality of the training data. Subsequently, using the pre-built A high-quality feature set consisting of phase jump reference values and phase fluctuation asymmetry reference values is further optimized through feature engineering methods to fully reveal the deep correlation between phase jumps and these features, providing a high-quality, highly discriminative data foundation for model input. After completing feature preparation, a machine learning algorithm suitable for the application scenario (such as long short-term memory networks (LSTMs) in deep learning, gradient boosting decision trees such as XGBoost, or random forest algorithms) is selected for preliminary model structure design and parameter initialization. The feature data is then input into the model, and repeated iterative training and parameter adjustment (including hyperparameter adjustment, cross-validation, and model optimization) are carried out to achieve high generalization and prediction accuracy.

[0038] This operational implementation emphasizes that the machine learning model has moved beyond the laboratory environment and has been formally deployed in a field-based intelligent backup power supply switching system, entering the real-time operational phase and becoming an integral component of the intelligent backup power supply switching control strategy. This operational implementation requires the model to possess extremely high reliability, real-time performance, and robustness. It must be able to continuously and stably perform real-time data processing and predictive analysis in a real-world production environment, rapidly outputting transition coefficients for accurate judgment and dynamic prediction of real-time phase transition states and trends. Specifically, when the model is operational, it receives real-time phase data from the primary and backup power supplies, obtained through high-frequency, high-precision synchronous sampling. Based on previously constructed features (such as phase mutation reference values and phase fluctuation asymmetry reference values), it rapidly calculates feature vectors online and then inputs these feature vectors into the model. The pre-trained model then immediately outputs transition coefficients based on the feature vectors (which can be understood as a scoring metric to quantify the severity of phase transitions). This allows for accurate classification, assessment, and trend prediction of real-time phase states within milliseconds to seconds, providing real-time guidance for subsequent decision-making and delay adjustment mechanisms within the intelligent backup power supply switching strategy. This real-time prediction capability is a crucial consideration during the commissioning phase. The model must withstand the rigors of real-world, complex operating conditions, such as voltage fluctuations, frequency disturbances, load variations, and ambient temperature fluctuations, rather than being limited to ideal experimental data and controlled conditions. After long-term real-time application, the model must also possess online self-learning and retraining capabilities, automatically or semi-automatically adjusting its weights and parameters based on new real-time operating data to continuously improve its prediction accuracy and adaptability, truly meeting the technical requirements for intelligent, real-time, dynamic switching of backup power supplies. A pre-trained and operational machine learning model not only signifies that it has undergone rigorous and thorough offline training, validation, and optimization, but also emphasizes that the model has been successfully deployed in an actual backup power supply real-time operating system. It possesses the ability to stably process field data in real time, quickly and accurately determine the degree of phase jumps and dynamic trends, and generate jump coefficients in real time, truly meeting the demanding model reliability, real-time, and accuracy requirements of intelligent backup power supply switching control strategies in industrial applications.

[0039] The machine learning model is not specifically limited here, and can achieve the phase mutation reference value and phase fluctuation asymmetry reference value Perform comprehensive analysis to generate jump coefficients In order to realize the technical solution of the present invention, the present invention provides a specific implementation method; the jump coefficient The generated expression is: , where 、 Phase mutation reference values and phase fluctuation asymmetry reference value The preset scaling factor of 、 The preset proportional coefficient refers to the parameter used to adjust the weight of the influence of different indicators on the final result in the calculation formula of the jump coefficient, that is, and .in, is the weight coefficient of the phase mutation reference value, is the weight coefficient of the phase fluctuation asymmetry reference value. By presetting the values of these two coefficients, the contribution ratio of these two indicators in the comprehensive jump coefficient can be manually adjusted according to specific scenarios or experience. For example, if the system is more concerned about the risk brought by instantaneous mutation, Set it to a larger value; if the system pays more attention to the abnormality of the fluctuation trend, it can be increased This "preset" refers to a set of constants set based on actual needs, historical data statistics, or engineering experience before machine learning model training or actual deployment. It is used to keep the model structure simple and controllable while providing a certain degree of flexibility and interpretability.

