An internet-based charging line overload anomaly intelligent early warning method and system

Abstract

The application relates to the technical field of intelligent early warning of charging line overload abnormalities, in particular to an internet-based intelligent early warning method and system for charging line overload abnormalities, which comprises the following steps: collecting power supply voltage information; identifying and quantifying characteristic parameters reflecting the quality of electric energy according to the power supply voltage information; accumulating and updating a cumulative index indicating potential damage to connected equipment according to the degree and duration of the characteristic parameters; generating early warning information when the cumulative index reaches a preset risk threshold; sending the early warning information to a remote service platform through a wireless communicator; and pushing early warning notifications to user terminals after the remote service platform receives the early warning information. The application can effectively identify and quantify cumulative risks, and improve the comprehensiveness and accuracy of early warning.

Description

Technical Field

[0001] This invention relates to the technical field of intelligent early warning for charging cable overload anomalies, and specifically to an internet-based intelligent early warning method and system for charging cable overload anomalies. Background Technology

[0002] In today's world of ubiquitous electronic devices, charging cables serve as a crucial bridge connecting chargers and electrical appliances, and their safety performance directly impacts user experience and personal and property safety. Traditional charging cable safety warning systems typically assess overload risk by monitoring charging current and cable temperature. This method provides effective protection against immediate and obvious overloads caused by charger power mismatch or internal short circuits in the electrical device. However, with technological advancements and changes in user habits, more subtle and cumulative risks are emerging. These risks are often difficult for existing systems to detect, causing long-term damage to terminal devices and even triggering safety accidents. Summary of the Invention

[0003] The purpose of this invention is to address the aforementioned shortcomings by proposing an intelligent early warning method and system for overload anomalies in charging cables based on the Internet.

[0004] The present invention adopts the following technical solution: An intelligent early warning method for overload anomalies in charging cables based on the Internet, the method includes the following steps: Collect power supply voltage information; Based on the power supply voltage information, identify and quantify characteristic parameters that reflect power quality. Among them, the characteristic parameters include at least one of the following: voltage waveform fluctuation amplitude characteristics, voltage ripple characteristics, instantaneous voltage spike characteristics, voltage drop characteristics, and high-frequency noise energy characteristics. Based on the degree and duration of the characteristic parameters, a cumulative index indicating potential damage to the connected equipment is accumulated and updated. The degree of the characteristic parameter is the abnormal intensity of the characteristic parameter relative to a preset reference level, and the duration of the characteristic parameter is the continuous existence of the abnormal state in the time dimension. The cumulative index is used to quantify the degree of risk posed by power quality anomalies to the connected equipment. When the cumulative indicators reach the preset risk threshold, an early warning message is generated. The generation of the early warning message is independent of the monitoring results of charging current information and data line temperature information. The warning information is sent to the remote service platform via a wireless communication device; After receiving the early warning information through the remote service platform, the early warning notification is pushed to the user terminal.

[0005] Through this technical solution, this application can achieve intelligent early warning of charging cable overload anomalies by monitoring power quality characteristic parameters, independent of traditional current and temperature monitoring, effectively identifying and quantifying cumulative risks, thereby solving the problem of difficulty in detecting hidden risks in existing technologies and improving the comprehensiveness and accuracy of early warning.

[0006] This application also discloses an internet-based intelligent early warning system for charging cable overload anomalies, applied to the aforementioned internet-based intelligent early warning method for charging cable overload anomalies. The system includes: The data acquisition module collects power supply voltage information. The identification module identifies and quantifies characteristic parameters reflecting power quality based on the power supply voltage information. Among them, the characteristic parameters include at least one of the following: voltage waveform fluctuation amplitude characteristics, voltage ripple characteristics, instantaneous voltage spike characteristics, voltage drop characteristics, and high-frequency noise energy characteristics. The update module accumulates and updates the cumulative index indicating potential damage to the connected device based on the degree and duration of the characteristic parameter. The degree of the characteristic parameter is the abnormal intensity of the characteristic parameter relative to the preset reference level, and the duration of the characteristic parameter is the continuous existence duration of the abnormal state in the time dimension. The generation module generates an early warning message when the cumulative indicators reach a preset risk threshold. The generation of the early warning message is independent of the monitoring results of charging current information and data line temperature information. The communication module sends the early warning information to the remote service platform via a wireless communication device; The remote service module receives early warning information through the remote service platform and then pushes early warning notifications to user terminals.

[0007] This application provides a system for implementing the aforementioned early warning method. Through modular design, it can efficiently complete the collection of power supply voltage information, the identification and quantification of power quality characteristic parameters, the updating of cumulative indicators, the generation and transmission of early warning information, and the push of early warning notifications, thereby providing users with a comprehensive intelligent early warning service for charging cable overload anomalies.

[0008] This application can effectively identify and quantify the long-term damage risk to equipment caused by abnormal power quality, and realize intelligent early warning of charging cable overload abnormalities, thereby avoiding equipment damage or even safety accidents caused by hidden risks, and significantly improving charging safety and user experience.

[0009] To further understand the features and technical content of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are for reference and illustration only and are not intended to limit the present invention. Attached Figure Description

[0010] Figure 1 This is a flowchart of an intelligent early warning method for overload anomalies of charging cables based on the Internet, according to the present invention. Figure 2 This is a schematic diagram of the structure of an Internet-based intelligent early warning system for charging cable overload abnormalities according to the present invention. Detailed Implementation

[0011] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can understand the advantages and effects of the present invention from the content disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and various details in this specification can also be modified and changed based on different viewpoints and applications without departing from the spirit of the present invention. Furthermore, the accompanying drawings of the present invention are for simple illustrative purposes only and are not depictions of actual dimensions; this is stated in advance. The following embodiments will further describe the relevant technical content of the present invention in detail, but the disclosed content is not intended to limit the scope of protection of the present invention.

[0012] This embodiment provides an internet-based intelligent early warning method and system for charging cable overload anomalies, combined with... Figure 1 and Figure 2 As shown.

[0013] refer to Figure 1 An intelligent early warning method for overload anomalies in charging cables based on the Internet, the method includes the following steps: Collect power supply voltage information; Based on the power supply voltage information, identify and quantify characteristic parameters that reflect power quality. Among them, the characteristic parameters include at least one of the following: voltage waveform fluctuation amplitude characteristics, voltage ripple characteristics, instantaneous voltage spike characteristics, voltage drop characteristics, and high-frequency noise energy characteristics. Based on the degree and duration of the characteristic parameters, a cumulative index indicating potential damage to the connected equipment is accumulated and updated. The degree of the characteristic parameter is the abnormal intensity of the characteristic parameter relative to a preset reference level, and the duration of the characteristic parameter is the continuous existence of the abnormal state in the time dimension. The cumulative index is used to quantify the degree of risk posed by power quality anomalies to the connected equipment. When the cumulative indicators reach the preset risk threshold, an early warning message is generated. The generation of the early warning message is independent of the monitoring results of charging current information and data line temperature information. The warning information is sent to the remote service platform via a wireless communication device; After receiving the early warning information through the remote service platform, the early warning notification is pushed to the user terminal.

[0014] This application identifies and quantifies various characteristic parameters of power quality through in-depth analysis of power supply voltage information, and cumulatively assesses potential damage based on the degree and duration of these parameters. This enables the more comprehensive and earlier detection of hidden risks that are difficult to detect by traditional methods, effectively avoiding long-term damage to connected devices and improving charging safety.

[0015] Among them, "power supply voltage information" refers to real-time voltage data obtained from the charging cable or charging interface, which reflects the input of electrical energy during the charging process.

[0016] "Characteristic parameters" are quantitative indicators extracted from supply voltage information that reflect abnormal power quality, such as voltage waveform fluctuation amplitude characteristics, voltage ripple characteristics, instantaneous voltage spike characteristics, voltage drop characteristics, and high-frequency noise energy characteristics. The degree and duration of these abnormal parameters are key to assessing potential damage.

[0017] "Degree of characteristic parameter" refers to the abnormal intensity of the characteristic parameter deviating from the normal preset reference level, such as the magnitude of voltage spikes.

