CN115800274A - Automatic self-adaption method and device for feeder of 5G power distribution network and storage medium - Google Patents
Automatic self-adaption method and device for feeder of 5G power distribution network and storage medium Download PDFInfo
- Publication number
- CN115800274A CN115800274A CN202310077578.6A CN202310077578A CN115800274A CN 115800274 A CN115800274 A CN 115800274A CN 202310077578 A CN202310077578 A CN 202310077578A CN 115800274 A CN115800274 A CN 115800274A
- Authority
- CN
- China
- Prior art keywords
- time delay
- feeder automation
- time
- distribution network
- power distribution
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a feeder automation self-adaption method, a feeder automation self-adaption device and a storage medium for a 5G power distribution network, wherein the method comprises the following steps: acquiring real-time operation data and historical feeder automation data of a 5G power distribution network; the historical feeder automation data comprises historical data of fault processing in different feeder automation modes under different communication time delays; inputting the real-time operation data into a preset time delay dependency calculation model to obtain the current time delay dependency; calculating a time delay dependency interval of each feeder automation mode according to the maximum fault processing time delay and the average feeder automation time delay in different feeder automation modes; and comparing the current time delay dependency with the time delay dependency interval to adjust a feeder automation mode of the 5G power distribution network. The invention can switch the feeder automation mode of the fastest fault processing corresponding to different network time delays, and realizes the self-adaptive switching of the feeder automation caused by the accidental time delay of the 5G communication, thereby ensuring the operation safety of the power distribution network.
Description
Technical Field
The invention relates to the technical field of feeder automation, in particular to a feeder automation self-adaption method, a feeder automation self-adaption device and a storage medium for a 5G power distribution network.
Background
Feeder automation is an important power distribution network fault self-healing control technology, and power supply reliability is improved by detecting faults, isolating the faults and finally recovering power supply. The intelligent distributed feeder automation technology is used as a mainstream feeder automation mode at present, and measurement and control information is exchanged among power distribution terminals in a peer-to-peer mode in the same communication mode, so that control decisions are realized to complete rapid isolation, accurate positioning and self-healing recovery of power distribution network faults. At present, information interaction is realized by an intelligent distributed feeder automation technology mainly depending on an optical fiber communication mode, however, the construction cost of an optical fiber channel of a city power distribution network is high, and the implementation difficulty is high. With the rapid development and wide application of a novel communication technology represented by a 5G communication technology, the problem can be effectively solved, and the 5G communication technology has the characteristics of flexible networking mode, long transmission distance, high safety performance and the like, and is suitable for an intelligent distributed feeder automation technology.
However, the wireless communication technology has a problem of low reliability, and is easily affected by the external environment, so that the communication delay is increased, and it is difficult to meet the requirement of high real-time performance of communication interaction between power distribution terminals. Therefore, how to provide a feeder automation self-adaptive method for a 5G power distribution network aiming at uncertain fluctuation time delay of 5G communication becomes a problem which needs to be solved urgently at present.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a 5G power distribution network feeder automation self-adaption method, a device and a storage medium, based on real-time operation data of a 5G power distribution network and historical data of fault treatment in different feeder automation modes under different communication time delays, the feeder automation mode of fastest power distribution network fault response suitable for real-time network time delay is output through LSTM deep learning historical data, the feeder automation mode of fastest fault treatment can be switched corresponding to different network time delays, and the feeder automation self-adaption switching brought by 5G communication sporadic time delay is realized, so that the operation safety of the power distribution network is ensured.
In order to achieve the above object, an embodiment of the present invention provides a feeder automation adaptive method for a 5G power distribution network, including:
acquiring real-time operation data and historical feeder automation data of a 5G power distribution network; the historical feeder automation data comprises historical data of fault processing in different feeder automation modes under different communication time delays;
inputting the real-time operation data into a preset time delay dependency calculation model to obtain the current time delay dependency output by the time delay dependency calculation model; the time delay dependency calculation model is obtained by training according to the historical feeder automation data and comprises an improved long-time and short-time memory neural network model and a residual error module;
calculating a time delay dependency interval of each feeder automation mode according to the maximum fault processing time delay and the average feeder automation time delay in different feeder automation modes;
and comparing the current time delay dependency with the time delay dependency interval, and adjusting a feeder automation mode of the 5G power distribution network according to a comparison result.
