A circuit protection control method for input and output of the same circuit

Abstract

The present invention relates to the field of power electronics and circuit protection control, and in particular to a circuit protection control method for the same input and output circuit. The method comprises the following steps: obtaining the current, voltage and frequency signals in the input and output circuits, and combining the temperature and humidity parameters of the circuit environment, preprocessing and noise filtering the signals to generate a cleaned signal data sequence; based on the cleaned signal data, combining the adaptive algorithm with the multidimensional feature extraction model, analyzing the type and location of the circuit fault, and generating fault diagnosis data; inputting the fault diagnosis data into the protection mechanism model, dynamically adjusting the protection threshold and triggering the protection action; optimizing the circuit energy feedback path according to the protection status data and the cleaning signal data, and generating the optimized energy feedback parameters. The present invention uses an intelligent algorithm and a dynamic optimization mechanism to diagnose faults in real time and optimize the protection strategy, effectively improving the reliability and energy efficiency of the circuit system.

Description

Technical Field

[0001] The present invention relates to the technical field of power electronics and battery management, and in particular to a circuit protection control method for inputting and outputting the same circuit. Background Art

[0002] With the advancement of battery technology, especially its widespread application in electric vehicles and energy storage systems, battery protection technology and energy management technology have gradually become key issues in improving the safety and energy efficiency of battery systems. In traditional battery protection technology, protection mechanisms such as overload, overcharge, and over-discharge are usually implemented through fixed threshold settings. Although this method can ensure the basic safety of the battery, it has problems such as insufficient response time and insufficient precision of the protection mechanism.

[0003] Especially in complex battery systems with the same input and output circuits, due to the complexity of energy flow and load changes, traditional protection technologies are difficult to achieve real-time fault diagnosis and protection response, and it is difficult to optimize the energy feedback path, resulting in a waste of battery system efficiency. In addition, existing technologies usually separate battery protection from energy management, resulting in insufficient synergy between the protection system and energy feedback control, affecting the overall performance and reliability of the battery system.

[0004] Therefore, how to achieve real-time fault diagnosis based on multi-dimensional signals, dynamically adjust protection thresholds, and optimize energy feedback paths to improve the efficiency and safety of battery systems is a technical problem that needs to be solved in the current battery protection field. The present invention solves the above problems through intelligent algorithms and dynamic optimization mechanisms, and provides a more efficient, intelligent, and sustainable battery protection control solution. Summary of the invention

[0005] The present invention provides a circuit protection control method for the same input and output circuit, so as to solve the problem of how to analyze the fault status in real time and respond quickly based on multi-dimensional signals in a complex circuit system with the same input and output circuit, and realize the coordinated optimization of energy feedback and overload protection, so as to improve the reliability and energy efficiency of the circuit system.

[0006] In order to solve the above technical problems, the present invention provides a circuit protection control method for inputting and outputting the same circuit, comprising: Acquire the current, voltage and frequency signals in the input and output circuits, and perform preprocessing and noise filtering on the signals in combination with the temperature and humidity parameters of the circuit environment to generate a cleaned signal data sequence; Based on the cleaned signal data sequence, combining an adaptive algorithm with a multi-dimensional feature extraction model, analyzing the type and location of the loop fault to generate fault diagnosis data; Inputting the fault diagnosis data into a protection mechanism model, dynamically adjusting the protection threshold of the model, triggering corresponding protection actions, and generating protection status data; Based on the protection status data and the cleaning signal data sequence, the energy feedback path of the loop is optimized in combination with a dynamic adjustment algorithm to generate optimized energy feedback parameters; The cleaning signal data sequence, fault diagnosis data, protection status data and energy feedback parameters are uploaded to the cloud platform, and the fault diagnosis model and protection mechanism model are updated in combination with the historical operation data to generate an optimized real-time protection control strategy.

