Bypass flexible cable detection method and system

Abstract

The invention provides a bypass flexible cable detection method and system, and belongs to the technical field of cable fault detection, and the method comprises the steps: collecting the current data of a bypass flexible cable through a data collection module; the collected current data is transmitted to a cloud server through a data transmission module; the cloud server receives, stores and processes the current data, and sends the current data to the deep learning module; the deep learning module stores a pre-trained cable fault detection model, and performs fault detection on the bypass flexible cable based on the pre-trained cable fault detection model; and remotely checking the fault detection result of the bypass flexible cable through a mobile terminal application program. Data can be timely transmitted to the deep learning module for analysis, and once an abnormality is detected, the mobile terminal can immediately receive fault information, so that operation and maintenance personnel can know and process the fault at the first time, the influence of the fault on a power system is reduced, and the stability of power supply is ensured.

Description

Technical Field

[0001] The present invention belongs to the technical field of cable fault detection, and in particular relates to a bypass flexible cable detection method and system. Background Art

[0002] With the development of distribution network operation technology in the power industry, the bypass cable bears the load current during bypass system operation. The magnitude of the current fluctuates with the load used by the downstream power customers. If the current exceeds the rated current carrying value of the bypass cable, the consequences may be tripping and power outage at the least, or equipment damage and personal injury at the worst.

[0003] The current detection system currently configured in the existing bypass system can only measure the current of the bypass cable once every 30 minutes in accordance with the provisions of the distribution network operation technical guidelines. It cannot provide real-time warning of faults. If no detection is detected within the time limit, on-site staff will not be able to take timely response measures, which poses certain safety risks.

[0004] To this end, the present invention provides a bypass flexible cable detection method and system. Summary of the Invention

[0005] The present invention provides a bypass flexible cable detection method and system to at least solve the problem in the prior art that workers need to go to the site to detect the bypass system. If the detection is not completed within a certain time limit, the on-site workers cannot take timely countermeasures, which poses certain safety risks.

[0006] In a first aspect, an embodiment of the present application further provides a bypass flexible cable detection method, the method comprising: Step S1: collecting current data of the bypass flexible cable through a data acquisition module; Step S2: transmitting the collected current data to the cloud server through the data transmission module; Step S3: The cloud server receives, stores, and processes the current data, and sends the current data to the deep learning module; Step S4: the deep learning module stores a pre-trained cable fault detection model, and performs fault detection on the bypass flexible cable based on the pre-trained cable fault detection model; Step S5: remotely viewing the fault detection result of the bypass flexible cable through the mobile terminal application.

[0007] Furthermore, the construction and training process of the cable fault detection model is as follows: Step S41: Input current data Perform normalization processing, and the expression of normalization processing is:

[0008] in, Represents the normalized current data, whose value is [0,1]. represents the minimum value of the current data in the training data set, Represents the maximum value of the current data in the training data set; Step S42: Selecting a cable fault detection model to be trained and constructing a cable fault detection model; Step S43: Use historical current data with labels indicating whether the cable is overloaded to train the model and define a loss function; Step S44: adjusting the weights and biases of the model through a back-propagation algorithm to minimize the loss function, thereby training a cable fault detection model that can accurately determine whether the cable is overloaded; Step S45: Parameter update: Based on the calculated gradient, the stochastic gradient descent optimization algorithm is used to update the weights and biases of the model; Step S46: Iterative training: Repeat steps S42 to S45 until the loss function converges to a stable value or reaches a preset number of training rounds, thereby obtaining a trained cable fault detection model.

[0009] Furthermore, in step S42, the specific process of model construction is as follows: Step S421: Calculate the forget gate, which is expressed as:

[0010] Where, represents the Sigmoid activation function, represents the hidden layer The weight matrix of the output of each neuron to the forget gate of the output layer, Indicates that the hidden state of the previous moment and the current input To splice, Represents the bias term of the forget gate; Step S422: Calculate the input gate, which is expressed as:

[0011] Where, is the hidden layer The weight matrix of the output of each neuron to the input gate of the output layer; Step S423: Calculate candidate memory cells, the expression is:

[0012] Where, is the hidden layer The output of each neuron is used to generate a weight matrix for candidate memory units in the output layer; Represents candidate memory units; is the bias term of the input gate; is the bias term used to generate candidate memory units; represents the input gate; is the hyperbolic tangent activation function; Step S424: Update the memory unit, the expression is:

[0013] Where, Represents element-wise multiplication; Indicates the current time The state of the memory unit; Indicates the last moment The state of the memory unit; Step S425: Calculate the output gate, which is expressed as:

[0014] Where, represents the output gate; represents the hidden layer The weight matrix of the output of each neuron to the output gate of the output layer, Represents the bias term of the output gate; Step S426: Calculate the hidden state, which is expressed as:

[0015] Indicates the current time The hidden state of .

