Method and system for predicting running state of distribution network backup power supply

Abstract

The invention provides an operation state prediction method and system for a distribution network backup power supply. The running state prediction method of the distribution network backup power supply comprises the following steps: acquiring various target power supply parameters in real time, and performing fusion processing on the various target power supply parameters by adopting a Bayesian fusion algorithm to obtain a posterior decision probability; predicting working condition data of the distribution network backup power supply based on the target power supply parameter and the posterior decision probability; and further predicting the future operation state of the distribution network backup power supply according to the working condition data and outputting a prediction result. According to the method, the operation information of the distribution network backup power supply can be comprehensively acquired, association and uncertainty among various parameters are considered, dynamic computing power support can be carried out under different working conditions, and the future operation state of the backup power supply can be evaluated more comprehensively and accurately; besides, dynamic changes of various factors are considered, so that configuration, dynamic management and self-healing processing of the backup power supply are more scientific, and the problem of resource waste or insufficient power supply is avoided.

Description

Technical Field

[0001] This application relates to the field of power prediction technology, and in particular to a method and system for predicting the operating status of backup power in distribution networks. Background Technology

[0002] Backup power supplies for distribution networks are primarily used to reliably power local distribution network automation equipment such as switchgear, DTU / FTU terminal equipment, and communication equipment. With the continuous growth of electricity demand and the increasing complexity of power systems, the stable operation of distribution networks faces numerous challenges. In the event of sudden situations such as natural disasters or equipment failures, and when AC power is lost, the distribution network needs backup power supplies to be quickly deployed to ensure continuous power supply to critical switchgear, DTU / FTU terminal equipment, and communication equipment. However, current online status management of distribution network backup power supplies has many shortcomings, still accounting for a high proportion of fault statistics in distribution networks, and the associated operation and maintenance costs remain high. (1) Existing technology has a single monitoring method for backup power status: relying on only a few parameters, such as battery voltage and current, it is impossible to comprehensively and accurately assess the health status of backup power and the remaining available capacity of the battery. This may lead to situations in actual operation where the backup power is insufficient at critical moments and cannot meet the load demand.

[0003] (2) Lack of dynamic computing power allocation and inefficiency: Traditional static monitoring of backup power in distribution networks mostly relies on manual inspections, which cannot provide dynamic computing power support under different operating conditions, and cannot predict the capacity status and load capacity of backup power. When a local fault occurs in the distribution network or the backup power load changes suddenly, it cannot meet the requirements of rapid fault recovery speed and power supply reliability of the smart grid.

[0004] (3) Lack of accurate load capacity prediction model: When predicting the load capacity of the backup power supply of the distribution network, the existing technology cannot take into account the dynamic changes of factors such as battery degradation and power supply load margin, resulting in load prediction blind spots. Consequently, the configuration, dynamic management and self-healing of the backup power supply lack scientificity, resulting in resource waste or insufficient power supply. Summary of the Invention

[0005] To address the aforementioned technical problems, this application provides a method and system for predicting the operating status of backup power supplies in distribution networks.

[0006] Specifically, this application provides a method for predicting the operating status of a distribution network backup power source, comprising the following steps: real-time acquisition of multiple target power source parameters, and fusion processing of the multiple target power source parameters using a Bayesian fusion algorithm to obtain a posteriori decision probability; prediction of the operating condition data of the distribution network backup power source based on the target power source parameters and the posteriori decision probability; wherein the operating condition data includes at least load capacity and electrical characteristic parameters; and prediction of the future operating status of the distribution network backup power source based on the operating condition data, and outputting the prediction result.

[0007] In the above technical solution, real-time acquisition of multiple target power parameters enables comprehensive acquisition of the operating information of the distribution network backup power supply, providing a reliable foundation for subsequent state prediction. The use of a Bayesian fusion algorithm to fuse multiple target power parameters considers the correlation and uncertainty between various parameters, improving data reliability and allowing for a more comprehensive and accurate assessment of the future operating state of the backup power supply. By obtaining posterior decision probabilities through real-time data acquisition and fusion processing, and then predicting the operating condition data of the distribution network backup power supply based on these data and probabilities, the system can provide dynamic computing power support under different operating conditions, overcoming the limitations of traditional static monitoring relying on manual inspections, and further ensuring the reliability of future operating state predictions for the distribution network backup power supply. Furthermore, considering the dynamic changes of multiple factors makes the configuration, dynamic management, and self-healing processing of the backup power supply more scientific, avoiding resource waste or insufficient power supply.