[0007] Furthermore, before the step of obtaining the current, voltage and frequency signals in the input and output circuits, the method further includes: The real-time current and voltage signals in the loop are collected by sensors, and the signals are preliminarily normalized in combination with the collected ambient temperature and humidity data to obtain a preliminary signal sequence; Furthermore, the steps of signal preprocessing and noise filtering specifically include: Filtering the preliminary signal sequence to remove abnormal signal points, and denoising the collected current, voltage and frequency signals to generate a cleaned signal data sequence; Furthermore, the step of performing fault analysis based on the cleaned signal data sequence specifically includes: In combination with the signal data sequence, a multidimensional feature extraction model is used to extract the time domain features and frequency domain features of the signal to generate a signal feature matrix; The signal feature matrix is ​​classified and analyzed through an adaptive algorithm to determine the fault type and location in the circuit; Furthermore, the fault diagnosis data includes: Label of the type of fault that occurred in the circuit in question , the specific location where the circuit fault occurs , and the signal feature matrix related to the fault ; Furthermore, the step of triggering the corresponding protection action specifically includes: According to the fault type and fault location data, trigger the corresponding protection mechanism and adjust the protection threshold in the protection model to adapt to the current working state of the circuit; Among them, the protection mechanisms include overload protection, short circuit protection and ground protection; Furthermore, the protection status data includes: Time to trigger protection action , the type of protection action triggered , the adjusted protection threshold , and the protection action execution status; Furthermore, the step of optimizing the feedback path specifically includes: Combining the cleaning signal data sequence with the protection status data, analyzing the energy flow between the input and output in the loop through a dynamic adjustment algorithm, optimizing the energy feedback path, and generating new energy feedback parameters; The energy feedback path optimization algorithm adjusts the feedback path according to the operating state of the loop to reduce energy loss and improve the efficiency of the circuit system; Furthermore, the step of uploading to the cloud platform specifically includes: Uploading the cleaning signal data sequence, fault diagnosis data, protection status data and optimized energy feedback parameters to the cloud platform, wherein the data uploading process includes ensuring the security of the data through encrypted communication and performing data compression processing to improve transmission efficiency; Furthermore, the step of updating the fault diagnosis model and the protection mechanism model according to the historical operation data specifically includes: Based on the historical operation data stored in the cloud platform, the existing fault diagnosis model and protection mechanism model are optimized and updated, the accuracy of the model is trained and verified using historical data, and an optimized real-time protection control strategy is generated; The real-time protection control strategy includes adjustment parameters and protection actions for different circuit faults.

[0008] The key innovative features of the present invention include: (1) Real-time multi-dimensional signal analysis and fault diagnosis. This invention combines real-time analysis of multi-dimensional signals such as current, voltage, and frequency with fault diagnosis for the first time. It uses an adaptive algorithm and a multi-dimensional feature extraction model to accurately identify the fault type and location based on the actual operating status of the battery system. This innovation solves the problems of inaccurate diagnosis and delayed response in traditional circuit protection methods, allowing the battery system to respond quickly when a fault occurs, thereby improving the speed and accuracy of fault detection.

[0009] (2) Dynamically adjust the protection threshold and trigger the protection mechanism. The present invention adopts a strategy of dynamically adjusting the protection threshold. According to the fault diagnosis data and real-time signal characteristics, the threshold in the protection mechanism is automatically adjusted to trigger different protection actions, such as overload protection and short-circuit protection. Compared with the traditional fixed threshold method, this dynamic adjustment scheme based on actual operating data can better adapt to the battery protection requirements under different working conditions, and effectively improves the safety and adaptability of the battery system.

[0010] (3) Energy feedback path optimization. The present invention combines the protection status data with the cleaning signal data sequence and uses a dynamic adjustment algorithm to optimize the energy feedback path of the battery system. This innovation not only optimizes the energy flow during the battery charging and discharging process, but also enables the battery system to maximize energy feedback and reduce energy waste after the protection mechanism is triggered, significantly improving the energy efficiency of the system.

[0011] Beneficial effects of the present invention: (1) Improve the reliability and safety of the battery system. The present invention can quickly respond and trigger corresponding protection measures when an abnormality occurs in the battery system through real-time fault diagnosis and dynamic adjustment of protection thresholds, effectively avoiding damage to the battery caused by faults and significantly improving the reliability and safety of the battery system.

[0012] (2) Improve the energy efficiency of the battery system. By optimizing the energy feedback path, the present invention can maximize the energy recovery efficiency of the battery during the charging and discharging process, avoiding the problem of low battery efficiency caused by energy waste in traditional protection mechanisms, thereby improving the energy efficiency and economy of the entire battery system.