As an improvement of the above scheme, the training method of the delay dependency calculation model specifically includes:
inputting the historical feeder automation data serving as input data into an improved long-time and short-time memory neural network model for training to obtain initial time delay dependency output by the improved long-time and short-time memory neural network model;
inputting the initial time delay dependency into an improved data pooling layer and a residual error module in sequence to obtain the time delay dependency output by the residual error module;
and optimizing the improved long-time and short-time memory neural network model and the residual error module according to a preset loss function to obtain a trained time delay dependency calculation model.
As an improvement of the scheme, the improved long-time and short-time memory neural network model comprises a plurality of layers of long-time and short-time memory neural networks, and each layer of the long-time and short-time memory neural network is provided withNnA training unit; wherein the content of the first and second substances,
，nis shown asnThe layer length is memorized in the neural network.
As an improvement of the above scheme, the residual error module comprises a hole convolution, an activation function and a batch normalization function; wherein, the calculation formula of the receptive field in the cavity convolution is as follows:
in the formula (I), the compound is shown in the specification,
the receptive field of the current layer is represented,
denotes the receptive field of the next layer, m denotes the size of the equivalent convolution kernel, L denotes the distance from the first layer to the second layer
Product of layer step size.
As an improvement of the above scheme, the optimizing the improved long-and-short-term memory neural network model and the residual error module according to a preset loss function to obtain a trained time delay dependency calculation model specifically includes:
back propagation is carried out on the improved long-time memory neural network model and the residual error module according to a preset loss function;
and iteratively updating parameters in the improved long-time memory neural network model and the residual error module until preset iteration conditions are met, so as to obtain a trained time delay dependency calculation model.
As an improvement of the above scheme, the comparing the current delay dependency with the delay dependency interval, and adjusting the feeder automation mode of the 5G power distribution network according to the comparison result specifically includes:
comparing the current time delay dependency with the time delay dependency interval to obtain a time delay dependency interval in which the current time delay dependency falls;
and adjusting the feeder automation mode corresponding to the time delay dependency interval in which the current time delay dependency falls into to the current feeder automation mode of the 5G power distribution network.
As an improvement of the above scheme, the calculation formula of the time delay dependency is:
in the formula (I), the compound is shown in the specification,yin order to be a delay-dependent degree,
in order to maximize the processing delay for the fault,
the average delay for feeder automation.
The embodiment of the invention also provides a feeder automation self-adapting device for the 5G power distribution network, which comprises the following components:
the acquisition module is used for acquiring real-time operation data and historical feeder automation data of the 5G power distribution network; the historical feeder automation data comprises historical data of fault processing in different feeder automation modes under different communication time delays;
the first calculation module is used for inputting the real-time operation data into a preset time delay dependency calculation model to obtain the current time delay dependency output by the time delay dependency calculation model; the time delay dependency calculation model is obtained by training according to the historical feeder automation data and comprises an improved long-time and short-time memory neural network model and a residual error module;
the second calculation module is used for calculating the time delay dependency interval of each feeder automation mode according to the maximum fault processing time delay and the average feeder automation time delay in different feeder automation modes;
and the adjusting module is used for comparing the current time delay dependency with the time delay dependency interval and adjusting a feeder automation mode of the 5G power distribution network according to a comparison result.
The embodiment of the invention also provides a feeder automation adaptive device for a 5G power distribution network, which comprises a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the feeder automation adaptive method for the 5G power distribution network according to any one of the above items when executing the computer program.
The embodiment of the invention also provides a computer-readable storage medium, which includes a stored computer program, where when the computer program runs, the apparatus where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned 5G power distribution network feeder automation self-adaptation methods.
Compared with the prior art, the feeder automation self-adaption method, the feeder automation self-adaption device and the storage medium for the 5G power distribution network have the advantages that: real-time operation data and historical feeder automation data of a 5G power distribution network are acquired; the historical feeder automation data comprises historical data of fault processing in different feeder automation modes under different communication time delays; inputting the real-time operation data into a preset time delay dependency calculation model to obtain the current time delay dependency output by the time delay dependency calculation model; the time delay dependency calculation model is obtained by training according to the historical feeder automation data and comprises an improved long-time and short-time memory neural network model and a residual error module; calculating a time delay dependency interval of each feeder automation mode according to the maximum fault processing time delay and the average feeder automation time delay in different feeder automation modes; and comparing the current time delay dependency with the time delay dependency interval, and adjusting a feeder automation mode of the 5G power distribution network according to a comparison result. The embodiment of the invention outputs the feeder automation mode of the fastest distribution network fault response suitable for real-time network delay through LSTM deep learning historical data based on the real-time operation data of the 5G distribution network and the historical data of fault processing in different feeder automation modes under different communication delays, so that the feeder automation mode of the fastest fault processing can be switched corresponding to different network delays, the self-adaptive switching of the feeder automation caused by 5G communication sporadic delay is realized, and the operation safety of the distribution network is ensured.