[0018] "Duration of characteristic parameters" refers to the length of time that a certain abnormal state exists continuously in the time dimension, such as how long a voltage drop lasts.

[0019] The "cumulative index" is a dynamically updated value used to quantify the degree of risk posed by power quality anomalies to connected equipment. It comprehensively considers the severity and duration of characteristic parameters, reflecting long-term, cumulative damage risks.

[0020] The "preset risk threshold" is a pre-defined critical value. When the cumulative indicators reach or exceed this threshold, it is considered that there is a high risk and an early warning needs to be triggered.

[0021] "Warning information" is a notification generated by the system when it detects a potential risk. Its generation process is independent of the monitoring results of charging current information and data line temperature information, which enables this method to discover risks that traditional methods cannot identify.

[0022] A wireless communication device is a hardware module used to enable wireless data transmission between a device and a remote service platform.

[0023] The "Remote Service Platform" is an internet-based backend system responsible for receiving early warning information, processing data, and pushing early warning notifications to user terminals.

[0024] "User terminal" refers to a smartphone, tablet, or other device that can receive notifications used by a user.

[0025] The core of the method proposed in this application lies in achieving intelligent early warning of charging cable overload anomalies through refined analysis of power supply voltage information.

[0026] There are several methods for collecting power supply voltage information. For example, a miniature voltage sensor can be integrated inside the charging cable to monitor the voltage values ​​at both ends of the cable in real time, and the collected analog signals can be converted into digital signals via an analog-to-digital converter (ADC). Another method is to set up a voltage detection circuit at the charging interface of the charger or the device, and acquire voltage data through a voltage divider resistor network or a Hall effect sensor. This collected voltage information can be continuously collected at a fixed sampling rate to form voltage time series data.

[0027] In identifying and quantifying characteristic parameters that reflect power quality, different characteristic parameters can be extracted using signal processing techniques based on the collected power supply voltage information.

[0028] For example, the fluctuation amplitude characteristics of a voltage waveform can be quantified by calculating the peak value, standard deviation, or root mean square value of the voltage signal within a certain time window. When the voltage fluctuation amplitude exceeds the normal range, it is considered to be abnormal.

[0029] For voltage ripple characteristics, Fourier transform or wavelet analysis can be performed on the supply voltage information to identify and quantify ripple components within a specific frequency range. For example, the severity of voltage ripple can be assessed by calculating the energy proportion at the power frequency (such as 50Hz or 60Hz) and its harmonic frequencies.

[0030] For transient voltage spikes, detection can be achieved by setting voltage and duration thresholds. A transient voltage spike is identified when the voltage value suddenly increases and exceeds the preset threshold within a very short time (e.g., on the order of microseconds). Its quantification can include the amplitude and duration of the spike.

[0031] For voltage drop characteristics, it is possible to monitor situations where the voltage value suddenly drops below a preset threshold within a certain time period (e.g., milliseconds). Its quantification can include the depth and duration of the drop.

[0032] To identify the energy characteristics of high-frequency noise, a high-pass filter can be applied to the supply voltage information to extract the high-frequency noise components. Then, the energy or power spectral density of these high-frequency noise components can be calculated. For example, the energy distribution in the frequency range above 1 kHz can be analyzed to quantify the intensity of the high-frequency noise.

[0033] In accumulating and updating cumulative indicators that indicate potential damage to connected devices based on the degree and duration of characteristic parameters, methods based on weighted integrals or state machine models can be used.

[0034] For example, an anomaly intensity function can be defined for each characteristic parameter, mapping the actual value of the characteristic parameter to a numerical value representing the degree of anomaly. Simultaneously, a time weighting function can be defined for each characteristic parameter, assigning different weights based on the duration of the anomaly. The cumulative index can be updated by multiplying the anomaly degree of each characteristic parameter by its duration weight and then summing the results. For example, when a voltage spike is detected, the larger its amplitude and the longer its duration, the greater its contribution to the cumulative index. The cumulative index can be a dimensionless numerical value or a quantity with specific physical meaning, such as an "equipment fatigue index." The index can be updated continuously or periodically, for example, once per second.

[0035] When a cumulative indicator reaches a preset risk threshold, the system continuously monitors the cumulative indicator to generate an early warning message. Once the indicator exceeds the preset risk threshold, the system will immediately generate an early warning message. The early warning message may include the anomaly type, the time of occurrence, the current value of the cumulative indicator, and the possible risk level. It is worth noting that the generation of the early warning message is independent of the monitoring results of charging current information and data line temperature information. This means that even if the charging current and data line temperature are within the normal range, the system will still issue an early warning as long as the cumulative damage to power quality reaches the threshold. For example, a persistent, small voltage ripple, although it may not immediately cause overcurrent or overheating, may cause chronic damage to the delicate circuitry inside the device due to its cumulative effect; in this case, the system can provide a timely warning.

[0036] The warning information is then sent to a remote service platform via a wireless communication device. The generated warning information is transmitted through the device's built-in wireless communication device (such as a Wi-Fi module, Bluetooth module, or cellular network module). The wireless communication device encapsulates the warning information into data packets and transmits them to the preset remote service platform via the wireless network. For example, the device can send the warning information to a cloud server via the MQTT or HTTP protocol.

[0037] After receiving early warning information through a remote service platform, the platform analyzes and processes the information before sending it to user terminals. The platform can determine the type and severity of the warning information to send to the user terminal. For example, a minor risk might trigger a standard notification, while a serious risk might trigger a high-priority notification with a suggestion to check the charging device or replace the charging cable. User terminals (such as smartphones) receive the warning notifications through installed applications or system notification services, thus gaining timely awareness of potential risks associated with the charging cable.

[0038] The intelligent early warning method for overload anomalies in charging cables proposed in this application overcomes the limitations of traditional early warning methods that rely solely on charging current and cable temperature through comprehensive and in-depth analysis of power supply voltage information. First, the system continuously collects power supply voltage information, which is fundamental to reflecting the quality of power input. Next, the system utilizes advanced signal processing technology to identify and quantify various characteristic parameters of power quality from this voltage information, such as voltage waveform fluctuation amplitude, voltage ripple, instantaneous voltage spikes, voltage drops, and high-frequency noise energy. These characteristic parameters can reveal hidden power anomalies that are difficult to detect using traditional methods. Subsequently, based on the degree and duration of these characteristic parameters' anomalies, the system dynamically accumulates and updates a cumulative index indicating potential damage to connected equipment. This index quantifies the long-term, chronic risks caused by power quality anomalies to equipment, rather than just instantaneous overload. When the cumulative index reaches a preset risk threshold, the system immediately generates an early warning message, independent of the monitoring results of charging current and data cable temperature. This means that even when the current and temperature appear normal, if the power quality continues to deteriorate to the point of damaging the equipment, the system can issue a timely warning. Finally, the generated warning information is sent to the remote service platform via a wireless communication device. After receiving the information, the platform further pushes warning notifications to the user terminal to ensure that the user can be aware of potential risks in a timely manner and take corresponding measures, thereby effectively avoiding damage to the equipment due to long-term power quality problems and improving charging safety.

[0039] This application further proposes a step of accumulating and updating a cumulative indicator that indicates potential damage to the connected device based on the degree and duration of the characteristic parameters, including: Generate voltage perturbation signals; Apply a voltage perturbation signal to the supply voltage; Collect instantaneous change data of power supply voltage and charging current; Based on instantaneous change data, the compensation capability characteristics are quantified. The compensation capability characteristics are characteristic quantities used to characterize the ability of the internal power supply of the connected equipment to suppress, absorb and recover voltage micro-disturbances. Based on the characteristics of the compensation capability, a fatigue correction factor is calculated, whereby the fatigue correction factor is a correction parameter used to characterize the degree of attenuation of the internal power compensation capability of the connected equipment. Based on the fatigue correction factor, the degree of characteristic parameters, and the duration, cumulative indicators indicating potential damage to connected equipment are accumulated and updated.