[0013] (3) Intelligent and adaptive optimization. The present invention realizes real-time data upload and continuous optimization of the model through the cloud platform, so that the battery protection control strategy can continuously adapt to different environments and load changes, forming a self-learning and adaptive mechanism. This not only improves the intelligence level of battery protection, but also enables the system to continuously optimize according to historical data and real-time operation conditions, and maintain an efficient and stable operating state. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 A schematic flow chart of a circuit protection control method for a circuit with the same input and output circuit provided in an embodiment of the present application; Figure 2 A structural block diagram of a circuit protection control method for the same input and output loop provided in an embodiment of the present application. DETAILED DESCRIPTION

[0015] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0016] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0017] Example 1: Reference Figure 1 , is a flow chart of a circuit protection control method for inputting and outputting the same circuit provided by an embodiment of the present invention, and the flow chart may at least include steps S100-S500: S100, obtaining current, voltage and frequency signals in the input and output circuits, and preprocessing and noise filtering the signals in combination with the temperature and humidity parameters of the circuit environment to generate a cleaned signal data sequence.

[0018] S200 , based on the cleaned signal data sequence, combining an adaptive algorithm with a multi-dimensional feature extraction model, analyzing the type and location of the loop fault to generate fault diagnosis data.

[0019] S300: Input the fault diagnosis data into a protection mechanism model, dynamically adjust the protection threshold of the model, trigger corresponding protection actions, and generate protection status data.

[0020] S400 , based on the protection status data and the cleaning signal data sequence, the energy feedback path of the loop is optimized in combination with a dynamic adjustment algorithm to generate optimized energy feedback parameters.

[0021] S500, uploading the cleaning signal data sequence, fault diagnosis data, protection status data and energy feedback parameters to the cloud platform, updating the fault diagnosis model and protection mechanism model in combination with historical operation data, and generating an optimized real-time protection control strategy.

[0022] Step S100 at least includes steps S110-S130: S110: Signal sampling and normalization processing. Obtain the current signal of the input and output ends in the loop , and voltage signal . At the same time, the input frequency is collected and output frequency , record signal changes based on time t.

[0023] By setting the sampling period Discretize the real-time signal to generate a sampling point sequence: in

[0024] Normalize the sampling sequence using the minimum and maximum values ​​of the input and output signals Calculate the normalized signal sequence: Normalized signal sequence Used for subsequent environmental correction.

[0025] S120: Signal environment correction. The temperature and humidity parameters of the input and output loop environments are collected through intelligent sensors, and the temperature signal T(t) and humidity signal H(t) are recorded to calibrate the loop state. The temperature and humidity signals are sampled based on time t to form a discrete sequence:

[0026] Normalized signal sequence and Perform environmental correction. Assume that the function of temperature affecting the signal is , the humidity influence function is , the corrected signal sequence is: Corrected signal sequence and It will be passed as input to the S130 module for noise filtering.

[0027] S130: Signal noise filtering. Apply an outlier detection algorithm (such as three sigma method or moving average) to remove abnormal signal points that exceed the threshold range: Outlier threshold: in is the signal mean, is the signal standard deviation.

[0028] A bandpass filter is used to process the signal after outliers are removed to eliminate high-frequency and low-frequency noise components and generate a cleaned signal sequence:

[0029] The filtered cleaned signal data The data is passed to the next module S200 for feature extraction and fault analysis.

[0030] Step S200 at least includes steps S210-S230: S210: Signal feature extraction. Receive the cleaning signal data sequence transmitted from the S130 module, including the current signal , voltage signal and frequency signal , by time is the index.

[0031] A feature extraction model is applied to the cleaning signal data to calculate feature values ​​in the time domain and frequency domain.

[0032] X stands for or , N is the total number of sampling points.

[0033] Arrange the extracted eigenvalues ​​by dimension and construct the signal feature matrix:

[0034] S220: Fault type analysis. The signal feature matrix generated by S210 is Enter the adaptive fault classification model.

[0035] Adaptive algorithms are used to analyze abnormal patterns in the feature matrix. Specifically, the signal feature matrix is ​​classified into fault categories using distance measurement methods (such as Euclidean distance) or neural network-based classification models. Assume that the fault type output by the classification model is:

[0036] in, Indicates the fault type label, such as overload, short circuit, or ground fault.

[0037] Fault type data includes fault type labels and corresponding characteristic indicators , which is used to locate the fault location later.

[0038] S230: Fault location: Receive the fault type data T and signal feature matrix F generated in S220.