[0008] Furthermore, the target power parameters include at least voltage, current, and temperature; before fusing the target power parameters, the method further includes: data preprocessing and feature extraction of the multiple target power parameters; wherein, the data preprocessing includes: sequentially denoising, filtering, and normalizing the target power parameters to obtain a first power parameter, and performing posterior probability fusion calculation on the first power parameter to update the posterior probability at time k-1; the feature extraction includes: extracting key features reflecting the health status and operating performance of the backup power supply based on the first power parameter to obtain a second power parameter.

[0009] In the above technical solutions, denoising can remove interference information from the data, filtering can further smooth the data, normalization makes different parameters comparable, and the fusion calculation of posterior probability updates the posterior probability at time k-1, providing a more accurate basis for subsequent fusion processing; extracting key features reflecting the health status and operating performance of the backup power supply reduces data redundancy, highlights important information, and facilitates subsequent analysis and prediction.

[0010] Furthermore, the fusion process includes: predicting the prior probability at time k based on the posterior probability and state transition probability at time k-1, and updating the second power supply parameter based on the target power supply parameter at time k and the prior probability at time k; determining the decision probability based on the second power supply parameter, and fusing multiple decision probabilities to obtain the posterior decision probability.

[0011] In the above technical solution, by predicting the prior probability and updating the second power supply parameter, combined with the target power supply parameter at the current moment, the change in power supply state can be reflected more accurately; by fusing multiple decision probabilities to obtain the posterior decision probability, the reliability of the decision is further improved.

[0012] Furthermore, the step of predicting the operating condition data of the distribution network backup power supply based on the target power supply parameters and the posterior decision probability includes: filtering the second power supply parameters based on the posterior decision probability to obtain the third power supply parameters, and obtaining the electrical characteristic parameters and load-carrying capacity based on the third power supply parameters.

[0013] In the above technical solution, the third power supply parameter is screened based on the posterior decision probability, which can remove some unnecessary or inaccurate parameters, making the subsequent analysis more focused on key information.

[0014] Furthermore, obtaining the electrical characteristic parameters includes: using the Thevenin model, obtaining target response data under load changes through the circuit dynamic equation using the third power supply parameters, and obtaining the electrical characteristic parameters by fitting the target response data using the least squares method.

[0015] In the above technical solution, the Thevenin model can better describe the circuit characteristics of the power supply. It obtains the target response data through the circuit dynamic equation and then uses the least squares method to fit and obtain the electrical characteristic parameters, thereby improving the accuracy of obtaining the electrical characteristic parameters.

[0016] Furthermore, obtaining the load-carrying capacity includes: performing dynamic load coupling and parameter transient processing based on the third power supply parameters and the electrical characteristic parameters to obtain capacity decay degradation characteristics and internal resistance response data; obtaining dynamic load characteristic trends based on the capacity decay data and internal resistance response data, so as to obtain the load-carrying capacity based on the dynamic load characteristic trends.

[0017] The above technical solution takes into account dynamic load coupling and parameter transients, which can more realistically reflect the load-carrying capacity of the power supply in actual operation. The load-carrying capacity is obtained through capacity decay and degradation characteristics and internal resistance response data, which improves the accuracy of load-carrying capacity prediction.

[0018] Furthermore, the future operating status includes at least future performance degradation and remaining lifespan; the prediction of the future operating status of the distribution network backup power supply based on the operating condition data includes: obtaining the current operating status of the distribution network backup power supply based on the operating condition data, and determining the dynamic degradation inflection point and dynamic degradation characteristics of the backup power supply based on the current operating status; analyzing the operating condition change trend based on the operating condition data, and predicting the future performance degradation of the distribution network backup power supply in combination with the dynamic degradation inflection point; and predicting the remaining lifespan of the distribution network backup power supply in combination with the dynamic degradation characteristics and historical data.

[0019] In the above technical solution, by determining the current operating status and dynamic degradation inflection point and characteristics, and combining the operating condition change trend and historical data, the future performance degradation and remaining life of the distribution network backup power supply can be predicted more comprehensively and accurately, providing a strong basis for power supply maintenance and management.

[0020] Based on the same concept, this application also provides a system for predicting the operating status of a distribution network backup power source. The system includes: an external feature quantity synchronous detection and calculation module for real-time acquisition of multiple target power source parameters; a multi-parameter fusion module for fusing the multiple target power source parameters using a Bayesian fusion algorithm to obtain a posteriori decision probability; a load dynamic module for predicting the operating condition data of the distribution network backup power source based on the target power source parameters and the posteriori decision probability; wherein the operating condition data includes at least load capacity and electrical characteristic parameters; a dynamic computing power module for predicting the future operating status of the distribution network backup power source based on the operating condition data to obtain a prediction result; and a human-computer interaction module for outputting the prediction result.