[0040] Specifically, generating a voltage micro-perturbation signal refers to injecting a small, controllable voltage disturbance into the supply voltage without affecting the normal operation of the equipment. This disturbance signal can have a specific frequency, amplitude, and waveform; for example, it can be a high-frequency pulse, a low-amplitude sine wave, or a step signal. Its purpose is to actively detect the response characteristics of the internal power supply of the connected equipment to changes in external voltage. Applying the voltage micro-perturbation signal to the supply voltage can be understood as using a specific injection circuit or modulation technique to superimpose the generated micro-perturbation signal onto the normal supply voltage, allowing it to be input to the connected equipment along with the supply voltage. This step aims to simulate minor power quality fluctuations that may occur in actual use and to observe the immediate response of the equipment.

[0041] In practical applications, collecting instantaneous change data of supply voltage and charging current refers to using high-precision, high-sampling-rate sensors and data acquisition systems to monitor and record the rapid changes in supply voltage and charging current in real time after the application of micro-disturbance signals. This instantaneous data contains detailed information about the suppression, absorption, and recovery processes of micro-disturbances by the internal power supply of the equipment, forming the basis for quantifying the equipment's compensation capability. Furthermore, quantifying compensation capability characteristics based on instantaneous change data involves analyzing the collected instantaneous voltage and current data to extract characteristic quantities that reflect the equipment's internal power supply's ability to suppress, absorb, and recover from voltage micro-disturbances. For example, indicators such as recovery time after voltage dips, current fluctuation amplitude, and energy absorption efficiency can be analyzed. These indicators collectively constitute the equipment's compensation capability characteristics, aiming to objectively assess the equipment's inherent resistance to power quality anomalies.

[0042] Furthermore, based on the compensation capability characteristics, a fatigue correction factor is calculated. This involves calculating a correction parameter based on historical operating data of the equipment, the changing trends of the compensation capability characteristics, and a pre-defined attenuation model. This fatigue correction factor characterizes the degree of attenuation of the internal power compensation capability of the connected equipment. For example, after prolonged high load or frequent power quality anomalies, the equipment's compensation capability may decline. The fatigue correction factor quantifies this decline, aiming to reflect the dynamic changes in the equipment's "health" status and making the assessment of cumulative indicators more timely and accurate. Finally, based on the fatigue correction factor, the degree and duration of the characteristic parameters, cumulative indicators indicating potential damage to the connected equipment are accumulated and updated. This combines traditional accumulation methods based on the degree and duration of characteristic parameters with the equipment's current compensation capability characteristics and the fatigue correction factor. This means that even if the degree and duration of the characteristic parameters are the same, if the equipment has a strong compensation capability and a low fatigue level, the growth of the cumulative indicator will be relatively slow; conversely, if the equipment's compensation capability declines and its fatigue level is high, the growth of the cumulative indicator will accelerate, thus more accurately reflecting the risks faced by the equipment.

[0043] This application's solution actively generates and applies voltage micro-perturbation signals, enabling real-time and dynamic detection of the internal power supply response mechanism of connected equipment to power quality anomalies. By collecting instantaneous change data of supply voltage and charging current, the specific performance of the equipment in the face of external disturbances can be obtained, thereby quantifying its compensation capability characteristics. This combination of active detection and real-time response makes the assessment of the equipment's inherent resilience no longer static, but dynamic and targeted. Furthermore, by introducing a fatigue correction factor, this application's solution can take into account the attenuation of the equipment's compensation capability during long-term operation, avoiding the bias that may result from assessments based solely on instantaneous power quality parameters. It is precisely because of the comprehensive consideration of the equipment's internal response mechanism and long-term fatigue state that the update of cumulative indicators can more accurately reflect the actual potential damage risk of the equipment, thus effectively compensating for the limitations of assessments relying solely on external power quality parameters.

[0044] In some preferred embodiments, it is assumed that a smartphone is being charged via a charging cable. To more accurately assess the potential damage to the phone caused by abnormal charging cable overload, the system first generates a voltage micro-perturbation signal with a frequency of 1kHz and an amplitude of 50mV, and superimposes it onto the normal 5V supply voltage. Subsequently, the system collects real-time data on the instantaneous changes in the supply voltage and charging current entering the phone at a sampling rate of 100kHz. By analyzing this instantaneous data, for example, if the voltage recovers to more than 95% within 10 microseconds after a drop, and the charging current fluctuation is less than 5mA, a high compensation capability characteristic can be quantified. Simultaneously, the system calculates a fatigue correction factor based on the phone's historical charging records and past power quality anomalies. For example, if the phone has been used continuously for two years and has experienced multiple voltage spikes, the fatigue correction factor might be 1.2, indicating that its compensation capability has decreased by 20%. Finally, when accumulating and updating cumulative indicators, in addition to considering the degree and duration of the current voltage ripple, the aforementioned quantified compensation capability characteristics and fatigue correction factor are also incorporated into the calculation. For example, if the voltage ripple is high and lasts for a long time, but the phone has good compensation capabilities and a low fatigue correction factor, the growth of the cumulative index will be relatively slow; conversely, if the compensation capabilities are poor and the fatigue correction factor is high, the growth of the cumulative index will accelerate, thus more accurately reflecting the actual risks faced by the phone.

[0045] This application further proposes steps for quantifying the characteristics of compensation capability based on instantaneous change data, including: Analyze instantaneous change data and characteristic parameters to determine whether the instantaneous change data represents a new internal baseline state of the equipment, and obtain the analysis results; Based on the analysis results, the baseline response characteristics are updated, wherein the baseline response characteristics include at least one or more of the following: voltage response amplitude characteristics, voltage response decay time characteristics, and current fluctuation characteristics after micro-perturbation; The compensation capability characteristics are quantified based on the updated baseline response characteristics.

[0046] Specifically, analyzing instantaneous changes in data and characteristic parameters to determine whether these changes represent a new internal baseline state of the equipment involves continuously monitoring instantaneous changes in supply voltage and charging current, and combining this data with identified power quality characteristic parameters. This data is then analyzed in depth using specific algorithms or models. The aim is to identify whether the response pattern of the internal power supply to voltage perturbations has undergone significant and sustained changes, thereby determining whether the internal baseline state of the equipment has drifted to a new level. For example, this analysis may involve applying statistical tests, pattern recognition, or machine learning algorithms to detect deviations from historical baseline behavior.

[0047] Baseline response characteristics can be understood as the typical response pattern exhibited by the device's power supply to minor voltage disturbances after a change in the device's internal baseline state. Specifically, baseline response characteristics include at least one or more of the following: voltage response amplitude characteristics, voltage response decay time characteristics, and post-disturbance current fluctuation characteristics. Voltage response amplitude characteristics describe the maximum or average amplitude of the internal voltage fluctuation when the device is subjected to a minor voltage disturbance; voltage response decay time characteristics indicate the time required for the internal voltage to recover from the disturbance peak to a stable state; and post-disturbance current fluctuation characteristics reflect the fluctuation of the device's charging current or internal current after the voltage disturbance occurs. These characteristics collectively characterize the device's power supply's ability to suppress, absorb, and recover from external disturbances under a new baseline state.

[0048] In practical applications, quantifying compensation capability characteristics based on updated baseline response features means that after recognizing changes in the internal baseline state of the equipment and updating the corresponding baseline response features, the system re-evaluates and quantifies the equipment's compensation capability characteristics based on these new response features. For example, a dynamic model can be established, using the updated voltage response amplitude characteristics, voltage response decay time characteristics, and current fluctuation characteristics after micro-disturbances as inputs, to calculate a compensation capability characteristic value that better reflects the equipment's current actual state. The purpose is to ensure that the quantified compensation capability characteristics accurately reflect the equipment's true performance under the current baseline state, thereby providing a more reliable basis for subsequent fatigue correction factor calculations and cumulative index updates.

[0049] This application's solution effectively addresses the problem of inaccurate quantification of compensation capability characteristics that may arise from internal state drift in traditional methods by introducing a mechanism for judging and updating the internal baseline state of the equipment. Specifically, by analyzing instantaneous change data and characteristic parameters, the system can dynamically identify whether the response mode of the internal power supply to voltage perturbations has changed, thereby determining whether a new internal baseline state exists. Once baseline drift is detected, the system updates the baseline response characteristics in a timely manner based on the analysis results, ensuring that the response model used matches the current actual operating state of the equipment. Therefore, quantifying the compensation capability characteristics based on the updated baseline response characteristics can more accurately capture the equipment power supply's ability to suppress, absorb, and recover from voltage perturbations, avoiding evaluation bias caused by baseline drift.