Drawings
Fig. 1 is a schematic flow chart of a preferred embodiment of a feeder automation adaptive method for a 5G power distribution network provided by the present invention;
fig. 2 is a schematic structural diagram of a preferred embodiment of a feeder automation adaptive device for a 5G power distribution network provided by the present invention;
fig. 3 is a schematic structural diagram of another preferred embodiment of a feeder automation adaptive device for a 5G power distribution network provided by the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, fig. 1 is a schematic flow chart of a feeder automation adaptive method for a 5G distribution network according to a preferred embodiment of the present invention. The feeder automation self-adaptive method for the 5G power distribution network comprises the following steps:
s1, acquiring real-time operation data and historical feeder automation data of a 5G power distribution network; the historical feeder automation data comprises historical data of fault processing in different feeder automation modes under different communication time delays;
s2, inputting the real-time operation data into a preset time delay dependency calculation model to obtain the current time delay dependency output by the time delay dependency calculation model; the time delay dependency calculation model is obtained by training according to the historical feeder automation data and comprises an improved long-time and short-time memory neural network model and a residual error module;
s3, calculating a time delay dependency interval of each feeder automation mode according to the maximum fault processing time delay and the average feeder automation time delay in different feeder automation modes;
and S4, comparing the current time delay dependency with the time delay dependency interval, and adjusting a feeder automation mode of the 5G power distribution network according to a comparison result.
Specifically, the embodiment of the invention obtains real-time operation data and historical feeder automation data of the 5G power distribution network. The historical feeder automation data comprises historical data of fault processing in different feeder automation modes under different communication time delays. And inputting the real-time operation data into a preset time delay dependency calculation model to obtain the current time delay dependency output by the time delay dependency calculation model. The delay dependency calculation model is obtained by training according to historical feeder automation data and comprises an improved long-time and short-time memory neural network model LSTM and a residual error module. And calculating the time delay dependency interval of each feeder automation mode according to the maximum fault processing time delay and the average feeder automation time delay in different feeder automation modes. And comparing the current time delay dependency with the time delay dependency interval, and adjusting a feeder automation mode of the 5G power distribution network according to a comparison result.
It should be noted that, in the embodiment of the present invention, in consideration of the weak adaptability of the single-layer LSTM network, which may cause over-fitting and gradient disappearance, the data features of different depths are extracted from the data by using the multi-layer stacked LSTM, the processed data is subjected to fusion analysis, and the layer-by-layer stacked LSTM performs repeated training on the sample features to weaken the weak point of over-fitting and gradient disappearance. Because the number of network layers is increased, the model can generate a serious overfitting problem during training, so that the classification effect is not added, a residual block is added to improve the neural network, an output result is used as input and is transmitted to an improved data pooling layer, the extraction of characteristic information is optimized by using the improved data pooling layer and a residual module, and the training performance is further improved.
The embodiment of the invention outputs the feeder automation mode of the fastest distribution network fault response suitable for real-time network delay through LSTM deep learning historical data based on the real-time operation data of the 5G distribution network and the historical data of fault processing in different feeder automation modes under different communication delays, so that the feeder automation mode of the fastest fault processing can be switched corresponding to different network delays, the self-adaptive switching of the feeder automation caused by 5G communication sporadic delay is realized, and the operation safety of the distribution network is ensured.
In another preferred embodiment, the training method of the delay dependency calculation model specifically includes:
inputting the historical feeder automation data serving as input data into an improved long-time and short-time memory neural network model for training to obtain initial time delay dependency output by the improved long-time and short-time memory neural network model;
inputting the initial time delay dependency into an improved data pooling layer and a residual error module in sequence to obtain the time delay dependency output by the residual error module;
and optimizing the improved long-time and short-time memory neural network model and the residual error module according to a preset loss function to obtain a trained time delay dependency calculation model.