[0021] The above technical solution enables real-time intelligent management of the health status of the power distribution network backup power supply, achieving the effects of timely detection of anomalies to improve maintenance efficiency, accurate troubleshooting to improve fault diagnosis efficiency, and early prevention of accidents to improve accident handling efficiency.

[0022] Furthermore, the system also includes a communication module for transmitting the prediction results to the human-computer interaction module.

[0023] In the above technical solution, the prediction results are transmitted to the human-computer interaction module, which realizes the effective transmission and sharing of data and improves the practicality of the system.

[0024] Furthermore, the system also includes a power control module, which provides power support for the external characteristic synchronous detection and calculation module, the multi-parameter fusion module, the load dynamic module, the dynamic computing power module, the human-computer interaction module, and the communication module.

[0025] In the above technical solution, the power control module provides power support for each module, ensuring the stable operation of the system.

[0026] Compared with the prior art, the beneficial effects of this application are as follows: This application can comprehensively obtain the operation information of the backup power supply of the distribution network, and takes into account the correlation and uncertainty between various parameters. At the same time, it can provide dynamic computing power support under different operating conditions, and can more comprehensively and accurately assess the future operating status of the backup power supply. In addition, it considers the dynamic changes of various factors, making the configuration, dynamic management and self-healing of the backup power supply more scientific, and avoiding the problems of resource waste or insufficient power supply. Attached Figure Description

[0027] Figure 1 This is a flowchart of the method for predicting the operating status of backup power supply in distribution networks as described in this application.

[0028] Figure 2 This is a framework diagram of the external feature synchronous detection and calculation module described in this application.

[0029] Figure 3 This is a framework diagram of the distribution network backup power supply operation status prediction system described in this application.

[0030] Figure 4 This is a framework diagram of the power control module described in this application.

[0031] Figure 5 This is a framework diagram of the main control unit for the algorithm described in this application. Detailed Implementation

[0032] The following describes in further detail a method and system for predicting the operating status of a distribution network backup power source according to specific embodiments and accompanying drawings.

[0033] Please see Figure 1 This application provides a method for predicting the operating status of a backup power supply in a distribution network, including the following steps S100-S300.

[0034] Step S100: Collect multiple target power parameters in real time, and use a Bayesian fusion algorithm to fuse the multiple target power parameters to obtain the posterior decision probability.

[0035] The target power parameters include at least voltage, current, and temperature. Before fusing the target power parameters, the method further includes: data preprocessing and feature extraction of the multiple target power parameters. The data preprocessing includes: sequentially denoising, filtering, and normalizing the target power parameters to obtain a first power parameter, and performing posterior probability fusion calculation on the first power parameter to update the posterior probability at time k-1. The feature extraction includes: extracting key features reflecting the health status and operating performance of the backup power supply based on the first power parameter to obtain a second power parameter.

[0036] In some embodiments, the external characteristic synchronous detection and calculation module 6 receives control instructions from the CPU and processes analog signals through filtering, linear opto-isolation, operational amplification, etc., before sending them back to the CPU for dynamic data acquisition and calculation. This enables real-time acquisition and dynamic monitoring of parameters such as input / output voltage, charging / discharging current, temperature, and backup battery internal resistance and capacity. Please refer to [link to relevant documentation]. Figure 2 The external characteristic synchronous detection and calculation module 6 includes a voltage sensor 61, a current sensor 62, an AD conversion circuit 63, a temperature sensor 64, and an operational amplifier 65; the temperature sensor 64 is preferably a surface-mount semiconductor temperature sensor.

[0037] The voltage divider circuit 61 divides the voltage according to the level of the backup power supply voltage of the distribution network, and then inputs it to the AD conversion circuit 63 after passing through an operational amplifier to realize the extraction and analog-to-digital conversion of voltage data; the current sensor 62, based on magneto-electric induction, inputs it to the AD conversion circuit 63 after integral conversion and operational amplification to realize the extraction and analog-to-digital conversion of current data; the AD conversion circuit 63, based on the analog-to-digital conversion principle, converts the analog voltage and current quantities and the analog temperature data quantities into digital quantities, and then sends them to the CPU chip in the main control circuit 1 for data processing; the temperature sensor 64 is composed of a power supply and a thermistor, and forms an analog voltage value that is linearly correlated with the temperature change based on the resistance change caused by the temperature change of the measured object to realize the temperature sensing of the measured object; the operational amplifier 65 amplifies the analog voltage value of the temperature sensor 64 into the analog voltage quantity required by the AD conversion circuit 63 according to the required amplitude of the acquisition.