[0050] In some preferred embodiments, it is assumed that after prolonged use, the charging and discharging characteristics of a smartphone's internal battery management system undergo slight changes due to battery aging, resulting in a different response pattern to external voltage fluctuations compared to the factory baseline. When the charging cable is connected to the phone and charging begins, the system continuously collects instantaneous change data of the supply voltage and charging current, and analyzes this data in conjunction with identified power quality characteristics such as voltage ripple. Through in-depth analysis of this data, such as using machine learning-based anomaly detection algorithms, the system can identify persistent changes in the instantaneous data that do not conform to the historical baseline response pattern, thereby determining that the phone's internal baseline state has drifted.

[0051] Based on this analysis, the system updates the phone's baseline response characteristics, such as adjusting the reference values ​​for voltage response amplitude and voltage response decay time characteristics. For example, if the system detects a slight increase in voltage response amplitude and a slight increase in decay time when the phone is subjected to the same minor perturbation, these updated characteristics will be used to reflect its current aging state. Subsequently, the system will requantify the phone's compensation capability characteristics based on these updated baseline response characteristics. For example, using a dynamic model with the updated voltage response amplitude and voltage response decay time characteristics as input, a more accurate compensation capability characteristic value will be calculated, reflecting the phone's true ability to suppress voltage minor perturbations under its current aging state. Ultimately, this dynamically adjusted and updated compensation capability characteristic will be used to calculate fatigue correction factors and cumulative indices, thereby ensuring that the assessment of potential damage risks to the phone is based on its latest internal state, improving the accuracy and reliability of early warnings.

[0052] The steps described above for analyzing transient data and characteristic parameters to determine whether the transient data represents a new internal baseline state of the device include: Frequency component analysis is performed on instantaneous change data to obtain the spectral distribution of the instantaneous change data; Synchronous frequency component analysis is performed on the characteristic parameters to obtain their spectral distribution. By comparing the spectral distribution of instantaneously changing data with the spectral distribution of characteristic parameters, the energy differences and phase relationships of the spectral distribution within a specific frequency range can be identified. When the energy difference and phase relationship exhibit consistent and continuous change characteristics over multiple charging cycles, and the fluctuation range of the characteristic parameters is within the preset fluctuation range, it is determined that there is a component in the instantaneous change data caused by the internal baseline drift of the device, and thus it is determined that the internal baseline of the device has drifted to a new baseline level. The degree of baseline drift within the device is quantified based on the direction and magnitude of the change characteristics.

[0053] Instantaneous change data refers to the instantaneous changes in the supply voltage and charging current collected after applying a voltage perturbation signal. Frequency component analysis of instantaneous change data aims to reveal its energy distribution at different frequencies, thereby identifying specific frequency responses that may be related to changes in the internal state of the equipment. For example, methods such as Fast Fourier Transform or Wavelet Transform can be used to convert the time-domain signal into a frequency-domain signal to obtain the spectral distribution of the instantaneous change data.

[0054] Simultaneously, synchronous frequency component analysis is performed on the characteristic parameters to obtain the frequency domain representation of external power quality anomalies. By performing frequency analysis in sync with instantaneous data, comparisons can be made within the same time window and frequency resolution, thereby more accurately separating external disturbances from internal responses.

[0055] Furthermore, by comparing the spectral distribution of instantaneously changing data with the spectral distribution of characteristic parameters, energy differences and phase relationships within a specific frequency range can be identified. Energy differences can indicate changes in the device's ability to absorb or amplify disturbances at a specific frequency, while phase relationships can reflect the delay or lead characteristics of the device's internal circuitry in signal processing. These differences and relationships are key indicators for determining whether the device's internal baseline state has changed.

[0056] Specifically, when the aforementioned energy differences and phase relationships exhibit a consistent and continuous change over multiple charging cycles, this indicates a persistent, non-random structural or performance change within the device. Furthermore, if the fluctuation range of the characteristic parameters is within a preset range, the possibility of misjudgment due to drastic fluctuations in external power quality is ruled out. Under these conditions, it can be determined that the instantaneous data contains components caused by baseline drift within the device, thus confirming that the device's internal baseline has drifted to a new baseline level. Baseline drift refers to the slow, continuous change over time in the device's internal power supply's ability to suppress, absorb, and recover from minor voltage disturbances.

[0057] Ultimately, the degree of baseline drift within the device can be quantified based on the direction and magnitude of the change characteristics. For example, the direction of change can indicate whether the baseline is drifting upwards (performance improvement) or downwards (performance degradation), while the magnitude quantifies the severity of the drift.

[0058] This application's solution, by introducing frequency component analysis and a synchronous comparison mechanism, can effectively distinguish the impact of external power quality anomalies and internal baseline drift on instantaneous change data. Traditional methods, when quantifying compensation capability characteristics, may struggle to accurately isolate external environmental factors, leading to biases in the assessment of the true state of the internal power supply. By performing fine-grained frequency domain analysis on instantaneous change data and characteristic parameters, and identifying the energy differences and phase relationships between the two within a specific frequency range, the response characteristics of the internal equipment to micro-disturbances can be captured more accurately. When these differences and relationships exhibit consistent and continuous changes, combined with the condition that the fluctuation amplitude of the characteristic parameters is within a preset range, it is possible to reliably determine whether the internal baseline of the equipment has drifted. This judgment mechanism based on frequency domain characteristics means that the assessment of the internal power supply compensation capability of the equipment no longer relies solely on the instantaneous response in the time domain, but incorporates deeper-level changes in physical characteristics, thereby improving the accuracy and robustness of the judgment.

[0059] The steps for quantifying the degree of baseline drift within the device, based on the direction and magnitude of the change characteristics, include: Multi-scale frequency analysis is performed on instantaneous change data to identify the energy distribution and change patterns within the frequency range corresponding to different subsystems. The change patterns are used to describe the changes in instantaneous change data over time within the corresponding frequency range. Decompose the direction and magnitude of the change characteristics to obtain multiple change components; Based on the energy distribution and variation patterns within the frequency range corresponding to different subsystems, the variation components are mapped to the frequency range corresponding to the subsystems. Based on the statistical analysis of the historical instantaneous change data of the subsystem over multiple charging cycles, a subsystem drift feature set is established. The subsystem drift feature set is a feature set used to describe the drift of the subsystem baseline over time. Compare the variation components with the subsystem drift feature set to quantify the independent contribution of the subsystem to the baseline drift; The independent contributions of each subsystem to baseline drift are summarized to obtain the overall extent of baseline drift within the device; Based on the independent contribution of each subsystem to the baseline drift, subsystem-specific drift information is generated, which is structured information used to indicate the baseline drift state or drift degree of the corresponding subsystem.

[0060] Specifically, "instantaneous change data" refers to the response data of the supply voltage and charging current within a very short time after a voltage micro-perturbation signal is applied. It contains information about the dynamic response of the device's internal circuitry to external disturbances. "Multi-scale frequency analysis" can be understood as analyzing instantaneous change data at different time scales and frequency resolutions, such as using wavelet transform, empirical mode decomposition, or variational mode decomposition. Its purpose is to reveal the frequency components and energy distributions hidden within the data that are related to different physical processes or subsystems. "Subsystem" refers to the functional structural unit within the connected device that exhibits differences in frequency response characteristics, energy distribution, or phase behavior in response to supply voltage disturbances, such as power management units, battery charging circuits, and processor power supply modules.

[0061] Here, "change pattern" refers to the specific form of instantaneous change data evolving over time within a specific frequency range, such as periodic increases or decreases in energy, frequency drift, or amplitude decay, reflecting the dynamic behavior of the subsystem under different operating conditions. "Change component" is a finer-grained component obtained after decomposing the direction and amplitude of the change feature; each component may correspond to a specific drift behavior of one or more subsystems. The "subsystem drift feature set" is a database built based on historical data, describing typical patterns and thresholds of baseline drift over time for each subsystem, aiming to provide a reference benchmark for subsequent comparisons and quantification.