Specifically, based on the situation that faults are processed in different feeder automation modes with different communication delays in the historical situation, in order to select the feeder mode which is the fastest to process the faults in real-time communication delays, historical data of the fault processing are used as training samples in the different feeder automation modes with different historical communication delays, and a long-time and short-time memory neural network model (LSTM) is constructed to train the samples. Considering that the adaptability of the single-layer LSTM network is weak, in order to prevent situations such as overfitting and gradient disappearance in the training process, the embodiment of the invention adopts the multi-layer stacked LSTM network to solve the problem of weak adaptability of the single-layer LSTM network, the LSTM stacked layer by layer trains sample characteristics repeatedly to weaken the weak points of overfitting and gradient disappearance, and processed data is subjected to fusion analysis. And inputting the historical feeder automation data serving as input data into the improved long-time and short-time memory neural network model for training to obtain the initial time delay dependency output by the improved long-time and short-time memory neural network model. And sequentially inputting the initial time delay dependence degree into the improved data pooling layer and the residual error module to obtain the time delay dependence degree output by the residual error module. And optimizing the improved long-time and short-time memory neural network model and the residual error module according to a preset loss function to obtain a trained time delay dependency calculation model.
The typical LSTM structure is divided into three layers, including an input layer, an output layer, and a hidden layer, where the input layer is used to input a feature sequence set to be trained, and there is no one-to-one correspondence between input and output.
An input node: this node is marked as
Which behaves as a normal neuron, receives as output the output of the hidden node at the previous time point and the current input.
Wherein the content of the first and second substances,
the tan h activation function is expressed as,
an input representing the time delay of the current distribution network,
the time delay dependence output of the power distribution network in the last stage is shown,
and
respectively represents the power distribution network time delay dependency weight of the input node introduced into the power distribution network historical time delay input and the output of the previous stage,
and the distribution network time delay input deviation item of the input node is represented.
An input gate: to be provided with
The input gate is named, the input multiplication gate is a remarkable feature of the LSTM model, and like the input nodes, the gate inputs are the output and the current input of the hidden node of the previous time node.
Wherein the content of the first and second substances,
representing the activation of the sigmoid function,
representing the time delay input of the current power distribution network,
representing the time delay dependence output of the power distribution network in the last stage,
and
respectively representing the historical time delay input of the power distribution network at the input gate and the time delay dependency weight of the power distribution network output at the last stage,
and the distribution network time delay input deviation item represents an input door.
Forgetting to remember the door: by this method, the network can learn to refresh the contents of the internal state.
Wherein the content of the first and second substances,
and
respectively represents the power distribution network time delay input forgotten to refresh and the power distribution network time delay dependency weight output in the last stage,
and representing a distribution network time delay refreshing deviation item.
An output gate: the input of the gate is the output of the hidden node at the last moment and the current input, and the input of the gate plays a role of controlling output information.
Wherein the content of the first and second substances,
and
respectively representing the power distribution network time delay input of the output door and the power distribution network time delay dependency weight of the output in the previous stage,
and outputting deviation terms representing the time delay dependence of the power distribution network.
Internal state: the core of each memory cell is a node with linear activation, the cell is indexed by the linear activation node c, thereby indexing the internal state of the cell and naming the internal state as
：
Wherein the content of the first and second substances,
representing the input at the input node at time t,
indicating that the gate input is entered at time t,
represents the internal state quantity at the time t-1,
indicating the internal state refresh content at time t.
Will represent
Multiplying the output gate by the internal state
To generate a power distribution network delay dependency output by the memory unit
And based on network delay dependence
And analyzing and selecting a feeder automatic switching mode with the fastest fault response.
Wherein, the first and the second end of the pipe are connected with each other,
representing a multiplication by one.
Due to the fact that the number of the network layers is increased, the model can generate a serious overfitting problem during training, the classification effect is not added, the residual block is added for connection, and the problem of feature information loss is solved by adding the pooling layer and the residual block. And adding a residual error block to improve a residual error neural network, transmitting an output result as input to an improved data pooling layer, and optimizing the extraction of the characteristic information by using the improved data pooling layer and the residual error block, thereby further improving the training performance and ensuring that the classification effect is better.
Preferably, the length of the improvementThe time memory neural network model comprises a plurality of layers of long and short time memory neural networks, and each layer of the long and short time memory neural networks is provided withNnA training unit; wherein the content of the first and second substances,
，ndenotes the firstnThe layer length is memorized in the neural network.