[0016] Furthermore, the expression of the pre-trained cable fault detection model is:

[0017] Where, represents the predicted value output by the cable fault detection model; represents the Sigmoid activation function; represents the number of neurons in the hidden layer; represents the hidden layer The output of each neuron has a weight matrix for the output layer; represents the output gate; Indicates the current time The state of the memory unit; represents the bias term.

[0018] Furthermore, the collected current data is transmitted to the cloud server through the data transmission module, including the following specific steps: The data transmission module uses modulation technology to modulate the current data; The data transmission module transmits the modulated current data to the cloud server via a wireless network.

[0019] Furthermore, the cloud server receives, stores, and processes the current data, and sends the current data to the deep learning module, including the following specific steps: After receiving the data, the cloud server demodulates it and restores the original current data; The cloud server inputs the recovered raw current data into the deep learning module; The demodulation process includes: Mixing processing and low-pass filtering processing.

[0020] Furthermore, the bypass flexible cable fault detection results can be viewed remotely through the mobile terminal application, including: When an overload fault is detected in the output of the cable fault detection model, the fault detection result is sent via a mobile terminal application; The mobile terminal application displays cable operating status, current data and overload fault information in real time.

[0021] In a second aspect, an embodiment of the present application provides a bypass flexible cable detection system applied to the bypass flexible cable detection method as described in the above aspects, the system comprising: a data acquisition module, a data transmission module, a cloud service module, a deep learning module and a mobile terminal; The data acquisition module is used to collect current data of the bypass flexible cable; The data transmission module is used to transmit the collected current data to the cloud server; The cloud server is used to receive, store and process current data; The deep learning module is used to store the trained cable fault detection model for fault detection of bypass flexible cables; The mobile terminal supports a mobile terminal application for remotely viewing the fault detection results of the bypass flexible cable.

[0022] Furthermore, the data acquisition module includes: a current transformer and a data acquisition unit; The current transformer is installed on the bypass flexible cable to collect the current data of the bypass flexible cable in real time; The data acquisition unit is connected to the current transformer and is used to acquire current data collected by the current transformer; Furthermore, the current transformer is one of the following current transformers: Open-type current transformer, through-type current transformer.

[0023] It can be seen from the above technical solutions that the present invention has the following advantages: In the bypass flexible cable detection method and system provided in the present application, current data is collected in real time through the data acquisition module, and in conjunction with the data transmission module, cloud service module, deep learning module and mobile terminal, the data can be transmitted to the deep learning module for analysis in a timely manner. Once an abnormality is detected, the mobile terminal can immediately receive the fault information, allowing the operation and maintenance personnel to know and handle the fault in the first time, reducing the impact of the fault on the power system and ensuring the stability of the power supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for the description. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0025] Figure 1 Flowchart of the bypass flexible cable detection method. DETAILED DESCRIPTION

[0026] The bypass flexible cable detection method and system will be described in detail below, and various embodiments of the present disclosure will be described more fully. The present disclosure is capable of various embodiments, and modifications and variations may be made therein. However, it should be understood that there is no intention to limit the various embodiments of the present disclosure to the specific embodiments disclosed herein, but rather that the present disclosure is to encompass all modifications, equivalents, and / or alternatives falling within the spirit and scope of the various embodiments of the present disclosure.

[0027] Hereinafter, the terms "include" or "may include" as used in various embodiments of the present disclosure indicate the presence of disclosed functions, operations, or elements, and do not limit the addition of one or more functions, operations, or elements. In addition, as used in various embodiments of the present disclosure, the terms "include," "have," and their cognates are intended only to indicate specific features, numbers, steps, operations, elements, components, or combinations of the foregoing, and should not be understood as excluding the presence of one or more other features, numbers, steps, operations, elements, components, or combinations of the foregoing, or the possibility of adding one or more features, numbers, steps, operations, elements, components, or combinations of the foregoing.