[0062] In practical applications, "quantifying the independent contribution of subsystems to baseline drift" refers to assessing the proportion or degree of influence of each subsystem in the current equipment baseline drift by comparing the current change components with the subsystem drift characteristic set, thereby achieving precise location of the drift source. "Summarizing the independent contribution of subsystems to baseline drift" involves weighted or unweighted summation of the contributions of each subsystem to obtain a precise quantitative value of the overall equipment baseline drift. "Subsystem-specific drift information" is structured data, such as subsystem ID, drift degree, drift trend, and recommended measures, aimed at providing users or remote service platforms with detailed and actionable early warning and diagnostic information.

[0063] This application's solution effectively addresses the problems of coarse quantification of baseline drift within equipment and difficulty in locating specific drift sources in existing technologies by introducing multi-scale frequency analysis and subsystem-independent contribution quantification. Specifically, by performing multi-scale frequency analysis on instantaneous data, the complex overall signal can be decomposed into frequency components related to different subsystems, thereby identifying the energy distribution and variation patterns of each subsystem within different frequency ranges. Subsequently, the direction and amplitude of the overall variation characteristics are decomposed into multiple variation components, which can more precisely reflect the dynamic response of different subsystems. By mapping these variation components to the corresponding subsystem frequency ranges and comparing them with a subsystem drift feature set established based on historical data, the independent contribution of each subsystem to the overall baseline drift can be accurately quantified. This detailed analysis allows the system to delve from the macroscopic overall drift level to the microscopic subsystem level, identifying which subsystem(s) are experiencing baseline drift, as well as the degree and pattern of the drift. Finally, specific drift information for each subsystem is generated based on these independent contributions, providing a more accurate and actionable basis for subsequent diagnosis and early warning.

[0064] In some preferred embodiments, it is assumed that instantaneous changes in the supply voltage and charging current of a smartphone are collected during charging. First, wavelet transform is performed on these instantaneous data, decomposing them into multiple scale components, each representing signal characteristics within a different frequency range. For example, high-frequency components may be related to the switching noise of the power management unit, mid-frequency components may be related to the dynamic response of the battery charging circuit, and low-frequency components may be related to load changes in the processor power supply module. By analyzing the energy distribution and time-varying patterns of these scale components within different frequency ranges, the activity state of each subsystem can be preliminarily identified.

[0065] Furthermore, the variation features extracted from the instantaneous data and characteristic parameters (e.g., the energy difference and phase relationship of the voltage waveform within a specific frequency range) are decomposed to obtain multiple variation components. For example, one variation component may indicate a decline in the performance of the power management unit's filter capacitor, resulting in a weakening of high-frequency noise suppression capability; another variation component may indicate an increase in the internal impedance of the battery charging IC, resulting in a change in the charging current fluctuation pattern.

[0066] Subsequently, these variable components are mapped to the corresponding subsystem frequency ranges. For example, variable components related to decreased high-frequency noise suppression capability are mapped to the power management unit's frequency range. Simultaneously, the system performs statistical analysis on subsystems such as the power management unit and battery charging circuit based on historical instantaneous change data accumulated over multiple charging cycles of the smartphone, establishing their respective subsystem drift characteristic sets. For example, the power management unit's drift characteristic set might include typical high-frequency noise energy distribution patterns of its filter capacitors under different aging conditions.

[0067] Next, the currently identified variation components are compared with the set of drift characteristics of these subsystems. If a variation component closely matches the drift characteristic of a degraded filter capacitor performance in the power management unit, the independent contribution of the power management unit to the current baseline drift can be quantified. For example, the quantification result might show that the power management unit contributes 60% of the overall drift, while the battery charging circuit contributes 30%.

[0068] Finally, the independent contributions of these subsystems to baseline drift are summarized to obtain the overall degree of baseline drift within the device. Simultaneously, based on the independent contribution of the power management unit, subsystem-specific drift information is generated, such as: "Subsystem: Power Management Unit; Drift Degree: Moderate; Drift Trend: Continuously Worsening; Recommendation: Check or replace filter capacitors." This information can then be sent to a remote service platform and pushed to user terminals, enabling users to promptly understand specific problems within the device and take appropriate measures.

[0069] Specifically, in some of the above embodiments, in order to more accurately identify and quantify the degree of baseline drift within the device, this application provides a clear definition of "subsystem".

[0070] The aforementioned subsystems are functional structural units within the connected device that exhibit differences in frequency response characteristics, energy distribution, or phase behavior in response to power supply voltage disturbances. A subsystem can be understood as an independent or semi-independent module within the connected device that possesses a specific function and whose response to external power supply voltage disturbances (such as frequency response, energy absorption or release, signal phase changes, etc.) differs significantly from other functional units. For example, in a smartphone, the power management unit, charging control circuit, display driver circuit, and processor core power supply module can all be considered different subsystems. When dealing with power supply voltage fluctuations, these subsystems exhibit unique frequency response characteristics, energy distribution patterns, or phase behaviors due to differences in their internal circuit design, operating frequency, and load characteristics. By dividing the device's internal structure into these subsystems with differentiated response characteristics, the impact of power supply voltage disturbances on different parts of the device can be analyzed more precisely.

[0071] The proposed solution divides the device internally into subsystems with different response characteristics, enabling more accurate identification and separation of instantaneous variation components caused by different subsystems during multi-scale frequency analysis. It is precisely because of the differences in frequency response characteristics, energy distribution, or phase behavior among the subsystems that multi-scale frequency analysis of instantaneous data can effectively distinguish the energy distribution and variation patterns within the frequency range corresponding to different subsystems. For example, some subsystems may be more sensitive to high-frequency noise, exhibiting specific high-frequency energy absorption or reflection patterns; while other subsystems may exhibit unique compensation or recovery behaviors during low-frequency voltage drops. This differentiated response characteristic forms the basis for independent component separation and mapping of variation components to specific subsystem frequency ranges, thereby enabling the quantification of the independent contribution of each subsystem to baseline drift.

[0072] The steps described above for performing multi-scale frequency analysis on instantaneous data to identify the energy distribution and variation patterns within the frequency range corresponding to different subsystems may include the following operations: Multi-scale decomposition is performed on instantaneously changing data to obtain multiple scale components; Multiple scale components are separated into multiple independent components. Based on the frequency characteristics of the independent components, the independent components are classified into the corresponding subsystem frequency ranges; The energy distribution and variation patterns of the categorized independent components are analyzed to identify the energy distribution and variation patterns within the frequency range corresponding to the subsystem.

[0073] Multi-scale decomposition of instantaneous data aims to break down the original instantaneous data into components with specific characteristics at different frequencies or time scales. For example, techniques such as wavelet transform, empirical mode decomposition, or variational mode decomposition can be used to decompose complex instantaneous signals into a series of scale components with different frequency ranges. The purpose of this is to reveal the intrinsic structure and dynamic characteristics of the signal at different time scales, laying the foundation for subsequent detailed analysis.

[0074] Furthermore, independent component separation is performed on multiple scale components. The aim is to separate any possible mixed signals within these scale components into statistically independent components. Independent component separation is a powerful blind source separation technique capable of recovering potential independent source signals from observed mixed signals. This process effectively isolates signal components from different subsystems from the mixed scale components, thereby reducing mutual interference and improving the accuracy of subsequent analyses.

[0075] After obtaining multiple independent components, they are categorized into corresponding subsystem frequency ranges based on their frequency characteristics. Specifically, spectral analysis can be performed on each independent component to identify its main frequency components and energy distribution. Based on predefined or learned typical characteristics of each subsystem's frequency response, independent components with similar frequency characteristics are assigned to the corresponding subsystems. For example, some subsystems may exhibit significant energy in the low-frequency region, while others have characteristic responses in the high-frequency region.

[0076] Finally, the energy distribution and variation patterns of the categorized independent components are analyzed to identify the energy distribution and variation patterns within the corresponding frequency ranges of the subsystems. This includes calculating the energy intensity of the independent components within each subsystem's frequency range, the energy variation trend over time, and their dynamic response patterns before and after specific events. Through these analyses, a deeper understanding of the specific behavior of each subsystem under supply voltage disturbances can be obtained, such as its energy absorption, release, or conversion patterns, and how these patterns evolve over time, thus providing crucial information for assessing the health status of the subsystems and baseline drift.