In the embodiment of the invention, the single-layer LSTM network is considered to be weak in adaptability, and in order to prevent situations such as overfitting and gradient disappearance in the training process, the multi-layer stacked LSTM network is adopted to solve the problem of weak adaptability of the single-layer LSTM network, the LSTM stacked layer by layer is used for repeatedly training sample characteristics to weaken the weak points of overfitting and gradient disappearance, and processed data is subjected to fusion analysis. Wherein, the stacking layer number of the LSTM stacked in multiple layers is suitable for 2-4 layers, and each layer is provided with a long-term memory neural networkNnA training unit; wherein, the first and the second end of the pipe are connected with each other,
，nis shown asnThe layer length memorizes the neural network.
In yet another preferred embodiment, the residual module includes a hole convolution, an activation function, and a bulk normalization function; wherein, the calculation formula of the receptive field in the cavity convolution is as follows:
in the formula (I), the compound is shown in the specification,
the receptive field of the current layer is represented,
denotes the receptive field of the next layer, m denotes the size of the equivalent convolution kernel, L denotes the distance from the first layer to the second layer
Product of layer step size.
Specifically, in the embodiment of the present invention, because the number of LSTM network layers may increase the number of models during training to cause a severe overfitting problem, and thus the classification effect is not added, a residual block is added to improve the neural network, an output result is used as an input to be transmitted to an improved data pooling layer, the improved data pooling layer and a residual module are used to optimize the extraction of feature information, and the training performance is further improved, so that the classification effect is better. The residual module comprises a hole convolution, an activation function and a batch normalization function. The hole convolution expands the convolution kernel receptive field by adding zero padding in the common convolution, so that the convolution receptive field is larger on the premise of not changing the characteristic resolution, thereby increasing the range of perception information. Wherein, the calculation formula of the receptive field in the cavity convolution is as follows:
in the formula (I), the compound is shown in the specification,
the receptive field of the current layer is represented,
denotes the receptive field of the next layer, m denotes the size of the equivalent convolution kernel, L denotes the distance from the first layer to the second layer
Product of layer step size.
In another preferred embodiment, the optimizing the improved long-and-short-term memory neural network model and the residual error module according to a preset loss function to obtain a trained time delay dependency calculation model specifically includes:
back propagation is carried out on the improved long-time memory neural network model and the residual error module according to a preset loss function;
and iteratively updating parameters in the improved long-time memory neural network model and the residual error module until preset iteration conditions are met, so as to obtain a trained time delay dependency calculation model.
In another preferred embodiment, the comparing the current delay dependency with the delay dependency interval, and adjusting a feeder automation mode of the 5G power distribution network according to a comparison result specifically includes:
comparing the current time delay dependency with the time delay dependency interval to obtain a time delay dependency interval in which the current time delay dependency falls;
and adjusting the feeder automation mode corresponding to the time delay dependency interval in which the current time delay dependency falls into to the current feeder automation mode of the 5G power distribution network.
Specifically, the embodiment of the present invention compares the current delay dependency with the delay dependency interval to obtain the delay dependency interval in which the current delay dependency falls. And adjusting the feeder automation mode corresponding to the time delay dependency interval in which the current time delay dependency falls into to the current feeder automation mode of the 5G power distribution network.
In a further preferred embodiment, the calculation formula of the time delay dependency is:
in the formula (I), the compound is shown in the specification,yin order to be a delay-dependent degree,
in order to maximize the processing delay for the fault,
the average delay for feeder automation.
Specifically, the feeder automation system based on current common use can be classified into three categories, namely an interactive feeder automation technology, an interrogating feeder automation technology and a voltage-current feeder automation technology, according to different degrees of dependence on communication. And dividing a time delay dependency interval according to the dependency degree of different feeder automation modes on communication time delay so as to facilitate the output classification of the LSTM model. The calculation formula of the time delay dependency is as follows:
in the formula (I), the compound is shown in the specification,yin order to be a delay-dependent degree,
in order to maximize the processing delay for the fault,
the average delay for feeder automation.
The delay dependency intervals of the three feeder automation modes can be divided as follows according to the delay dependency:
a time delay dependence interval [0 to 0.3] of a voltage-current type feeder automation mode;
a delay dependence interval [0.3 to 0.6] of an interrogation feeder automation mode;
the time delay dependence interval of the interactive feeder automation mode is [0.6 to 1].
It should be noted that, in the embodiment of the present invention, a layer of Softmax may be used for classification at the end of the delay dependency calculation model, a fastest fault response feeder automation mode for different network delays is obtained according to the classification result, and an appropriate feeder mode is switched according to the result to realize fast processing of the power distribution network fault for different real-time communication delays.