[0039] Based on fault type data and the feature matrix , combined with the node distribution of the input and output circuits, the location of the fault is calculated using the changes in node electrical parameters (such as current gradient or voltage difference). The positioning formula is: in Indicates the fault location. is the measured value, is the reference value.

[0040] Fault diagnosis data including fault location , Fault type and related characteristic indicators The combined data set is used to trigger the subsequent protection mechanism.

[0041] Step S300 at least includes steps S310-S330: S310: Analyze the fault type and determine the protection strategy. Receive the fault diagnosis data generated from S230, including the fault type , signal feature matrix and fault location .

[0042] Through the protection mechanism model, for the fault type (such as overload fault, short circuit fault, ground fault) and fault location , analyze the protection requirements and calculate the initial protection strategy. Suppose the protection strategy set is:

[0043] in, Indicates protection action under specific fault conditions, such as overload current limiting, circuit breaker disconnection, or grounding shielding.

[0044] According to the signal feature matrix The characteristic distribution (such as the peak current and voltage mean ), mapping the protection parameters corresponding to the protection action : in, It is the triggering threshold of the protection action (such as current limit, time delay parameter, etc.).

[0045] S320: Dynamically adjust the protection threshold and trigger the protection action. and protection policy collection , dynamically adjust the protection threshold Specifically, according to the adaptive adjustment rules in the protection mechanism model, the adjusted protection threshold is calculated: ;in, Represents the incremental adjustment function of the fault feature to the threshold.

[0046] By monitoring the real-time input signal , when the protection condition is met When the corresponding protection action is triggered For example: Overload protection: limit current to .

[0047] Short circuit protection: disconnects the circuit path.

[0048] Ground fault protection: switches to alternate ground path.

[0049] S330: Generate and transmit protection status data. Collect relevant status information of protection action execution, including trigger time , Trigger conditions , protection action , constitute the protection status data:

[0050] The protection status data generated Passed to the energy optimization module for subsequent energy feedback path adjustment.

[0051] Step S400 at least includes steps S410-S430: S410: Analyze the energy feedback path. Obtain protection status data from S330 , and the cleaning signal data sequence obtained from S130, including current and frequency .

[0052] Based on the trigger time in the protection status data and trigger protection actions , before and after extraction protection The cleaning signal sequence is used to calculate the corresponding energy distribution information:

[0053] in, and Represent the energy values ​​of the input and output circuits respectively, Indicates the combined fault location The energy transfer function.

[0054] According to the energy distribution information Combine the loop topology diagram to identify the current energy feedback path and generate a path distribution matrix in Represents the distribution weight of energy from node i to node j.

[0055] S420: Optimize the energy feedback path. Receive the path distribution matrix generated in S410 and energy distribution information .

[0056] Using the dynamic adjustment algorithm, the weights in the path distribution matrix are optimized and adjusted according to the input energy and output demand to generate the optimized path matrix. The adjustment amount Calculated based on protection status data and input signal characteristics: in, It is a dynamic adjustment function.

[0057] The optimized path matrix Applied in the circuit to redistribute input energy To the feedback path, ensure the stability of the loop operation.

[0058] S430: Generate optimized energy feedback parameters. Based on the optimized path matrix and energy distribution , calculate the energy feedback parameter set of the loop: is the optimized energy feedback value, and the calculation formula is: ; Feedback parameters of generated energy Applied to the circuit, it dynamically adjusts the energy flow and ensures efficient energy transfer.

[0059] The optimized energy feedback parameters are transferred to the cloud platform module for subsequent data storage and analysis.

[0060] Step S500 at least includes steps S510-S530: S510: Upload data to the cloud platform for storage. Arrange data from the previous module, including: Clean signal data sequence: from S130, including Fault diagnosis data: from S230, including fault type , Fault location Sum signal characteristic matrix ; Protection status data: from S330, including trigger time Protection Action , adjusted threshold Optimized energy feedback parameters: from S430, including optimized path matrix and feedback energy .

[0061] Structuring the above data into a unified format , defined as: Using secure communication protocols to transfer data Upload to the cloud platform storage system as input for model optimization and historical data analysis.

[0062] S520: Model training and optimization. Call historical operation data from the cloud platform, including historical cleaning signals, fault characteristics and protection status data: Based on uploaded data and historical data , train the fault diagnosis model and protection mechanism model, and use optimization algorithms (such as gradient descent or reinforcement learning) to generate an optimized set of model parameters:

[0063] in are the optimized fault diagnosis model parameters, are the optimized protection mechanism model parameters.