[0077] This application's scheme decomposes transient data into multi-scale components, breaking down complex signals into components of different frequencies and time scales, thus revealing the signal's intrinsic structure. Based on this, independent component separation effectively separates signal components from different subsystems within these scale components, yielding statistically independent components. This separation allows each independent component to more purely reflect the behavior of a specific subsystem. Subsequently, these independent components are categorized into their corresponding subsystem frequency ranges based on their frequency characteristics, ensuring the analysis's focus. Finally, by analyzing the energy distribution and variation patterns of the categorized independent components, the energy distribution of each subsystem within a specific frequency range and its time-varying patterns can be accurately identified. This series of processes enables the system to meticulously extract and understand the independent responses of each subsystem to supply voltage disturbances from macroscopic transient data, providing accurate and reliable input data for subsequent quantification of the independent contribution of subsystems to baseline drift.

[0078] This application further proposes a step for independent component separation of the above-mentioned multiple scale components to obtain multiple independent components, including: Statistical property evaluation of multiple scale components is performed to identify components with Gaussian distribution characteristics; Components with Gaussian distribution characteristics are preprocessed; Independent component separation is performed on the preprocessed scale components, and non-Gaussianity is used as the criterion for component separation to obtain multiple mutually independent components. By combining the energy distribution and frequency characteristics of each independent component, it is determined whether each independent component matches the preset subsystem signal characteristics, whereby the subsystem signal characteristics are used to describe the typical signal performance of each subsystem under abnormal conditions. Independent components that do not conform to the characteristics of the subsystem signal are labeled as residual noise or interference.

[0079] Specifically, evaluating the statistical properties of multiple scale components involves calculating statistical measures such as skewness and kurtosis of each scale component, or performing a Gaussian fit goodness-of-fit test, to determine whether its distribution approximates a Gaussian distribution. The aim is to distinguish signal components with Gaussian distribution characteristics that may be generated by random noise or certain specific physical processes, so that targeted processing can be performed.

[0080] Preprocessing components with Gaussian distribution characteristics can be understood as filtering, denoising, or enhancing their non-Gaussianity through transformation. For example, adaptive filters and wavelet thresholding can be used to effectively suppress the influence of Gaussian noise while retaining useful information, thus providing a cleaner input for subsequent independent component separation.

[0081] In practical applications, performing independent component separation on preprocessed scale components, using non-Gaussian properties as the separation criterion, refers to using the Independent Component Analysis (ICA) algorithm. The core idea of ​​this algorithm is to find a set of statistically independent source signals, whose linear combination constitutes the observed mixed signal. Since the independence of Gaussian signals is difficult to measure using higher-order statistics, and most practical source signals (such as specific operating signals within equipment) are often non-Gaussian distributed, maximizing non-Gaussianity as the separation criterion can more effectively separate these non-Gaussian source signals from the mixed signal, obtaining multiple mutually independent components.

[0082] Furthermore, by combining the energy distribution and frequency characteristics of each independent component, determining whether each independent component matches the preset subsystem signal characteristics involves performing spectral analysis on each separated independent component after separation to obtain its energy distribution across different frequency ranges and extracting its main frequency characteristics. These characteristics are then compared with the preset subsystem signal characteristics. The subsystem signal characteristics are established based on prior knowledge or historical data analysis of the typical signal performance of each functional structural unit within the equipment under normal or abnormal operating conditions. For example, a subsystem may have significant energy peaks in a specific frequency range or possess a specific harmonic structure. Through this comparison, it can be confirmed whether the separated independent components truly represent the activity of a specific subsystem.

[0083] Therefore, independent components that do not conform to the characteristics of the subsystem signal are marked as residual noise or interference. The purpose is to identify those components that cannot match the behavior of any known subsystem as non-target signals, so as to exclude them in subsequent analysis and avoid them interfering with subsystem identification and state assessment.

[0084] The proposed solution first evaluates the statistical properties of multiple scale components, effectively identifying those with Gaussian distribution characteristics. These components are often random noise or background interference. Subsequently, preprocessing these Gaussian components effectively suppresses their influence on the subsequent independent component separation process, allowing the independent component analysis algorithm to focus more on separating signals with non-Gaussian properties that truly represent the activities of the internal subsystems of the equipment. It is precisely because independent component separation uses non-Gaussian properties as a criterion that it becomes possible to extract mutually independent signals related to specific subsystems from complex, transiently changing data. Based on this, by combining the energy distribution and frequency characteristics of each independent component and comparing them with preset subsystem signal characteristics, the physical meaning of the separated components can be further verified, ensuring that they indeed correspond to a specific functional structural unit within the equipment. Components that do not conform to the subsystem signal characteristics are marked as residual noise or interference, thereby avoiding interference from these irrelevant components in subsequent analysis and improving the accuracy and reliability of subsystem identification.

[0085] In some preferred embodiments, it is assumed that a series of scale components are obtained by multi-scale decomposition of the instantaneous change data collected by a charging device during the charging process. To accurately identify the independent signals representing different subsystems (e.g., power management unit, battery charging module, communication module, etc.) within these components, the statistical properties of these scale components are first evaluated. For example, by calculating the kurtosis value of each scale component, it is found that some components have a kurtosis value close to 3 (the kurtosis of a Gaussian distribution), indicating that they may mainly consist of random noise. For these components with Gaussian distribution characteristics, wavelet thresholding denoising can be used for preprocessing to reduce their noise level. Subsequently, the preprocessed scale components are input into an independent component separation module based on the FastICA algorithm. This algorithm aims to maximize non-Gaussianity and extract multiple statistically independent components from the mixed signal. For example, it may separate periodic signals representing the switching frequency of the power management unit, battery charging current ripple signals, and modulation signals of the communication module. Next, spectral analysis is performed on these separated independent components to obtain their energy distribution and frequency characteristics. For example, if an independent component has a significant energy peak around 100kHz and its harmonic structure matches the typical operating signal characteristics of a known power management unit (PMU), then that component can be considered representative of the PMU's activity. Conversely, if the energy distribution and frequency characteristics of an independent component do not match any pre-defined subsystem signal characteristics, it is labeled as residual noise or external interference and excluded from subsequent subsystem drift analysis. This ensures that subsequent quantification of subsystem drift is based on real and meaningful subsystem signals.

[0086] This application further proposes a preprocessing step for components with Gaussian distribution characteristics, including: Real-time spectral analysis is performed on components with Gaussian distribution characteristics to identify the energy distribution of components with Gaussian distribution characteristics in different frequency ranges; Synchronous spectrum analysis is performed on the target subsystem signal to obtain the energy distribution of the target subsystem signal within the same frequency range. The target subsystem signal is the signal component selected in the current analysis process and used to match and process the characteristics of the corresponding subsystem signal. By comparing the energy distribution of components with Gaussian distribution characteristics in different frequency ranges with the energy distribution of the target subsystem signal in the same frequency range, the energy overlap region can be determined. For regions of energy overlap, the filtering parameters used in the preprocessing process are dynamically adjusted; Real-time performance evaluation of the preprocessing method with adjusted filtering parameters is used to assess the balance between noise suppression effectiveness and signal fidelity of the target subsystem.

[0087] Specifically, real-time spectral analysis of components with Gaussian distribution characteristics aims to dynamically understand their energy distribution across different frequency ranges. This helps identify which frequency regions may contain important signal information and which regions are primarily noise. The target subsystem signal can be understood as the signal component selected for matching and processing during the current analysis process, based on predefined subsystem signal characteristics. Synchronous spectral analysis of this component is performed to obtain its energy distribution within the same frequency range, serving as a benchmark for subsequent comparisons. In practical applications, by comparing the energy distribution of Gaussian-distributed components across different frequency ranges with the energy distribution of the target subsystem signal within the same frequency range, frequency regions with energy overlap can be accurately determined. For these overlapping regions, the filtering parameters used in preprocessing are dynamically adjusted, such as adjusting the filter cutoff frequency, bandwidth, or gain, to suppress noise while maximizing the preservation of the target subsystem signal's integrity. Furthermore, real-time performance evaluation of the preprocessing method after parameter adjustment is conducted to continuously monitor its effectiveness and ensure an optimal balance between noise suppression and target subsystem signal fidelity. This can be achieved by calculating metrics such as signal-to-noise ratio, signal distortion, or similarity to a known reference signal.