Calculation of Softmax function:
wherein K is the total number of outputs, the denominator is normalization, which is composed of the sum of indexes on all output nodes, ensuring that the sum of outputs is 1, and the Softmax function is also called normalization index function, which is used for showing the result of multi-classification in a probability form. Wherein the content of the first and second substances,
is the output value of the kth node, the K of the denominator summation is the number of the output nodes,
the summation is started from the first output node. The output symbols are for highlightingAnd the judgment is easy.
When in use
When the system is used, the system adjusts and selects a voltage current type feeder automation technology to switch the feeder;
when in use
When the system is used, the system is adjusted and selects an inquiry type feeder automation technology to switch the feeders;
when in use
And when the system is adjusted, the interactive feeder automation technology is selected for feeder switching.
According to the embodiment of the invention, data are learned and trained through a multi-layer stacked LSTM long-time memory neural network model, an improved data pool and a hollow convolution residual block are added to optimize the extraction of characteristic information, data normalization and classification processing are finally carried out through a Softmax function, real-time power distribution network communication delay acquisition input is carried out, a feeder automation mode for processing faults most quickly in the current network delay is obtained according to the training and learning results, and a corresponding feeder automation strategy is deployed. The feeder switching mode aiming at real-time communication delay can eliminate the influence of the uncertainty problem of the communication delay on the feeder automation technology.
Correspondingly, the invention also provides a 5G power distribution network feeder automation self-adaption device which can realize all the processes of the 5G power distribution network feeder automation self-adaption method in the embodiment.
Referring to fig. 2, fig. 2 is a schematic structural diagram of a feeder automation adaptive device for a 5G distribution network according to a preferred embodiment of the present invention. Automatic self-adaptation device of 5G distribution network feeder includes:
the acquisition module 201 is used for acquiring real-time operation data and historical feeder automation data of the 5G power distribution network; the historical feeder automation data comprises historical data of fault processing in different feeder automation modes under different communication time delays;
the first calculation module 202 is configured to input the real-time operation data into a preset delay dependency calculation model, so as to obtain a current delay dependency output by the delay dependency calculation model; the time delay dependency calculation model is obtained by training according to the historical feeder automation data and comprises an improved long-time and short-time memory neural network model and a residual error module;
the second calculating module 203 is configured to calculate a delay dependency interval of each feeder automation mode according to the maximum fault processing delay and the average feeder automation delay in different feeder automation modes;
and the adjusting module 204 is configured to compare the current delay dependency with the delay dependency interval, and adjust a feeder automation mode of the 5G power distribution network according to a comparison result.
Preferably, the training method of the delay dependency calculation model specifically includes:
inputting the historical feeder automation data serving as input data into an improved long-time and short-time memory neural network model for training to obtain initial time delay dependency output by the improved long-time and short-time memory neural network model;
inputting the initial time delay dependency into an improved data pooling layer and a residual error module in sequence to obtain the time delay dependency output by the residual error module;
and optimizing the improved long-time and short-time memory neural network model and the residual error module according to a preset loss function to obtain a trained time delay dependency calculation model.
Preferably, the improved long-term and short-term memory neural network model comprises a plurality of layers of long-term and short-term memory neural networks, and each layer of the long-term and short-term memory neural network is provided withNnA training unit; wherein the content of the first and second substances,
，nis shown asnThe layer length is memorized in the neural network.
Preferably, the residual module comprises a hole convolution, an activation function and a batch normalization function; wherein, the calculation formula of the receptive field in the cavity convolution is as follows:
in the formula (I), the compound is shown in the specification,
the receptive field of the current layer is represented,
denotes the receptive field of the next layer, m denotes the size of the equivalent convolution kernel, L denotes the distance from the first layer to the second layer
Product of layer step size.
Preferably, the optimizing the improved long-and-short-term memory neural network model and the residual error module according to a preset loss function to obtain a trained time delay dependency calculation model specifically includes:
back propagation is carried out on the improved long-time memory neural network model and the residual error module according to a preset loss function;
and iteratively updating parameters in the improved long-time memory neural network model and the residual error module until preset iteration conditions are met, so as to obtain a trained time delay dependency calculation model.
Preferably, the adjusting module 204 is specifically configured to:
comparing the current time delay dependency with the time delay dependency interval to obtain a time delay dependency interval in which the current time delay dependency falls;
and adjusting the feeder automation mode corresponding to the time delay dependency interval in which the current time delay dependency falls into to the current feeder automation mode of the 5G power distribution network.