[0064] The optimized model parameter set Organize and store them in the cloud platform for subsequent use.

[0065] S530: Feedback the optimized model parameters to the local system. Synchronize from cloud platform to local system, including fault diagnosis model and protection mechanism model .

[0066] Update the fault diagnosis and protection mechanism model of the local system using the optimized model parameters:

[0067] Combined with the optimized local model parameters, a real-time protection control strategy is generated based on the cleaning signal and real-time data , defined as:

[0068] The key innovative features of the present invention include: (1) Real-time multi-dimensional signal analysis and fault diagnosis. This invention combines real-time analysis of multi-dimensional signals such as current, voltage, and frequency with fault diagnosis for the first time. It uses an adaptive algorithm and a multi-dimensional feature extraction model to accurately identify the fault type and location based on the actual operating status of the battery system. This innovation solves the problems of inaccurate diagnosis and delayed response in traditional circuit protection methods, allowing the battery system to respond quickly when a fault occurs, thereby improving the speed and accuracy of fault detection.

[0069] (2) Dynamically adjust the protection threshold and trigger the protection mechanism. The present invention adopts a strategy of dynamically adjusting the protection threshold. According to the fault diagnosis data and real-time signal characteristics, the threshold in the protection mechanism is automatically adjusted to trigger different protection actions, such as overload protection and short-circuit protection. Compared with the traditional fixed threshold method, this dynamic adjustment scheme based on actual operating data can better adapt to the battery protection requirements under different working conditions, and effectively improves the safety and adaptability of the battery system.

[0070] (3) Energy feedback path optimization. The present invention combines the protection status data with the cleaning signal data sequence and uses a dynamic adjustment algorithm to optimize the energy feedback path of the battery system. This innovation not only optimizes the energy flow during the battery charging and discharging process, but also enables the battery system to maximize energy feedback and reduce energy waste after the protection mechanism is triggered, significantly improving the energy efficiency of the system.

[0071] Beneficial effects of the present invention: (1) Improve the reliability and safety of the battery system. The present invention can quickly respond and trigger corresponding protection measures when an abnormality occurs in the battery system through real-time fault diagnosis and dynamic adjustment of protection thresholds, effectively avoiding damage to the battery caused by faults and significantly improving the reliability and safety of the battery system.

[0072] (2) Improve the energy efficiency of the battery system. By optimizing the energy feedback path, the present invention can maximize the energy recovery efficiency of the battery during the charging and discharging process, avoiding the problem of low battery efficiency caused by energy waste in traditional protection mechanisms, thereby improving the energy efficiency and economy of the entire battery system.

[0073] (3) Intelligent and adaptive optimization. The present invention realizes real-time data upload and continuous optimization of the model through the cloud platform, so that the battery protection control strategy can continuously adapt to different environments and load changes, forming a self-learning and adaptive mechanism. This not only improves the intelligence level of battery protection, but also enables the system to continuously optimize according to historical data and real-time operation conditions, and maintain an efficient and stable operating state.

[0074] Embodiment 2: Figure 2 FIG. 2 is a block diagram showing a circuit protection control method for a circuit with the same input and output loop according to an embodiment of the present invention. Figure 2 As shown, the structure may include: The data acquisition module 10 is used to obtain signal data related to the same input and output loop from the circuit system in real time, including current, voltage, frequency and environmental data (such as temperature and humidity). The module obtains signal data through high-precision sensors and ensures the timeliness and accuracy of the data. Specifically: Current signal acquisition: obtain current data of input circuit and output circuit; Voltage signal acquisition: obtain voltage data of input circuit and output circuit; Frequency signal acquisition: obtain the frequency signals of input and output circuits; Environmental data collection: Acquire environmental temperature and humidity data for subsequent signal correction.

[0075] The data acquisition module ensures the high quality of signal data and provides the necessary basis for subsequent signal preprocessing, feature extraction and model analysis.

[0076] The signal preprocessing module 20 preprocesses the collected current, voltage, frequency and environmental data, including data denoising, normalization, environmental correction and signal cleaning. Specifically: Signal denoising: Filter the noise of the collected signals, remove abnormal data, and ensure the reliability of the data; Signal normalization: normalize the current and voltage signals to eliminate the dimension difference and facilitate subsequent analysis; Environmental correction: Combined with the ambient temperature and humidity data, signal correction is performed to eliminate the impact of environmental factors on the signal.