[0038] Furthermore, the multi-parameter fusion module 3 performs preprocessing operations such as denoising, filtering, and normalization on the acquired target power parameters, which is then used to obtain the first power parameters. Through the above preprocessing, interference and outliers in the data are eliminated, and the dynamic response patterns of key parameters such as voltage, current, capacity status, and temperature over time are obtained. Data with different dimensions are unified to the same scale, providing a high-quality data foundation for subsequent fusion analysis, including voltage dynamic response, capacity estimation accuracy, power limit constraints, and temperature effects.

[0039] Furthermore, for the first power supply parameters B1, B2, ..., Bn, it is necessary to fuse multi-source information to update the posterior probability. Assuming that the multi-dimensional input observation data are independent, the joint likelihood probability is the product of the likelihoods of each multi-dimensional input observation data, i.e., P(B1, B2, ..., Bn|A) = Пi=1nP(Bi|A).

[0040] The posterior probability is updated by fusing multi-source information according to the Bayesian fusion algorithm P(A|B1,B2,...,Bn)=P(B1,B2,...,Bn)|A)*P(A) / P(B1,B2,...,Bn).

[0041] The newly acquired fusion factor P(B1,...,Bn)=∑AP(B1,...,Bn|A)P(A) is updated in real time and used to normalize and update the newly acquired fusion probability.

[0042] Furthermore, the extraction of key features specifically includes: by analyzing the battery voltage and current change curves over time, extracting features such as battery charging and discharging characteristics and internal resistance changes; then using temperature and humidity data to analyze the impact of environmental factors on backup power performance; finally obtaining the fault model and its feature quantities; and using the multi-parameter fusion algorithm program to capture and analyze the dynamic data distortion features of the backup power supply in the distribution network, thereby obtaining the second power parameters.

[0043] Furthermore, the fusion process includes: predicting the prior probability at time k based on the posterior probability and state transition probability at time k-1, and updating the second power supply parameter based on the target power supply parameter at time k and the prior probability at time k; determining the decision probability based on the second power supply parameter, and fusing multiple decision probabilities to obtain the posterior decision probability.

[0044] In some embodiments, the posterior probability at time k-1 is known. ( (Based on the first k-1 acquisitions), predict the prior probability at time k: .

[0045] in, Let be the state transition probability, representing the state from time k-1. Transition to state at time k The probability of.

[0046] Combined with observations at time k Update to obtain new data or features: .

[0047] in, For the original input parameter model, It is a characteristic factor.

[0048] Furthermore, the final decision is obtained by outputting the failure type probabilities for multiple hypotheses. Let the original input parameters of the acquired model be relative to the hypotheses. The decision probability is The fusion formula is: .

[0049] in, Assumption The prior probability, For the original input parameters against the true hypothesis The correct decision update probability is obtained, and finally the posterior decision probability is obtained through normalization.

[0050] Step S200: Predict the operating condition data of the distribution network backup power supply based on the target power supply parameters and the posterior decision probability; wherein the operating condition data includes at least load capacity and electrical characteristic parameters.

[0051] The step of predicting the operating condition data of the distribution network backup power supply based on the target power supply parameters and the posterior decision probability includes: filtering the second power supply parameters based on the posterior decision probability to obtain the third power supply parameters, and obtaining the electrical characteristic parameters and load-carrying capacity based on the third power supply parameters.

[0052] In some embodiments, the posterior decision probability can help filter and refine the second power supply parameters, making the model input more accurate.

[0053] Furthermore, obtaining the electrical characteristic parameters includes: using the Thevenin model, obtaining target response data under load changes through the circuit dynamic equation using the third power supply parameters, and obtaining the electrical characteristic parameters by fitting the target response data using the least squares method.

[0054] In some embodiments, the load dynamic module 4 constructs a dynamic prediction model based on the multi-dimensional characteristic parameters (i.e., the third power supply parameters) monitored as input parameters. This model is used to determine the dynamic response of key parameters such as voltage, current, internal resistance, capacity status, and temperature of the backup power supply over time under dynamic loads (including constant current, pulse current, random power, etc.). The model is combined with the battery electrochemical characteristics and external load characteristics to establish a dynamic change model of load characteristics.

[0055] Using the Thevenin model, multiple RC branches (polarization resistors) are added to the steady-state scenario with ohmic internal resistance. +Polarized capacitor This can describe the dynamic polarization effect at different time scales, and the formula is: .