Preferably, the calculation formula of the delay dependency is as follows:
in the formula (I), the compound is shown in the specification,yin order to be a delay-dependent degree,
in order to maximize the processing delay for the fault,
the average delay for feeder automation.
In a specific implementation, the working principle, the control flow and the technical effect of the 5G power distribution network feeder automation adaptive device provided in the embodiment of the present invention are the same as those of the 5G power distribution network feeder automation adaptive method in the above embodiment, and are not described herein again.
Referring to fig. 3, fig. 3 is a schematic structural diagram of another preferred embodiment of a feeder automation adaptive device for a 5G distribution network according to the present invention. The 5G power distribution network feeder automation adaptive device comprises a processor 301, a memory 302 and a computer program stored in the memory 302 and configured to be executed by the processor 301, wherein the processor 301 implements the 5G power distribution network feeder automation adaptive method according to any of the above embodiments when executing the computer program.
Preferably, the computer program may be divided into one or more modules/units (e.g., computer program 1, computer program 2, … …) that are stored in the memory 302 and executed by the processor 301 to implement the present invention. The one or more modules/units may be a series of instruction segments of a computer program capable of performing specific functions, and the instruction segments are used for describing the execution process of the computer program in the feeder automation adaptive device of the 5G distribution network.
The Processor 301 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, a discrete hardware component, etc., the general purpose Processor may be a microprocessor, or the Processor 301 may be any conventional Processor, the Processor 301 is a control center of the 5G distribution network feeder automation adaptation apparatus, and various interfaces and lines are used to connect various parts of the 5G distribution network feeder automation apparatus.
The memory 302 mainly includes a program storage area that may store an operating system, an application program required for at least one function, and the like, and a data storage area that may store related data and the like. In addition, the memory 302 may be a high speed random access memory, a non-volatile memory such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card (Flash Card), and the like, or the memory 302 may be other volatile solid state memory devices.
It should be noted that the above 5G distribution network feeder automation adaptive device may include, but is not limited to, a processor and a memory, and those skilled in the art will understand that the schematic structural diagram of fig. 3 is only an example of the above 5G distribution network feeder automation adaptive device, and does not constitute a limitation to the above 5G distribution network feeder automation adaptive device, and may include more or less components than those shown in the drawings, or combine some components, or different components.
The embodiment of the present invention further provides a computer-readable storage medium, where the computer-readable storage medium includes a stored computer program, and when the computer program runs, a device where the computer-readable storage medium is located is controlled to execute the 5G distribution network feeder automation self-adaptive method according to any of the above embodiments.
The embodiment of the invention provides a feeder automation self-adaption method, a feeder automation self-adaption device and a storage medium for a 5G power distribution network, wherein real-time operation data and historical feeder automation data of the 5G power distribution network are obtained; the historical feeder automation data comprises historical data of fault processing in different feeder automation modes under different communication time delays; inputting the real-time operation data into a preset time delay dependency calculation model to obtain the current time delay dependency output by the time delay dependency calculation model; the time delay dependency calculation model is obtained by training according to the historical feeder automation data and comprises an improved long-time and short-time memory neural network model and a residual error module; calculating a time delay dependency interval of each feeder automation mode according to the maximum fault processing time delay and the average feeder automation time delay in different feeder automation modes; and comparing the current time delay dependency with the time delay dependency interval, and adjusting a feeder automation mode of the 5G power distribution network according to a comparison result. The embodiment of the invention outputs the feeder automation mode of fastest distribution network fault response suitable for real-time network delay through LSTM deep learning historical data based on real-time operation data of a 5G distribution network and historical data of fault processing by different feeder automation modes under different communication delays, so that the feeder automation mode of fastest fault processing can be switched corresponding to different network delays, self-adaptive switching of feeder automation caused by 5G communication sporadic delay is realized, and the operation safety of the distribution network is ensured.
It should be noted that the above-described system embodiments are merely illustrative, where the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. In addition, in the drawings of the embodiment of the system provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, and may be specifically implemented as one or more communication buses or signal lines. One of ordinary skill in the art can understand and implement it without inventive effort.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention.