[0028] In various embodiments of the present disclosure, the expression "or" or "at least one of A or / and B" includes any or all combinations of the words listed simultaneously. For example, the expression "A or B" or "at least one of A or / and B" may include A, may include B, or may include both A and B.

[0029] The expressions (such as "first", "second", etc.) used in the various embodiments of the present disclosure may modify the various constituent elements in the various embodiments, but may not limit the corresponding constituent elements. For example, the above expressions do not limit the order and / or importance of the elements. The above expressions are only used to distinguish one element from other elements. For example, a first user device and a second user device indicate different user devices, although both are user devices. For example, without departing from the scope of the various embodiments of the present disclosure, a first element may be referred to as a second element, and similarly, a second element may also be referred to as a first element.

[0030] It should be noted that when a component is described as being “connected” to another component, the first component may be directly connected to the second component, and a third component may be “connected” between the first and second components. Conversely, when a component is described as being “directly connected” to another component, it can be understood that there is no third component between the first and second components.

[0031] The term “user” used in various embodiments of the present disclosure may indicate a person who uses an electronic device, and may be a monitoring person, a test person, or an operator.

[0032] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0033] The embodiments of the present application provide a bypass flexible cable detection method and system to solve the current urgent need for a technical problem in which staff are required to go to the site to detect the bypass system. If the detection is not completed within the time limit, the on-site staff cannot take timely response measures, which poses certain safety risks.

[0034] The technical solutions proposed in the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0035] An embodiment of the present application provides a bypass flexible cable detection method, the method comprising: S1: Collect current data of the bypass flexible cable through the data acquisition module.

[0036] The current transformer is installed on the bypass flexible cable to collect current data of the bypass flexible cable in real time, and the data acquisition unit acquires the current data collected by the current transformer.

[0037] Among them, the expression of the collected current data is:

[0038] Where, express Current data collected at all times; express Cable current at every moment; Represents the characteristic function of the current transformer. The current transformer collects the current data of the bypass flexible cable in real time. The data acquisition unit performs analog-to-digital conversion and signal processing on the collected current data. The data acquisition unit transmits the processed current data to the data transmission module via the data bus.

[0039] S2: Transmit the collected current data to the cloud server through the data transmission module; The data transmission module uses modulation technology to modulate the current data. The expression is as follows:

[0040] Where: represents the modulated signal after modulation; Indicates the signal amplitude; Indicates the carrier frequency; Indicates modulation sensitivity; indicates the change in carrier instantaneous phase caused by a change in unit current data; express Current data collected at all times; express The phase of the moment.

[0041] The data transmission module transmits the modulated current data to the cloud server via a wireless network.

[0042] S3: The cloud server receives, stores, and processes the current data, and sends the current data to the deep learning module; After receiving the data, the cloud server demodulates it and recovers the original current data; the recovered original current data is input into the deep learning module; The demodulation process includes: Mixing processing, its expression is:

[0043] Where, represents the mixed signal; where, middle, is the signal amplitude, which determines the strength of the signal; The carrier frequency is the high-frequency signal frequency used to carry the original signal during the modulation process; Indicates the modulation sensitivity, reflecting the raw current data Ability to modulate carrier phase; yes The current data collected at all times is collected by the current transformer, and its expression is , for Cable current at the moment, is the current transformer characteristic function; for The phase at the moment affects the initial phase of the carrier. During demodulation, the received modulated signal With local carrier Multiply to achieve spectrum shifting, moving the spectrum of the modulated signal from the carrier frequency Move nearby to a low frequency band for subsequent processing.

[0044] Low-pass filtering, its expression is:

[0045] Where, is the low-pass filter function; is the original current data restored after low-pass filtering. The mixed signal Contains multiple frequency components. Through the low-pass filter, high-frequency noise and unnecessary frequency components are filtered out, and only the low-frequency original current data related signals are retained, thereby restoring the original current data. Here we assume that the low-pass filter function is , which can extract The baseband signal components in . It is the original current data restored after low-pass filtering. The mixed signal Contains multiple frequency components. Through the low-pass filter, high-frequency noise and unnecessary frequency components are filtered out, and only the low-frequency original current data related signals are retained, thereby restoring the original current data. The cloud server will restore the original current data Input into the deep learning module, the deep learning module Perform analysis to determine whether the cable is overloaded.