[0056] Among them, polarization voltage (t) satisfies the dynamic equation: ; is the time constant.

[0057] The electrical characteristic parameters of backup power sources in a distribution network can be simulated using an equivalent circuit model. These electrical characteristic parameters include at least the ohmic internal resistance. Polarization internal resistance Polarized capacitors And open-circuit voltage OCV, specifically: Equivalent circuit model parameters ( The curve function OCV=f(SOC) is fitted by the static parameter open-circuit voltage OCV, and the voltage transient response curve under transient current is inferred from the voltage response data using the least squares algorithm to fit the polarization resistance. and polarization capacitor .

[0058] Furthermore, obtaining the load-carrying capacity includes: performing dynamic load coupling and parameter transient processing based on the third power supply parameters and the electrical characteristic parameters to obtain capacity decay degradation characteristics and internal resistance response data; obtaining dynamic load characteristic trends based on the capacity decay data and internal resistance response data, so as to obtain the load-carrying capacity based on the dynamic load characteristic trends.

[0059] In some embodiments, input features I(t), I(t−1), U(t−1), SOC(t), T(t) are divided with output labels U(t), SOC(t+1), I(t). The parameters are iteratively optimized using dynamic training set data, and the dynamic response accuracy (such as root mean square error RMSE) is evaluated. The structure is adjusted or the amount of data is increased to dynamically optimize the model.

[0060] Furthermore, dynamic load coupling and instantaneous parameter processing are performed. The SOC parameter curve is fitted using the function R0=f(SOC) which describes the variation of parameters such as OCV and internal resistance with capacity and lifetime. This curve is then called in real time within the model. The internal resistance response is obtained through the parameter-temperature function R0=f(SOC,T), and the temperature input in the model is also obtained. Simultaneously, an aging factor is introduced through R0=f(SOC,T,N) to monitor the effect of cycle number on capacity decay and internal resistance response. Finally, the dynamic load characteristic trend is obtained by calculating the errors (such as RMSE and mean absolute error MAE) between the model's predicted voltage, internal resistance, capacity, and load capacity and the measured voltage, internal resistance, capacity, and load capacity.

[0061] Step S300: Predict the future operating status of the power distribution network backup power supply based on the operating condition data, and output the prediction results.

[0062] The future operating status includes at least future performance degradation and remaining lifespan. Predicting the future operating status of the distribution network backup power supply based on the operating condition data includes: obtaining the current operating status of the distribution network backup power supply based on the operating condition data, and determining the dynamic degradation inflection point and dynamic degradation characteristics of the backup power supply based on the current operating status; analyzing the operating condition change trend based on the operating condition data, and predicting the future performance degradation of the distribution network backup power supply in conjunction with the dynamic degradation inflection point; and predicting the remaining lifespan of the distribution network backup power supply in conjunction with the dynamic degradation characteristics and historical data.

[0063] In some embodiments, the dynamic computing module 5 monitors the operating conditions of the backup power supply in the distribution network in real time, including parameters such as voltage, current, capacity, load, temperature, and charging / discharging status, and completes the backup power supply fault model trigger point. By analyzing these operating condition data, it determines the current operating status of the backup power supply in the distribution network, completes the backup power supply fault model and trigger processing, and completes the fault information and load change status, such as normal operation, overload, and fault. Then, based on the capture of the distortion characteristics of the battery dynamic data and the capture of the online capacity inflection point of the battery cell, it triggers the dynamic judgment logic program of the health status, realizes the identification of the battery's load-carrying capacity and outputs the result.

[0064] In summary, the proposed method collects multiple target power parameters in real time, enabling comprehensive acquisition of operational information of the distribution network's backup power supply and providing a reliable foundation for subsequent state prediction. The use of a Bayesian fusion algorithm to fuse multiple target power parameters considers the correlation and uncertainty between various parameters, improving data reliability and allowing for a more comprehensive and accurate assessment of the backup power supply's future operational status. By obtaining posterior decision probabilities through real-time data acquisition and fusion processing, and then predicting the operating conditions of the distribution network's backup power supply based on these data and probabilities, the system can provide dynamic computing power support under different operating conditions, overcoming the limitations of traditional static monitoring that relies on manual inspections, further ensuring the reliability of future operational status predictions for the distribution network's backup power supply. Furthermore, considering the dynamic changes of multiple factors makes the configuration, dynamic management, and self-healing of backup power supplies more scientific, avoiding resource waste or insufficient power supply.