[0088] This application's solution addresses the potential blindness of traditional preprocessing by performing refined preprocessing on components exhibiting Gaussian distribution characteristics. Specifically, through real-time spectrum analysis and synchronous spectrum analysis with the target subsystem signal, it accurately identifies energy overlap regions where noise and the target signal may overlap. Therefore, the filtering parameters during preprocessing are no longer fixed but dynamically adjusted for these overlap regions, avoiding the problems of over-filtering leading to useful signal loss or under-filtering resulting in noise residue. Furthermore, the real-time performance evaluation mechanism ensures the adaptability and optimization of the preprocessing, enabling subsequent independent component separation to be performed on a cleaner and more discriminative signal, thereby improving the accuracy and efficiency of independent component separation.

[0089] As a specific implementation, it is assumed that after multi-scale decomposition of instantaneous data, a certain scale component is evaluated as having Gaussian distribution characteristics. For preprocessing, real-time spectral analysis is first performed on this Gaussian component, revealing significant energy in the 100Hz-200Hz and 500Hz-600Hz frequency ranges. Simultaneously, based on preset subsystem signal characteristics, the target subsystem signal to be matched (e.g., a specific harmonic response of a power module) is identified and subjected to synchronous spectral analysis, revealing that its main energy is concentrated in the 180Hz-220Hz and 580Hz-620Hz ranges. Comparison reveals overlapping energy regions in the 180Hz-200Hz and 580Hz-600Hz ranges. For these overlapping regions, the system dynamically adjusts the parameters of the bandpass filter; for example, slightly narrowing the filter bandwidth in the 180Hz-200Hz region and adjusting its gain to maximize the preservation of the target subsystem signal's energy while suppressing background noise. During preprocessing, the filtering effect is evaluated in real time by monitoring the signal-to-noise ratio (SNR) of the filtered signal and its mean square error relative to the ideal target signal. If the SNR is lower than a preset threshold or the mean square error is too large, the filtering parameters are further fine-tuned until the optimal balance between noise suppression and signal fidelity is achieved. In this way, the quality of the input signal for subsequent independent component separation is ensured, thereby improving the accurate extraction of subsystem signals.

[0090] refer to Figure 2 This application proposes an Internet-based intelligent early warning system for charging cable overload anomalies, applied to the aforementioned Internet-based intelligent early warning method for charging cable overload anomalies. The system includes: The data acquisition module collects power supply voltage information. The identification module identifies and quantifies characteristic parameters reflecting power quality based on the power supply voltage information. Among them, the characteristic parameters include at least one of the following: voltage waveform fluctuation amplitude characteristics, voltage ripple characteristics, instantaneous voltage spike characteristics, voltage drop characteristics, and high-frequency noise energy characteristics. The update module accumulates and updates the cumulative index indicating potential damage to the connected device based on the degree and duration of the characteristic parameter. The degree of the characteristic parameter is the abnormal intensity of the characteristic parameter relative to the preset reference level, and the duration of the characteristic parameter is the continuous existence duration of the abnormal state in the time dimension. The generation module generates an early warning message when the cumulative indicators reach a preset risk threshold. The generation of the early warning message is independent of the monitoring results of charging current information and data line temperature information. The communication module sends the early warning information to the remote service platform via a wireless communication device; The remote service module receives early warning information through the remote service platform and then pushes early warning notifications to user terminals.

[0091] Specifically, the acquisition module can be understood as a hardware or software unit responsible for acquiring real-time power supply voltage data. For example, it can be a voltage sensor integrated into a charging device or power adapter, with the purpose of providing raw data input for subsequent power quality analysis.

[0092] The identification module receives the power supply voltage information provided by the acquisition module and processes it to extract characteristic parameters of power quality. This module may include a signal processing unit and analysis algorithms, such as using Fourier transform, wavelet analysis, and other techniques to identify voltage waveform fluctuation amplitude characteristics, voltage ripple characteristics, instantaneous voltage spike characteristics, voltage drop characteristics, and high-frequency noise energy characteristics. Its purpose is to transform the raw voltage data into quantifiable power quality indicators.

[0093] In practical applications, the update module dynamically accumulates and updates cumulative indicators that indicate potential damage to connected equipment based on the severity and duration of the characteristic parameters output by the identification module. This module can employ time-weighted or event-driven accumulation algorithms. For example, the higher the intensity and the longer the duration of the abnormal characteristic parameters, the faster the cumulative indicator grows. The purpose is to accurately reflect the potential risks posed by power quality anomalies to equipment.

[0094] Furthermore, the generation module is responsible for triggering and generating early warning information when the cumulative indicators reach a preset risk threshold. The unique feature of this module is that its early warning information generation is independent of the monitoring results of charging current and data line temperature. This means that even when the charging current and data line temperature are normal, the system can still issue an early warning as long as the power quality abnormalities reach the risk threshold, aiming to provide more comprehensive and earlier risk warnings.

[0095] In addition, the communication module transmits the generated early warning information to a remote service platform via wireless communication devices such as Wi-Fi, Bluetooth, and cellular networks. The purpose is to achieve remote transmission and centralized management of early warning information.

[0096] Finally, upon receiving the warning information, the remote service module is responsible for pushing the warning notification to user terminals, such as smartphones and tablets. This module can be a cloud server or data center, and its purpose is to ensure that users receive the warning in a timely manner and take appropriate measures.

[0097] This application's solution visualizes each logical step in the aforementioned internet-based intelligent early warning method for charging cable overload anomalies as independent system modules, thereby achieving the physical deployment and automated execution of the method. Specifically, the acquisition module, as the input end of the method, is responsible for acquiring power supply voltage information in real time, providing a data foundation for subsequent analysis. The identification module undertakes the function of identifying and quantifying power quality characteristic parameters in the method, transforming raw voltage data into meaningful anomaly indicators. The update module dynamically assesses the potential damage risk to the equipment by accumulating and calculating the degree and duration of these characteristic parameters, which directly corresponds to the step of accumulating and updating cumulative indicators in the method. The generation module generates an early warning independently of other monitoring results when the risk reaches a preset threshold, ensuring the timeliness and relevance of the warning. Finally, the communication module and the remote service module jointly complete the remote transmission of warning information and user notification, enabling the entire early warning process to be closed-loop, thus effectively transforming the abstract method into a practically operable intelligent early warning system.

[0098] Through the above technical solution, this application provides a specific and feasible hardware / software architecture for implementing an intelligent early warning method for overload anomalies in internet-based charging cables. This systematic implementation not only ensures the automated and real-time operation of the early warning method but also improves the system's maintainability and scalability through modular design. In particular, by decoupling the various functions into independent modules, the system can be more flexibly adjusted and optimized to meet different application scenarios or technological upgrades, thereby significantly improving the accuracy, timeliness, and user experience of the early warning, and effectively reducing the risk of equipment damage caused by abnormal power quality.

[0099] The content disclosed above is only a preferred and feasible embodiment of the present invention, and is not intended to limit the scope of protection of the present invention. Therefore, all equivalent technical changes made based on the content of the present invention specification and drawings are included within the scope of protection of the present invention. Furthermore, the elements therein can be updated as technology develops.

Claims

1. A smart early warning method for overload anomalies in charging cables based on the Internet, characterized in that, The method includes the following steps: Collect power supply voltage information; Based on the power supply voltage information, identify and quantify characteristic parameters that reflect power quality. Among them, the characteristic parameters include at least one of the following: voltage waveform fluctuation amplitude characteristics, voltage ripple characteristics, instantaneous voltage spike characteristics, voltage drop characteristics, and high-frequency noise energy characteristics. Based on the degree and duration of the characteristic parameters, a cumulative index indicating potential damage to the connected equipment is accumulated and updated. The degree of the characteristic parameter is the abnormal intensity of the characteristic parameter relative to a preset reference level, and the duration of the characteristic parameter is the continuous existence of the abnormal state in the time dimension. The cumulative index is used to quantify the degree of risk posed by power quality anomalies to the connected equipment. When the cumulative indicators reach the preset risk threshold, an early warning message is generated. The generation of the early warning message is independent of the monitoring results of charging current information and data line temperature information. The warning information is sent to the remote service platform via a wireless communication device; After receiving the early warning information through the remote service platform, the early warning notification is pushed to the user terminal.