[0046] S4: The deep learning module stores a pre-trained cable fault detection model and performs fault detection on the bypass flexible cable based on the pre-trained cable fault detection model; Among them, the expression of the pre-trained cable fault detection model is:

[0047] Where, represents the predicted value output by the cable fault detection model; represents the Sigmoid activation function; represents the number of neurons in the hidden layer; represents the hidden layer The output of each neuron has a weight matrix for the output layer; represents the output gate; Indicates the current time The state of the memory unit; represents the bias term.

[0048] Current moment The expression of the memory cell state is:

[0049] Where, represents the forget gate, Indicates the last moment The memory cell state, represents the input gate, Represents a candidate memory unit.

[0050] The expression of the forget gate is:

[0051] Where, represents the hidden layer The weight matrix of the output of each neuron to the forget gate of the output layer, represents the bias term of the forget gate, Indicates the hidden state at the previous moment, Indicates the current data being input at the current moment.

[0052] The expression of the input gate is:

[0053] Where, represents the hidden layer The weight matrix of the output of each neuron to the input gate of the output layer, Represents the bias term of the input gate.

[0054] The expression of candidate memory unit is:

[0055] Where, represents the candidate memory unit, represents the hidden layer The output of each neuron is used to generate the weight matrix of candidate memory units in the output layer. Represents the bias term used to generate candidate memory cells.

[0056] The expression of the output gate is:

[0057] Where, represents the hidden layer The weight matrix of the output of each neuron to the output gate of the output layer, Represents the bias term of the output gate.

[0058] The current data Input into the trained cable fault detection model and output the predicted value ; If the predicted value , it is judged that the cable is overloaded; if , then it is judged that the cable is not overloaded; Indicates the threshold value.

[0059] S5: Remotely view the fault detection results of the bypass flexible cable through the mobile terminal application.

[0060] The cable fault detection model outputs the fault detection results of the bypass flexible cable and sends them to a mobile terminal application. When an overload fault occurs in the bypass flexible cable, relevant personnel can view the fault information through the mobile terminal application and take timely measures.

[0061] The mobile terminal application displays cable operating status, current data, and overload fault information in real time. The mobile terminal application supports remote viewing, allowing staff to monitor cable operation anytime and anywhere.

[0062] In this embodiment, the construction and training process of the cable fault detection model is as follows: Step 1: Normalization: Since the collected current data may have different magnitudes and distribution ranges, in order to improve the training efficiency and accuracy of the cable fault detection model, the input current data needs to be normalized. Perform normalization processing, and the expression of normalization processing is:

[0063] in, represents the normalized current data, represents the minimum value of the current data in the training data set, Represents the maximum value of the current data in the training dataset; by normalizing, the current data is mapped to the [0,1] interval, making it easier for the cable fault detection model to learn the characteristics of the data.

[0064] Step 2: Forward propagation: Select the cable fault detection model to be trained. The specific process of cable fault detection model construction is as follows: Calculate the forget gate, its expression is:

[0065] Where, represents the Sigmoid activation function, represents the hidden layer The weight matrix of the output of each neuron to the forget gate of the output layer, Indicates that the hidden state of the previous moment and the current input To splice, Represents the bias term of the forget gate; the forget gate determines which information in the memory unit at the previous moment can be retained to the current moment.

[0066] Calculate the input gate, its expression is:

[0067] Where, is the hidden layer The weight matrix of the output of each neuron to the input gate of the output layer. During the calculation of the input gate, Responsible for the hidden state of the previous moment and the current input Perform linear transformation to determine the influence of the current input data and the hidden state at the previous moment on the input gate. Different weight values determine the importance of different input information. The weights are continuously adjusted during model training. value, so that the model can learn the optimal weight combination, thereby better controlling the situation where new information enters the memory unit at the current moment.