[0065] Based on the same concept, please refer to Figure 3 This application also provides a system for predicting the operating status of a distribution network backup power source. The system includes: an external feature quantity synchronous detection and calculation module 6, used to collect multiple target power source parameters in real time; a multi-parameter fusion module 3, used to fuse the multiple target power source parameters using a Bayesian fusion algorithm to obtain a posteriori decision probability; a load dynamic module 4, used to predict the operating condition data of the distribution network backup power source based on the target power source parameters and the posteriori decision probability; wherein the operating condition data includes at least load capacity and electrical characteristic parameters; a dynamic computing power module 5, used to predict the future operating status of the distribution network backup power source based on the operating condition data to obtain prediction results; and a human-computer interaction module 8, used to output the prediction results.

[0066] Furthermore, the system also includes a communication module 7 for transmitting the prediction results to the human-computer interaction module 8.

[0067] The communication module 7, through differential level signal isolation and IC integration technology, realizes an RS485 protocol transceiver unit that integrates power isolation, signal isolation, RS485 communication and bus protection, easily realizing RS485 protocol network connection function, and controlling the sending and receiving of information with the host computer (configured with human-machine interaction module 8) using the standard MODBUS protocol format, while controlling the opening and closing of the port.

[0068] Furthermore, the system also includes a power control module 1, which provides power support for the external characteristic synchronous detection and calculation module 6, the multi-parameter fusion module 3, the load dynamic module 4, the dynamic computing power module 5, the human-computer interaction module 8, and the communication module 7.

[0069] It should be noted that the power control module 1 is the main unit for backup power supply and charging / discharging, connected to the AC power supply and load, backup battery, and switching between main power supply and backup power; please refer to [link to relevant documentation]. Figure 4 The power control module 1 includes an EMC input terminal 11, a rectifier 12, a PFC power factor control circuit 13, a high-frequency power conversion driver on the high-voltage side 14, an isolation step-down converter 15, an output-side rectifier and voltage regulation control 16, and a DC / DC constant voltage and constant current control 17, which are used to complete the power conversion process from AC high voltage to DC low voltage and constant voltage and constant current charging unit.

[0070] The EMC input terminal 11 consists of a fuse, a varistor, an NTC thermistor, a common-mode inductor, a differential-mode inductor, an X capacitor, a Y capacitor, and a discharge resistor. The fuse is used to cut off the circuit and protect other components in case of overcurrent or other faults. The varistor can quickly conduct when the voltage rises abnormally, clamping the overvoltage to a certain level and providing overvoltage protection. The NTC thermistor can suppress surge current during power-on. The common-mode inductor is a symmetrically wound bi-wire inductor connected in series with the phase and neutral lines. It exhibits a large inductance for common-mode interference signals and a relatively small inductance for normal differential-mode signals, effectively suppressing common-mode interference from passing through the normal power supply signal. The differential-mode inductor consists of one or more inductors connected between the positive and negative terminals of the power supply, used to block the propagation of differential-mode noise, acting as a low-pass filter. However, in some cases, the differential mode inductor can be omitted; the X capacitor is connected between the live wire (L) and the neutral wire (N), and its capacitance is generally large and needs to withstand high voltage to withstand transient fluctuations in the AC power supply, and is used to filter differential mode interference signals; the Y capacitor is a line bypass capacitor, which is connected across the floating ground (FG) and the live wire (L) / neutral wire (N). To prevent leakage current from affecting safety, its capacitance is generally small, and it is used to guide common mode interference signals to ground; the discharge resistor is connected in parallel with the X capacitor to quickly release the charge on the X capacitor after the power is turned off, so as to avoid the risk of electric shock or interference caused by residual voltage.

[0071] The diodes in the rectifier 12 are the core components of the rectifier circuit. The full-wave bridge rectifier circuit consists of four diodes connected in a bridge structure. Regardless of whether the input AC current is in the positive or negative half-cycle, it can be converted into DC pulse output in the same direction. In actual design, the four diodes are packaged in a modular rectifier bridge to complete the rectification.

[0072] The PFC (Power Factor Control) circuit 13 consists of a control IC, a power switch, a fast recovery diode, an inductor, an output capacitor, and sensing elements. The control IC is the core of the PFC circuit, generating PWM signals to control the switching action of the power switch. It also adjusts the duty cycle by detecting the input voltage and current to ensure the current and voltage are in phase. The power switch has high switching speed and low on-resistance, enabling rapid switching at high frequencies and controlling energy transfer in the circuit; it is a key component for achieving PFC functionality. The fast recovery diode has low losses during reverse recovery and provides a freewheeling path for the inductor current when the switch is off. The inductor stores and releases energy; it stores energy when the switch is on and releases it to the output when the switch is off, while simultaneously adjusting the waveform of the input current to synchronize it with the input voltage waveform.