[0040] It can be seen from the jump coefficient that within the monitoring window, the greater the performance value of the phase mutation reference value generated by in-depth feature engineering processing of the accumulated values of the phase angle jump frequency and amplitude per unit time, the greater the performance value of the phase fluctuation asymmetry reference value generated after in-depth feature engineering processing of the skewness of the phase difference change, that is, the greater the performance value of the jump coefficient generated when the real-time phase jump state during the backup power supply switching is dynamically predicted by the machine learning model, the more serious the phase jump between the main and backup power supplies is, and vice versa, the less serious the phase jump between the main and backup power supplies is.

[0041] Based on the output jump prediction results, the system evaluates the phase jump risk in real time, automatically identifies the current phase synchronization safety level between the primary and backup power sources, and adaptively delays the reference switching time threshold based on the phase jump risk level to ensure that the final switching is completed when the phase is stable. Automatically identify the current safety level of phase synchronization between the primary and backup power supplies and adaptively delay the reference switching time threshold based on the phase jump risk level. The specific steps are as follows: During the real-time switching of the backup power supply, the machine learning model outputs the current jump coefficient, which is compared with the set jump coefficient reference threshold to evaluate the phase synchronization risk level between the current main and backup power supplies. A risk level function is introduced to quantify the risk. The expression of the risk level function is: in: is the jump coefficient of the model output. The larger the value, the more serious the phase jump. The jump coefficient reference threshold set for the system is usually set by historical data experience or model tuning. The phase jump risk level coefficient reflects the intensity ratio of the current jump degree relative to the preset safety threshold; This step is used to standardize the jump degree and convert it into a universal phase jump risk level coefficient. , which facilitates adaptive adjustment of switching strategies in different scenarios.

[0042] Based on the phase jump risk level coefficient Calculate an adaptive delay factor to adjust the reference switching time threshold. The adaptive delay factor is calculated based on the exponential smoothing delay function. The calculation expression is: in: is the adaptive delay factor, which determines the adjustment multiple of the actual switching time. is the maximum delay gain coefficient, which controls the upper limit of the delay. is the jump response sensitivity coefficient, which controls the response steepness of the delay function; This step controls the delay amplitude through exponential smoothing, allowing the system to maintain high efficiency when the risk is low, and automatically slowing down the switching rhythm when the jump is severe, thereby improving robustness and system stability.

[0043] Apply the adaptive delay factor to the original reference switching time threshold , get the adaptive delay switching threshold actually used to control the switching action, the expression is: in: is the delay switching time threshold used to control execution. The system preset benchmark switching time (such as 10 seconds), It is the maximum allowable switching delay time, used to prevent excessive delay from affecting power supply.

[0044] A maximum switching time limit is set. If phase synchronization cannot be achieved within the maximum delay threshold, an alarm system is automatically triggered. This alarm is quickly pushed to operations and maintenance personnel, managers, and engineers through various channels, including the human-machine interface, monitoring platform, SMS alarms, and voice prompts. The alarm details the severity of the phase jump and recommends appropriate action (such as manual switching, emergency disconnection of the backup power supply, and generator inspection). Timely human intervention prevents serious incidents from escalating and ensures safe and stable system operation. This provides a critical safety fallback mechanism, ensuring that if automatic control cannot handle the phase jump risk, human intervention is swift, minimizing system operational risks.

[0045] This IoT-based intelligent monitoring and switching control method for backup power supplies achieves a critical leap from "scheduled switching" to "state-aware, driven switching." This enables the system to fully perceive the phase synchronization status between the primary and backup power sources before switching. Combined with machine learning models, it dynamically predicts and intelligently judges phase jump trends, effectively avoiding the risks of inrush current, equipment damage, and regional tripping caused by forced switching due to phase inconsistencies. This method significantly improves the safety, accuracy, and intelligence of the backup power switching process, ensuring power continuity while providing enhanced power operation assurance for critical load scenarios (such as hospitals, data centers, and industrial production lines).