[0068] Calculate the candidate memory unit, the expression is:

[0069] Where, is the hidden layer The output of each neuron is used to generate a weight matrix for candidate memory units in the output layer; Represents candidate memory units; weight matrix used to generate candidate memory units Similarly, the hidden state of the previous moment and the current input Perform linear transformation to generate candidate memory units , the candidate memory unit Contains the new information that needs to be added to the memory unit at the current moment, The value of is continuously optimized during the training process to generate candidate memory content that better meets the model requirements; is the bias term of the input gate, the bias term of the input gate A fixed offset is introduced into the calculation of the input gate, which is independent of the input data and hidden state. The role of the bias term is to add additional learning ability to the model, helping the model to better fit the data, avoiding underfitting during the training process, and making the input gate more flexible in adjusting the output during calculation; It is a bias term used to generate candidate memory units. It provides a fixed offset for the calculation of generating candidate memory units, helping the model better capture data features, optimize the generation process of candidate memory units, and improve model performance; Represents the input gate, which is a result calculated by the Sigmoid function and has a value range between 0 and 1. It plays the role of a control gate, and its output value determines how much new information at the current moment can enter the memory unit. When it is close to 1, it means that new information at the current moment can enter the memory unit in large quantities; when When it is close to 0, it means that new information at the current moment can rarely enter the memory unit. is the hyperbolic tangent activation function.

[0070] Update the memory unit, the expression is:

[0071] Where, Represents element-level multiplication; updates the content of the memory unit through the synergy of the forget gate and the input gate; Indicates the current time The memory unit state at the current moment The information of the memory unit at the previous moment is integrated (through the forget gate filter retention) and the new information at the current moment (through the input gate Control entry) is the result of the model fusing historical information with current information. This state is continuously updated as time steps advance, and is used to store and transmit long-term dependency information in time series data. It is crucial for the model to understand the context when processing sequence data. In the cable fault detection model that determines whether the cable is overloaded, The important features related to the time series of current data are preserved, which helps the model comprehensively consider historical current changes and thus more accurately judge the current cable status; Indicates the last moment The memory unit state is one of the important bases for updating the memory unit at the current moment. Memory unit with the current moment Related, forget gate Decided Which information in the memory can be retained to the current moment, so that the model can selectively forget or retain historical information, avoid the accumulation of too much useless information in the memory unit, and ensure that the model can focus on historical data features that are valuable for current judgment.

[0072] Calculate the output gate, its expression is:

[0073] Where, represents the output gate; It is a vector with a value between 0 and 1, calculated by the Sigmoid function. The function of the output gate is to control the output of information in the memory unit and determine the state of the memory unit. Which information in will be used to generate the output at the current moment ,In the cable fault detection model to determine whether the cable is overloaded, Determines how much of the current data stored in the memory cell can be used to generate the output required to ultimately determine the cable status. By screening the memory cell information, the model output becomes more targeted and effective. represents the hidden layer The weight matrix of the output of each neuron to the output gate of the output layer, Represents the bias term of the output gate.

[0074] Calculate the hidden state, its expression is:

[0075] Indicates the current time The hidden state of the model; it is the output of the model at the current moment, which integrates the state of the memory unit The information selected by the output gate and the input information at the current moment (integrated into the memory unit through a series of calculations such as the previous input gate and forget gate). In the cable overload judgment task, This includes the model's understanding and integration of current current data and historical data characteristics, which is an important basis for subsequent judgment of cable overload. The model's final prediction result is often obtained through further processing based on hidden states, which carries the model's dynamic understanding and expression of the input data time series.

[0076] Step 3: Calculate the loss: Use historical current data with labels indicating whether the cable is overloaded to train the model and define the loss function, which is expressed as:

[0077] Where, represents the number of samples, represents the true label, where Indicates overload, Indicates no overload. Represents the label predicted by the model; calculates the error between the model prediction value and the true label. The value of the loss function reflects the current prediction accuracy of the model.

[0078] Step 4: Adjust the weights and biases of the model through the back-propagation algorithm to minimize the loss function, thereby training a cable fault detection model that can accurately determine whether the cable is overloaded.

[0079] The back-propagation algorithm is used to adjust the model's weights and biases to minimize the loss function, thereby training a cable fault detection model that can accurately determine whether the cable is overloaded. The specific steps include the following: Back propagation: The gradient of the loss function to the parameters (weight matrix and bias term) of the cable fault detection model is calculated by the back propagation algorithm, and the calculation of the variant long short-term memory network (LSTM) at each time step is derived. In this embodiment, the weight matrix of the forget gate is sequentially , the bias term of the forget gate , the weight matrix of the input gate , the bias term of the input gate , the weight matrix used to generate candidate memory units , bias term used to generate candidate memory units , the weight matrix of the output gate , the bias term of the output gate Perform gradient calculation.