[0077] The processed signal data will be passed as input to the subsequent feature extraction module for further analysis.

[0078] The feature extraction and analysis module 30 uses a multi-dimensional feature extraction algorithm to extract key signal features based on the processed signal data for subsequent fault diagnosis and protection mechanism application. Specifically: Feature extraction: extract features from current, voltage, frequency and other signals, such as mean, variance, peak value, spectrum center, etc.; Fault pattern recognition: Use machine learning or statistical analysis methods to classify signal features and identify normal and fault signals; Feature screening: Based on the relevance and predictive power of features, the most valuable feature data for fault diagnosis is screened out.

[0079] The extracted feature data will be input into the fault diagnosis module for further analysis.

[0080] The fault diagnosis and protection decision module 40 detects and classifies the faults in the circuit using an adaptive algorithm based on the signal feature data transmitted from the feature extraction and analysis module, and triggers the corresponding protection strategy according to the fault type. Specifically: Fault diagnosis: Combine feature matrix with machine learning algorithm to analyze fault types (such as overload, short circuit, ground fault, etc.); Protection strategy determination: Dynamically determine the protection strategy based on the fault type, such as starting overload protection, short circuit protection or grounding protection; Protection threshold adjustment: According to the fault diagnosis results, adjust the protection threshold in the protection mechanism to trigger the corresponding protection action.

[0081] The fault diagnosis data and protection decision results will be passed to the energy optimization module for subsequent processing.

[0082] The energy feedback path optimization module 50 optimizes the energy feedback path through a dynamic adjustment algorithm based on the fault diagnosis data and the cleaning signal data. Specifically: Energy distribution analysis: Analyze the energy flow and distribution in the circuit, and calculate the energy values ​​of the input and output circuits; Path optimization: Use optimization algorithms to adjust the energy distribution of the feedback path to ensure that energy is effectively used during the protection process; Feedback parameter generation: Generate optimized energy feedback parameters to ensure that the circuit can maintain efficient operation during the protection process.

[0083] The optimized energy feedback parameters will be transmitted to the cloud platform module for data storage and model updating.

[0084] The cloud platform data storage and optimization module 60 is used to store the uploaded signal data, fault diagnosis data, protection status data and optimized energy feedback parameters, and update the fault diagnosis model and protection mechanism model in combination with the historical operation data. Specifically: Data upload: Upload cleaning signal data, fault diagnosis data, protection status data and optimized energy feedback parameters to the cloud platform; Model training and optimization: Based on the uploaded data and historical operation data, the fault diagnosis model and protection mechanism model are trained and optimized to generate optimized model parameters; Real-time strategy generation: Generates real-time protection control strategies based on optimized model parameters and feeds them back to the local system.

[0085] The cloud platform module supports the system's self-learning capabilities, enabling continuous optimization of protection and control strategies and improving the intelligence level of the circuit system.

[0086] The local system update and feedback module 70 is responsible for receiving the optimization model parameters returned from the cloud platform and applying them to the local fault diagnosis and protection mechanism model. Specifically: Model parameter feedback: Feedback the optimized model parameters updated by the cloud platform to the local system; Real-time protection strategy execution: Based on the updated fault diagnosis and protection mechanism model, generate and execute real-time protection control strategies to ensure the safe operation of the circuit system; Data feedback and optimization: Feedback the local system operation data to the cloud platform for subsequent model optimization.

[0087] The present invention provides a circuit protection control method for inputting and outputting the same circuit, which has the following beneficial effects: Efficient energy utilization: By optimizing the energy feedback path, the effective utilization of energy in the circuit is maximized, and the energy efficiency of the circuit system is improved; Intelligent fault detection and protection: Combining multi-dimensional feature analysis and adaptive algorithms, it can quickly and accurately diagnose the fault type and automatically trigger the corresponding protection mechanism, significantly improving the reliability and safety of the circuit; Self-learning and dynamic optimization: Through the cloud platform, historical data and real-time data are continuously learned and optimized, and the system can adjust the protection strategy in real time to adapt to the complex and changing working environment; Real-time and stability: Combined with data interaction between the local system and the cloud platform, real-time fault diagnosis and protection decision-making can be achieved to ensure the stable operation of the circuit system under various working conditions.