[0046] The present invention provides Figure 2 The IoT-based backup power supply intelligent monitoring and switching control system shown includes a benchmark switching strategy configuration module, a real-time phase acquisition and analysis module, a phase jump feature construction module, a phase jump prediction and identification module, and an adaptive switching duration control module; The benchmark switching strategy configuration module combines historical operating data, load characteristics, and technical specifications of backup power supply switching to set an optimal benchmark switching time threshold and clarify the initial safe switching period; The real-time phase acquisition and analysis module uses a high-precision synchronous sampling unit to perform real-time, high-frequency synchronous sampling of the voltage waveforms of the backup power supply and the active power supply during the benchmark waiting period for the backup power supply switching execution. It uses digital signal processing technology to extract the real-time phase angle of the main and backup power supply voltage waveforms in real time; The phase jump feature construction module extracts key indicators that reflect the phase jump characteristics between the main and backup power supplies based on the collected real-time phase data. It then performs in-depth feature engineering on these key indicators to construct a set of features that characterize the degree of phase jump. The phase jump prediction and recognition module converts the constructed feature set into a feature vector and inputs it into a pre-trained and operational machine learning model to accurately determine and dynamically predict the real-time phase jump status and evolution trend during backup power supply switching. The adaptive switching duration control module evaluates the risk level of phase jump in real time based on the output jump prediction results, automatically identifies the current safety level of phase synchronization between the main and standby power supplies, and adaptively delays the benchmark switching time threshold according to the phase jump risk level.

[0047] The IoT-based intelligent monitoring and switching control method for backup power supply provided in an embodiment of the present invention is realized by the above-mentioned IoT-based intelligent monitoring and switching control system for backup power supply. The specific methods and processes of the IoT-based intelligent monitoring and switching control system for backup power supply are detailed in the embodiment of the IoT-based intelligent monitoring and switching control method for backup power supply, which will not be repeated here.

[0048] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0049] The above description is merely illustrative of certain exemplary embodiments of the present invention. It goes without saying that those skilled in the art will be able to modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the scope of protection of the claims.

[0050] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. The intelligent monitoring and switching control method of backup power supply based on the Internet of Things is characterized by: The following steps are involved: Combine historical operating data, load characteristics, and technical specifications of backup power supply switching to set an optimal benchmark switching time threshold and clarify the initial safe switching period; During the benchmark waiting period for the backup power supply switching execution, a high-precision synchronous sampling unit is used to perform real-time, high-frequency synchronous sampling of the voltage waveforms of the backup power supply and the active power supply, and the real-time phase angle of the main and backup power supply voltage waveforms is extracted in real time through digital signal processing technology; Based on the collected real-time phase data, key indicators that can reflect the phase jump characteristics between the main and backup power supplies are extracted. In-depth feature engineering processing is performed on the key indicators to construct a set of features that characterize the degree of phase jump. The constructed feature set is converted into a feature vector and input into a pre-trained and operational machine learning model to accurately determine and dynamically predict the real-time phase jump state and its evolution trend during the backup power supply switching period. Based on the output jump prediction results, the risk level of phase jump is evaluated in real time, the safety level of phase synchronization between the current main and standby power supplies is automatically identified, and the benchmark switching time threshold is adaptively delayed according to the risk level of phase jump.

2. The method for intelligent monitoring and switching control of backup power supply based on Internet of Things according to claim 1 is characterized in that: The real-time phase angle of the main and standby power supply voltage waveforms is extracted in real time through fast Fourier transform. The specific steps are as follows: S1. Preprocess the sampled data and perform window function processing to reduce spectrum leakage; S2. Perform fast Fourier transform to convert the time domain signal to the frequency domain, find the amplitude peak near the detected main frequency and read its phase; S3, using the read phase as a real-time reference point to calculate the phase angle between the main power supply and the backup power supply; S4. Continuously repeat S1-S3 and periodically refresh the phase results to achieve real-time monitoring of the phase of the main and standby power supply voltage waveforms.

3. The method for intelligent monitoring and switching control of backup power supply based on Internet of Things according to claim 1, characterized in that: Signal processing technology is used to extract key indicators that can reflect the phase jump characteristics between the main and standby power supplies. The extracted indicators include the cumulative value of the phase angle jump frequency and amplitude per unit time and the skew degree of the phase difference change. Within the monitoring window, the extracted indicators are subjected to in-depth feature engineering processing to generate phase mutation reference values and phase fluctuation asymmetry reference values respectively. A set of feature sets that characterize the degree of phase jump are constructed based on the phase mutation reference values and phase fluctuation asymmetry reference values.

4. The method for intelligent monitoring and switching control of backup power supply based on Internet of Things according to claim 3 is characterized in that: The feature set constructed by the phase mutation reference value and the phase fluctuation asymmetry reference value is converted into a feature vector and input into a pre-trained and put into use machine learning model. The jump coefficient is generated based on the machine learning model, and the real-time phase jump status and its evolution trend during the backup power supply switching are accurately judged and dynamically predicted through the jump coefficient.