[0080] Weight matrix for forget gate Perform gradient calculation: First calculate , according to the chain rule (because ,right Find the partial derivative); Then calculate ,because , according to the sigmoid function derivative formula: , and finally get: ,in, is the total amount of time; similarly, calculate the bias term of the forget gate , the weight matrix of the input gate , the bias term of the input gate , the weight matrix used to generate candidate memory units , bias term used to generate candidate memory units , the weight matrix of the output gate , the bias term of the output gate gradient.

[0081] Step 5: Parameter update: Based on the calculated gradient, use the stochastic gradient descent optimization algorithm to update the weights and biases of the model. The update formula is expressed as:

[0082]

[0083]

[0084]

[0085]

[0086]

[0087]

[0088]

[0089] Where, Represents the learning rate, which is used to control the step size of parameter updates.

[0090] Step 6: Iterative training: Repeat steps 2 to 5 until the loss function converges to a stable value or reaches the preset number of training rounds to obtain a trained cable fault detection model.

[0091] The present invention also provides a bypass flexible cable detection system, comprising: a data acquisition module, a data transmission module, a cloud service module, a deep learning module, and a mobile terminal; the data acquisition module comprises: a current transformer and a data acquisition unit; the current transformer is installed on the bypass flexible cable to collect current data of the bypass flexible cable in real time; the data acquisition unit is connected to the current transformer and is used to obtain the current data collected by the current transformer; the data transmission module is used to transmit the collected current data to a cloud server; the cloud server is used to receive, store, and process the current data; the deep learning module is used to store a trained cable fault detection model for fault detection of the bypass flexible cable. The mobile terminal supports a mobile terminal application for remotely viewing the operating status of the bypass flexible cable and receiving early warning information.

[0092] The bypass flexible cable detection system collects current data in real time through the data acquisition module. In conjunction with the data transmission module, cloud service module, deep learning module and mobile terminal, it can promptly transmit the data to the deep learning module for analysis. Once an abnormality is detected, the mobile terminal can immediately receive the fault information, allowing operation and maintenance personnel to know and handle the fault in the first time, reducing the impact of the fault on the power system and ensuring the stability of the power supply.

[0093] Exemplarily, the current transformer is one of the following: an open-type current transformer and a through-type current transformer. Both open-type and through-type current transformers are available for different scenarios and cable specifications. The open-type current transformer is opened and secured around the cable during installation, which can be completed in less than 5 minutes, significantly reducing installation time compared to traditional current transformers. The through-type current transformer adapts to complex factory environments and fluctuating cable positions, is simple to install, and can be adjusted at any time to ensure monitoring accuracy and stability.

[0094] In the embodiments of the bypass flexible cable detection system provided in the above embodiments of the present disclosure, the bypass flexible cable detection system and the bypass flexible cable detection method of the above embodiments belong to the same inventive concept. For details not fully described in the embodiments of the bypass flexible cable detection system, reference can be made to the embodiments of the above-mentioned bypass flexible cable detection method.

[0095] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the composition and steps of each example according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0096] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the devices, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. Exemplarily, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0097] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, or can be electrical, mechanical or other forms of connection.

[0098] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present invention. However, it will be appreciated by those skilled in the art that the technical solutions of the present invention can be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring various aspects of the present invention.

[0099] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

[0100] For those skilled in the art, designing different forms of control circuits according to the teachings of the present invention does not require creative work. These changes, modifications, substitutions and variations of the embodiments without departing from the principles and spirit of the present invention still fall within the scope of protection of the present invention.

Claims

1. A bypass flexible cable detection method, characterized in that: The method comprises: Step S1: collecting current data of the bypass flexible cable through a data acquisition module; Step S2: transmitting the collected current data to the cloud server through the data transmission module; Step S3: The cloud server receives, stores, and processes the current data, and sends the current data to the deep learning module; Step S4: the deep learning module stores a pre-trained cable fault detection model, and performs fault detection on the bypass flexible cable based on the pre-trained cable fault detection model; Step S5: remotely viewing the fault detection result of the bypass flexible cable through the mobile terminal application.