[0073] The high-voltage side power conversion high-frequency drive 14 and the high-frequency converter isolation step-down 15 are the core modules for achieving high-efficiency energy conversion. They consist of a drive chip (controller), power switching transistors, a bootstrap circuit, a high-frequency transformer and resonant elements, and a signal detection and feedback network. The drive chip (controller) is responsible for generating precise PWM signals to control the switching transistors, including voltage and current loops. It dynamically adjusts the duty cycle by detecting the bus voltage and inductor current to ensure the switching transistors operate in optimal condition. The power switching transistors work with the drive chip to achieve efficient switching. The gate capacitance and gate charge of the switching transistors directly affect the drive power consumption; the charging and discharging speed is optimized through the drive resistor. The bootstrap circuit includes a self-... The system consists of a bootstrap capacitor, a bootstrap diode, and a bootstrap resistor. The bootstrap capacitor stores charge to power the high-side drive, and its capacitance must meet the gate charge requirements. The bootstrap diode prevents backflow of charge and must be an ultra-fast recovery diode with a reverse withstand voltage higher than the bus voltage. The bootstrap resistor limits the charging current, reduces EMI, and protects the diode. In the high-frequency transformer and resonant components, the high-frequency transformer achieves voltage transformation and electrical isolation, while the resonant converter achieves zero-voltage switching (ZVS) through frequency modulation. The signal detection and feedback network includes current and voltage detection circuits and temperature and fault monitoring components. The current and voltage detection circuits provide feedback signals to the driver chip, while the temperature and fault monitoring components monitor the temperature of the switching transistor or transformer and trigger overheat protection.

[0074] The core of the output-side rectification and voltage regulation control 16 is to convert the AC power output from the secondary side of the high-frequency transformer into unidirectional pulsating DC power. It is a coordinated working process of "rectification → filtering → voltage regulation → protection". The rectifier converts high-frequency AC into pulsating DC, the LC filter network smooths the ripple, the closed-loop control circuit dynamically adjusts the output stability through feedback, and the protection circuit ensures reliability under extreme operating conditions.

[0075] The DC / DC constant voltage and constant current control 17 is designed to meet the application requirements of several different levels of DC power supplies in the circuit. It integrates a DC / DC conversion chip, a reference voltage source, filtering, feedback, etc., to realize the conversion and constant voltage and constant current control of several different levels of DC power supplies.

[0076] Please see below. Figure 5 It also includes an algorithm processing main control unit 2, which uses a domestically produced high-performance 32-bit MCU microcontroller chip to form a microcontroller, and completes the intelligent performance of A / D conversion, dynamic status data operation, dynamic judgment of health status, data storage, output alarm drive, etc. It includes peripheral crystal oscillator circuit 21, clock circuit 22, storage circuit 23, multi-dimensional data acquisition unit 24, multi-parameter fusion algorithm program 25, capacity decay model algorithm program 26, and dynamic judgment program of health status 27.

[0077] Among them, the peripheral crystal oscillator circuit 21 is an external circuit that provides precise timing for the CPU; the clock circuit 22 and the storage circuit 23 are integrated circuits inside the CPU; the clock circuit 22 provides precise time to record accurate time points; and the storage circuit 23 is used to store set values, cached data, and device information. The multi-dimensional data acquisition unit 24 completes data acquisition, i.e., executes the external feature quantity synchronous detection and calculation module 6. The multi-parameter fusion algorithm program 25 completes the fault model and feature quantity calculation of the power supply for the distribution network backup power, i.e., executes the multi-parameter fusion module 3. The capacity decay model algorithm program 26 completes the dynamic capacity change curve management and accurate capture of the battery degradation inflection point during the operation of the power supply for the distribution network backup power, i.e., executes the load dynamic module 4. The dynamic health status judgment program 27 completes the identification and result output of the load-carrying capacity of the power supply for the distribution network backup power, i.e., executes the dynamic computing power module 5.

[0078] Although exemplary embodiments have been described herein with reference to the accompanying drawings, it should be understood that the above exemplary embodiments are merely illustrative and are not intended to limit the scope of this application. Various changes and modifications can be made therein by those skilled in the art without departing from the scope and spirit of this application. All such changes and modifications are intended to be included within the scope of this application as claimed in the appended claims.

[0079] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0080] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed.