5. The method for intelligent monitoring and switching control of backup power supply based on Internet of Things according to claim 4 is characterized in that: Automatically identify the current safety level of phase synchronization between the primary and backup power supplies and adaptively delay the reference switching time threshold based on the phase jump risk level. The specific steps are as follows: During the real-time switching of the backup power supply, the machine learning model outputs the current jump coefficient, which is compared with the set jump coefficient reference threshold to evaluate the phase synchronization risk level between the current main and backup power supplies. A risk level function is introduced to quantify the risk. The expression of the risk level function is: in: is the jump coefficient of the model output, The jump coefficient reference threshold set by the system, The phase jump risk level coefficient reflects the intensity ratio of the current jump degree relative to the preset safety threshold; Based on the phase jump risk level coefficient Calculate an adaptive delay factor to adjust the reference switching time threshold. The calculation expression is: in: is the adaptive delay factor, which determines the adjustment multiple of the actual switching time. is the maximum delay gain coefficient, which controls the upper limit of the delay. It is the jump response sensitivity coefficient, which controls the response steepness of the delay function.

6. The method for intelligent monitoring and switching control of backup power supply based on Internet of Things according to claim 5, characterized in that: Apply the adaptive delay factor to the original reference switching time threshold , get the adaptive delay switching threshold actually used to control the switching action, the expression is: in: is the delay switching time threshold used to control execution. is the preset reference switching time, The maximum allowed switching delay time.

7. The method for intelligent monitoring and switching control of backup power supply based on Internet of Things according to claim 3, characterized in that: Within the monitoring window, the steps for performing in-depth feature engineering on the accumulated values of the phase angle jump frequency and amplitude per unit time to generate the phase mutation reference value are as follows: A transition energy function is constructed to capture the joint strength of frequency and amplitude; The jump energy is normalized and the sequential perturbation expansion rate is introduced to generate the phase jump reference value.

8. The method for intelligent monitoring and switching control of backup power supply based on Internet of Things according to claim 3, characterized in that: Within the monitoring window, the specific steps for performing in-depth feature engineering on the skewness of the phase difference change to generate a phase fluctuation asymmetry reference value are as follows: Analyze the phase difference curve of the main and standby power supplies in the monitoring window, identify all amplitude extreme points, record the phase difference value corresponding to each extreme point, and form an amplitude extreme point sequence in sequence; For all extreme points, positive extreme values and negative extreme values are distinguished, and then they are classified into positive amplitude extreme value sets. and negative amplitude extreme value set ; For the positive amplitude extreme value set and negative amplitude extreme value set Normalization processing is performed, and the normalized reference value is used to calculate the relative offset of each extreme point. Then, the phase fluctuation asymmetry reference value is calculated according to the ratio of the relative offsets.

9. An intelligent monitoring and switching control system for backup power supply based on the Internet of Things, used to implement the intelligent monitoring and switching control method for backup power supply based on the Internet of Things according to any one of claims 1 to 8, characterized in that: It includes a benchmark switching strategy configuration module, a real-time phase acquisition and analysis module, a phase jump feature construction module, a phase jump prediction and identification module, and an adaptive switching duration control module; The benchmark switching strategy configuration module combines historical operating data, load characteristics, and technical specifications of backup power supply switching to set an optimal benchmark switching time threshold and clarify the initial safe switching period; The real-time phase acquisition and analysis module uses a high-precision synchronous sampling unit to perform real-time, high-frequency synchronous sampling of the voltage waveforms of the backup power supply and the active power supply during the benchmark waiting period for the backup power supply switching execution. It uses digital signal processing technology to extract the real-time phase angle of the main and backup power supply voltage waveforms in real time; The phase jump feature construction module extracts key indicators that reflect the phase jump characteristics between the main and backup power supplies based on the collected real-time phase data. It then performs in-depth feature engineering on these key indicators to construct a set of features that characterize the degree of phase jump. The phase jump prediction and recognition module converts the constructed feature set into a feature vector and inputs it into a pre-trained and operational machine learning model to accurately determine and dynamically predict the real-time phase jump status and evolution trend during backup power supply switching. The adaptive switching duration control module evaluates the risk level of phase jump in real time based on the output jump prediction results, automatically identifies the current safety level of phase synchronization between the main and standby power supplies, and adaptively delays the benchmark switching time threshold according to the phase jump risk level.

Images