[0088] In general, the present invention realizes the intelligence, automation and high efficiency of circuit protection control, and is applicable to various power electronic equipment and industrial automation systems with high reliability requirements.

[0089] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A circuit protection control method for input and output of the same circuit, characterized in that: The steps include: Acquire the current, voltage and frequency signals in the input and output circuits, and perform preprocessing and noise filtering on the signals in combination with the temperature and humidity parameters of the circuit environment to generate a cleaned signal data sequence; Based on the cleaned signal data sequence, combining an adaptive algorithm with a multi-dimensional feature extraction model, analyzing the type and location of the loop fault to generate fault diagnosis data; Inputting the fault diagnosis data into a protection mechanism model, dynamically adjusting the protection threshold of the model, triggering corresponding protection actions, and generating protection status data; Based on the protection status data and the cleaning signal data sequence, the energy feedback path of the loop is optimized in combination with a dynamic adjustment algorithm to generate optimized energy feedback parameters; The cleaning signal data sequence, fault diagnosis data, protection status data and energy feedback parameters are uploaded to the cloud platform, and the fault diagnosis model and protection mechanism model are updated in combination with the historical operation data to generate an optimized real-time protection control strategy.

2. The circuit protection control method for input and output of the same circuit according to claim 1, characterized in that: Before the step of obtaining the current, voltage and frequency signals in the input and output loops, the method further includes: The real-time current and voltage signals in the loop are collected by sensors, and the signals are preliminarily normalized in combination with the collected ambient temperature and humidity data to obtain a preliminary signal sequence.

3. The circuit protection control method for input and output of the same circuit according to claim 2, characterized in that: The steps of signal preprocessing and noise filtering specifically include: The preliminary signal sequence is filtered to remove abnormal signal points, and the collected current, voltage and frequency signals are denoised to generate a cleaned signal data sequence.

4. The circuit protection control method for input and output of the same circuit according to claim 3, characterized in that: The step of performing fault analysis based on the cleaned signal data sequence specifically includes: In combination with the signal data sequence, a multidimensional feature extraction model is used to extract the time domain features and frequency domain features of the signal to generate a signal feature matrix; The signal feature matrix is ​​classified and analyzed through an adaptive algorithm to determine the fault type and location in the circuit.

5. The circuit protection control method for input and output of the same circuit according to claim 4, characterized in that: The fault diagnosis data includes: Label of the type of fault that occurred in the circuit in question , the specific location where the circuit fault occurs , and the signal feature matrix related to the fault .

6. The circuit protection control method for input and output of the same circuit according to claim 1, characterized in that: The step of triggering the corresponding protection action specifically includes: According to the fault type and fault location data, trigger the corresponding protection mechanism and adjust the protection threshold in the protection model to adapt to the current working state of the circuit; Among them, the protection mechanisms include overload protection, short circuit protection and grounding protection.

7. The circuit protection control method for input and output of the same circuit according to claim 6, characterized in that: The protection status data includes: Time to trigger protection action , the type of protection action triggered , the adjusted protection threshold , and the protection action execution status.

8. The circuit protection control method for input and output of the same circuit according to claim 1, characterized in that: The step of optimizing the feedback path specifically includes: Combining the cleaning signal data sequence with the protection status data, analyzing the energy flow between the input and output in the loop through a dynamic adjustment algorithm, optimizing the energy feedback path, and generating new energy feedback parameters; The energy feedback path optimization algorithm adjusts the feedback path according to the operating state of the loop to reduce energy loss and improve the efficiency of the circuit system.

9. The circuit protection control method for input and output of the same circuit according to claim 1, characterized in that: The steps of uploading to the cloud platform specifically include: The cleaning signal data sequence, fault diagnosis data, protection status data and optimized energy feedback parameters are uploaded to the cloud platform. The data uploading process includes ensuring the security of the data through encrypted communication and performing data compression processing to improve transmission efficiency.

10. The circuit protection control method for input and output of the same circuit according to claim 1, characterized in that: The step of updating the fault diagnosis model and the protection mechanism model according to the historical operation data specifically includes: Based on the historical operation data stored in the cloud platform, the existing fault diagnosis model and protection mechanism model are optimized and updated, the accuracy of the model is trained and verified using historical data, and an optimized real-time protection control strategy is generated; The real-time protection control strategy includes adjustment parameters and protection actions for different circuit faults.

Images