2. The bypass flexible cable detection method according to claim 1, wherein: The construction and training process of the cable fault detection model is as follows: Step S41: Input current data Perform normalization processing, and the expression of normalization processing is: in, Represents the normalized current data, whose value is [0,1]. represents the minimum value of the current data in the training data set, Represents the maximum value of the current data in the training data set; Step S42: Selecting a cable fault detection model to be trained and constructing a cable fault detection model; Step S43: Use historical current data with labels indicating whether the cable is overloaded to train the model and define a loss function; Step S44: adjusting the weights and biases of the model through a back-propagation algorithm to minimize the loss function, thereby training a cable fault detection model that can accurately determine whether the cable is overloaded; Step S45: Parameter update: Based on the calculated gradient, the stochastic gradient descent optimization algorithm is used to update the weights and biases of the model; Step S46: Iterative training: Repeat steps S42 to S45 until the loss function converges to a stable value or reaches a preset number of training rounds, thereby obtaining a trained cable fault detection model.

3. The bypass flexible cable detection method according to claim 2, wherein: In step S42, the specific process of model construction is as follows: Step S421: Calculate the forget gate, which is expressed as: Where, represents the Sigmoid activation function, represents the hidden layer The weight matrix of the output of each neuron to the forget gate of the output layer, Indicates that the hidden state of the previous moment and the current input To splice, Represents the bias term of the forget gate; Step S422: Calculate the input gate, which is expressed as: Where, is the hidden layer The weight matrix of the output of each neuron to the input gate of the output layer; Step S423: Calculate candidate memory cells, the expression is: Where, is the hidden layer The output of each neuron is used to generate a weight matrix for candidate memory units in the output layer; Represents candidate memory units; is the bias term of the input gate; is the bias term used to generate candidate memory units; represents the input gate; is the hyperbolic tangent activation function; Step S424: Update the memory unit, the expression is: Where, Represents element-wise multiplication; Indicates the current time The state of the memory unit; Indicates the last moment The state of the memory unit; Step S425: Calculate the output gate, which is expressed as: Where, represents the output gate; represents the hidden layer The weight matrix of the output of each neuron to the output gate of the output layer, Represents the bias term of the output gate; Step S426: Calculate the hidden state, which is expressed as: Indicates the current time The hidden state of .

4. The bypass flexible cable detection method according to claim 3, wherein: The expression of the pre-trained cable fault detection model is: Where, represents the predicted value output by the cable fault detection model; represents the Sigmoid activation function; represents the number of neurons in the hidden layer; represents the hidden layer The output of each neuron has a weight matrix for the output layer; represents the output gate; Indicates the current time The state of the memory unit; represents the bias term.

5. The bypass flexible cable detection method according to claim 4, characterized in that: The collected current data is transmitted to the cloud server through the data transmission module, including the following specific steps: The data transmission module uses modulation technology to modulate the current data; The data transmission module transmits the modulated current data to the cloud server via a wireless network.

6. The bypass flexible cable detection method according to claim 5, wherein: The cloud server receives, stores, and processes the current data, and sends it to the deep learning module. This involves the following specific steps: After receiving the data, the cloud server demodulates it and restores the original current data; The cloud server inputs the recovered raw current data into the deep learning module; The demodulation process includes: Mixing processing and low-pass filtering processing.

7. The bypass flexible cable detection method according to claim 6, wherein: Remotely view bypass flexible cable fault detection results via mobile terminal applications, including: When an overload fault is detected in the output of the cable fault detection model, the fault detection result is sent via a mobile terminal application; The mobile terminal application displays cable operating status, current data and overload fault information in real time.

8. A bypass flexible cable detection system applied to the bypass flexible cable detection method according to any one of claims 1 to 7, characterized in that: The system includes: a data acquisition module, a data transmission module, a cloud service module, a deep learning module and a mobile terminal; The data acquisition module is used to collect current data of the bypass flexible cable; The data transmission module is used to transmit the collected current data to the cloud server; The cloud server is used to receive, store and process current data; The deep learning module is used to store the trained cable fault detection model for fault detection of bypass flexible cables; The mobile terminal supports a mobile terminal application for remotely viewing the fault detection results of the bypass flexible cable.

9. The bypass flexible cable detection system according to claim 8, characterized in that: The data acquisition module includes: a current transformer and a data acquisition unit; The current transformer is installed on the bypass flexible cable to collect the current data of the bypass flexible cable in real time; The data acquisition unit is connected to the current transformer and is used to acquire current data collected by the current transformer.

10. The bypass flexible cable detection system according to claim 9, wherein: The current transformer is one of the following current transformers: Open-type current transformer, through-type current transformer.

Images