2. The intelligent early warning method for overload anomalies of charging cables based on the Internet as described in claim 1, characterized in that, The steps for accumulating and updating cumulative indicators that indicate potential damage to connected devices, based on the degree and duration of the characteristic parameters, include: Generate voltage perturbation signals; Apply a voltage perturbation signal to the supply voltage; Collect instantaneous change data of power supply voltage and charging current; Based on instantaneous change data, the compensation capability characteristics are quantified. The compensation capability characteristics are characteristic quantities used to characterize the ability of the internal power supply of the connected equipment to suppress, absorb and recover voltage micro-disturbances. Based on the characteristics of the compensation capability, a fatigue correction factor is calculated, whereby the fatigue correction factor is a correction parameter used to characterize the degree of attenuation of the internal power compensation capability of the connected equipment. Based on the fatigue correction factor, the degree of characteristic parameters, and the duration, cumulative indicators indicating potential damage to connected equipment are accumulated and updated.

3. The intelligent early warning method for overload anomalies of charging cables based on the Internet as described in claim 2, characterized in that, The steps for quantifying the characteristics of compensation capability based on instantaneous change data include: Analyze instantaneous change data and characteristic parameters to determine whether the instantaneous change data represents a new internal baseline state of the equipment, and obtain the analysis results; Based on the analysis results, the baseline response characteristics are updated, wherein the baseline response characteristics include at least one or more of the following: voltage response amplitude characteristics, voltage response decay time characteristics, and current fluctuation characteristics after micro-perturbation; The compensation capability characteristics are quantified based on the updated baseline response characteristics.

4. The intelligent early warning method for overload anomalies of charging cables based on the Internet as described in claim 3, characterized in that, The steps for analyzing transient change data and characteristic parameters to determine whether the transient change data represents a new internal baseline state of the device include: Frequency component analysis is performed on instantaneous change data to obtain the spectral distribution of the instantaneous change data; Synchronous frequency component analysis is performed on the characteristic parameters to obtain their spectral distribution. By comparing the spectral distribution of instantaneously changing data with the spectral distribution of characteristic parameters, the energy differences and phase relationships of the spectral distribution within a specific frequency range can be identified. When the energy difference and phase relationship exhibit consistent and continuous change characteristics over multiple charging cycles, and the fluctuation range of the characteristic parameters is within the preset fluctuation range, it is determined that there is a component in the instantaneous change data caused by the internal baseline drift of the device, and thus it is determined that the internal baseline of the device has drifted to a new baseline level. The degree of baseline drift within the device is quantified based on the direction and magnitude of the change characteristics.

5. The intelligent early warning method for overload anomalies of charging cables based on the Internet as described in claim 4, characterized in that, The steps for quantifying the degree of baseline drift within the device, based on the direction and magnitude of the change characteristics, include: Multi-scale frequency analysis is performed on instantaneous change data to identify the energy distribution and change patterns within the frequency range corresponding to different subsystems. The change patterns are used to describe the changes in instantaneous change data over time within the corresponding frequency range. Decompose the direction and magnitude of the change characteristics to obtain multiple change components; Based on the energy distribution and variation patterns within the frequency range corresponding to different subsystems, the variation components are mapped to the frequency range corresponding to the subsystems. Based on the statistical analysis of the historical instantaneous change data of the subsystem over multiple charging cycles, a subsystem drift feature set is established. The subsystem drift feature set is a feature set used to describe the drift of the subsystem baseline over time. Compare the variation components with the subsystem drift feature set to quantify the independent contribution of the subsystem to the baseline drift; The independent contributions of each subsystem to baseline drift are summarized to obtain the overall extent of baseline drift within the device; Based on the independent contribution of each subsystem to the baseline drift, subsystem-specific drift information is generated, which is structured information used to indicate the baseline drift state or drift degree of the corresponding subsystem.

6. The intelligent early warning method for overload anomalies of charging cables based on the Internet as described in claim 5, characterized in that, A subsystem is a functional structural unit within a connected device that exhibits differences in frequency response characteristics, energy distribution, or phase behavior in response to power supply voltage disturbances.

7. The intelligent early warning method for overload anomalies of charging cables based on the Internet as described in claim 5, characterized in that, The steps for performing multi-scale frequency analysis on instantaneous data to identify energy distribution and variation patterns within frequency ranges corresponding to different subsystems include: Multi-scale decomposition is performed on instantaneously changing data to obtain multiple scale components; Multiple scale components are separated into multiple independent components. Based on the frequency characteristics of the independent components, the independent components are classified into the corresponding subsystem frequency ranges; The energy distribution and variation patterns of the categorized independent components are analyzed to identify the energy distribution and variation patterns within the frequency range corresponding to the subsystem.

8. The intelligent early warning method for overload anomalies of charging cables based on the Internet as described in claim 7, characterized in that, The steps for performing independent component separation on multiple scale components to obtain multiple independent components include: Statistical property evaluation of multiple scale components is performed to identify components with Gaussian distribution characteristics; Components with Gaussian distribution characteristics are preprocessed; Independent component separation is performed on the preprocessed scale components, and non-Gaussianity is used as the criterion for component separation to obtain multiple mutually independent components. By combining the energy distribution and frequency characteristics of each independent component, it is determined whether each independent component matches the preset subsystem signal characteristics, whereby the subsystem signal characteristics are used to describe the typical signal performance of each subsystem under abnormal conditions. Independent components that do not conform to the characteristics of the subsystem signal are labeled as residual noise or interference.

9. The intelligent early warning method for overload anomalies of charging cables based on the Internet as described in claim 8, characterized in that, The steps for preprocessing components with Gaussian distribution characteristics include: Real-time spectral analysis is performed on components with Gaussian distribution characteristics to identify the energy distribution of components with Gaussian distribution characteristics in different frequency ranges; Synchronous spectrum analysis is performed on the target subsystem signal to obtain the energy distribution of the target subsystem signal within the same frequency range. The target subsystem signal is the signal component selected in the current analysis process and used to match and process the characteristics of the corresponding subsystem signal. By comparing the energy distribution of components with Gaussian distribution characteristics in different frequency ranges with the energy distribution of the target subsystem signal in the same frequency range, the energy overlap region can be determined. For regions of energy overlap, the filtering parameters used in the preprocessing process are dynamically adjusted; Real-time performance evaluation of the preprocessing method with adjusted filtering parameters is used to assess the balance between noise suppression effectiveness and signal fidelity of the target subsystem.

10. An Internet-based intelligent early warning system for charging cable overload anomalies, applied to the Internet-based intelligent early warning method for charging cable overload anomalies as described in claim 1, characterized in that, The system includes: The data acquisition module collects power supply voltage information. The identification module identifies and quantifies characteristic parameters reflecting power quality based on the power supply voltage information. Among them, the characteristic parameters include at least one of the following: voltage waveform fluctuation amplitude characteristics, voltage ripple characteristics, instantaneous voltage spike characteristics, voltage drop characteristics, and high-frequency noise energy characteristics. The update module accumulates and updates the cumulative index indicating potential damage to the connected device based on the degree and duration of the characteristic parameter. The degree of the characteristic parameter is the abnormal intensity of the characteristic parameter relative to the preset reference level, and the duration of the characteristic parameter is the continuous existence duration of the abnormal state in the time dimension. The generation module generates an early warning message when the cumulative indicators reach a preset risk threshold. The generation of the early warning message is independent of the monitoring results of charging current information and data line temperature information. The communication module sends the early warning information to the remote service platform via a wireless communication device; The remote service module receives early warning information through the remote service platform and then pushes early warning notifications to user terminals.

Images