[0081] The various component embodiments of this application can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some modules according to the embodiments of this application. This application can also be implemented as an apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such an implementation of this application can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

[0082] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0083] Although the description of this application has been made in conjunction with the specific embodiments described above, it will be apparent to those skilled in the art that many substitutions, modifications, and variations can be made based on the foregoing. Therefore, all such substitutions, modifications, and variations are included within the spirit and scope of the appended claims.

Claims

1. A method for predicting the operating state of a backup power supply of a power distribution network, characterized by, The method comprises the following steps: Real-time acquisition of multiple target power parameters, and fusion processing of the multiple target power parameters by using a Bayesian fusion algorithm to obtain posterior decision probability; Prediction of working condition data of the distribution network backup power supply based on the target power parameters and the posterior decision probability; wherein the working condition data at least includes load carrying capacity and electrical characteristic parameters; And prediction of future operating state of the distribution network backup power supply according to the working condition data, and output of the prediction result.

2. The method of claim 1, wherein, The target power parameters at least include voltage, current and temperature; before the fusion processing of the target power parameters, the multiple target power parameters are further subjected to data preprocessing and feature extraction; The data preprocessing includes denoising, filtering and normalization of the target power parameters in sequence to obtain first power parameters, and fusion calculation of posterior probability of the first power parameters to update the posterior probability at k-1 time; The feature extraction includes extraction of key features reflecting the health state and operating performance of the backup power supply based on the first power parameters to obtain second power parameters.

3. The method of claim 2, wherein, The fusion processing includes: Prediction of prior probability at k time according to the posterior probability at k-1 time and state transition probability, and updating of the second power parameters according to the target power parameters at k time and the prior probability at k time; Determination of decision probability according to the second power parameters, and fusion of multiple decision probabilities to obtain posterior decision probability.

4. The method of claim 3, wherein, The prediction of working condition data of the distribution network backup power supply based on the target power parameters and the posterior decision probability includes: Screening of the second power parameters based on the posterior decision probability to obtain third power parameters, and obtaining of electrical characteristic parameters and load carrying capacity based on the third power parameters.

5. The method of claim 4, wherein, The obtaining of electrical characteristic parameters includes: Using Thevenin model, obtaining target response data under load change by using circuit dynamic equation through the third power parameters, and fitting electrical characteristic parameters based on the target response data by least square method.

6. The method of claim 5, wherein, The obtaining of load carrying capacity includes: Dynamic load coupling and parameter instantaneous processing based on the third power parameters and the electrical characteristic parameters to obtain capacity attenuation degradation characteristics and internal resistance response data; Obtaining of dynamic load characteristic trend according to the capacity attenuation data and internal resistance response data, and obtaining of load carrying capacity based on the dynamic load characteristic trend.

7. The method of claim 6, wherein the operating state of the backup power supply is predicted based on the state of charge of the backup power supply. The future operating state at least includes future performance degradation and remaining life; the prediction of future operating state of the distribution network backup power supply according to the working condition data includes: Obtaining of current operating state of the distribution network backup power supply according to the working condition data, and determination of backup power supply dynamic degradation inflection point and dynamic degradation characteristics according to the current operating state; Analysis of working condition change trend based on the working condition data, and prediction of future performance degradation of the distribution network backup power supply in combination with the backup power supply dynamic degradation inflection point; Prediction of remaining life of the distribution network backup power supply in combination with the dynamic degradation characteristics and historical data.

8. A system for predicting the operating state of a backup power supply for network configuration, using the method according to any one of claims 1 to 7, characterized in that The system comprises: An external characteristic quantity synchronous detection and operation module is configured to collect multiple target power supply parameters in real time, A multi-parameter fusion module is configured to perform fusion processing on the multiple target power supply parameters by using a Bayesian fusion algorithm to obtain posterior decision probability; A load dynamic module is configured to predict working condition data of a distribution network backup power supply based on the target power supply parameters and the posterior decision probability, wherein the working condition data at least includes load carrying capacity and electrical characteristic parameters; A dynamic computing power module is configured to predict a future operating state of the distribution network backup power supply according to the working condition data to obtain a prediction result; And a human-computer interaction module is configured to output the prediction result.

9. The system of claim 8, wherein, The system further includes a communication module configured to transmit the prediction result to the human-computer interaction module.

10. The system of claim 8, wherein, The system further includes a power master control module configured to provide power supply support for the external characteristic quantity synchronous detection and operation module, the multi-parameter fusion module, the load dynamic module, the dynamic computing power module, the human-computer interaction module, and the communication module.

Images