Claims (10)
1. A feeder automation self-adaption method for a 5G power distribution network is characterized by comprising the following steps:
acquiring real-time operation data and historical feeder automation data of a 5G power distribution network; the historical feeder automation data comprises historical data of fault processing in different feeder automation modes under different communication time delays;
inputting the real-time operation data into a preset time delay dependency calculation model to obtain the current time delay dependency output by the time delay dependency calculation model; the time delay dependency calculation model is obtained by training according to the historical feeder automation data and comprises an improved long-time and short-time memory neural network model and a residual error module;
calculating a time delay dependency interval of each feeder automation mode according to the maximum fault processing time delay and the average feeder automation time delay in different feeder automation modes;
and comparing the current time delay dependency with the time delay dependency interval, and adjusting a feeder automation mode of the 5G power distribution network according to a comparison result.
2. The automatic adaptive method for feeder lines of a 5G power distribution network according to claim 1, wherein the training method of the delay dependency calculation model specifically comprises:
inputting the historical feeder automation data serving as input data into an improved long-time and short-time memory neural network model for training to obtain initial time delay dependency output by the improved long-time and short-time memory neural network model;
inputting the initial time delay dependency into an improved data pooling layer and a residual error module in sequence to obtain the time delay dependency output by the residual error module;
and optimizing the improved long-time and short-time memory neural network model and the residual error module according to a preset loss function to obtain a trained time delay dependency calculation model.
3. The automatic feeder adaptation method for the 5G power distribution network according to claim 2, wherein the improved long-term and short-term memory neural network model comprises a plurality of layers of long-term and short-term memory neural networks, and each layer of the long-term and short-term memory neural networks is provided withNnA training unit; wherein the content of the first and second substances,
，nis shown asnThe layer length is memorized in the neural network.
4. The 5G power distribution network feeder automation adaptation method of claim 3, wherein the residual module comprises a hole convolution, an activation function, and a bulk normalization function; wherein, the calculation formula of the receptive field in the cavity convolution is as follows:
in the formula (I), the compound is shown in the specification,
the receptive field of the current layer is represented,
denotes the receptive field of the next layer, m denotes the size of the equivalent convolution kernel, L denotes the distance from the first layer to the second layer
Product of layer step size.
5. The automatic self-adapting method for the feeder of the 5G power distribution network according to claim 4, wherein the optimization of the improved long-time memory neural network model and the residual error module according to a preset loss function to obtain a trained time delay dependency calculation model specifically comprises:
back propagation is carried out on the improved long-time memory neural network model and the residual error module according to a preset loss function;
and iteratively updating parameters in the improved long-time memory neural network model and the residual error module until preset iteration conditions are met, so as to obtain a trained time delay dependency calculation model.
6. The automatic feeder adaptation method for the 5G power distribution network according to claim 5, wherein the comparing the current delay dependency with the interval of the delay dependency and adjusting the automatic feeder mode of the 5G power distribution network according to the comparison result specifically comprises:
comparing the current time delay dependency with the time delay dependency interval to obtain a time delay dependency interval in which the current time delay dependency falls;
and adjusting the feeder automation mode corresponding to the time delay dependency interval in which the current time delay dependency falls into to the current feeder automation mode of the 5G power distribution network.
7. The 5G power distribution network feeder automation self-adapting method according to claim 6, wherein the time delay dependency is calculated by the formula:
in the formula (I), the compound is shown in the specification,yin order to be a delay-dependent degree,
in order to maximize the processing delay for the fault,
the average delay for feeder automation.
8. The utility model provides a 5G distribution network feeder automation self-adaptation device which characterized in that includes:
the acquisition module is used for acquiring real-time operation data and historical feeder automation data of the 5G power distribution network; the historical feeder automation data comprises historical data of fault processing in different feeder automation modes under different communication time delays;
the first calculation module is used for inputting the real-time operation data into a preset time delay dependency calculation model to obtain the current time delay dependency output by the time delay dependency calculation model; the time delay dependency calculation model is obtained by training according to the historical feeder automation data and comprises an improved long-time and short-time memory neural network model and a residual error module;
the second calculation module is used for calculating the time delay dependency interval of each feeder automation mode according to the maximum fault processing time delay and the average feeder automation time delay in different feeder automation modes;
and the adjusting module is used for comparing the current time delay dependency with the time delay dependency interval and adjusting a feeder automation mode of the 5G power distribution network according to a comparison result.
9. A feeder automation adaptation device for a 5G power distribution network, which is characterized by comprising a processor and a memory, wherein the memory stores a computer program, and the computer program is configured to be executed by the processor, and when the processor executes the computer program, the feeder automation adaptation device for the 5G power distribution network realizes the feeder automation adaptation method for the 5G power distribution network according to any one of claims 1 to 7.
10. A computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and wherein when the computer program is executed by a device, the computer-readable storage medium implements the method for feeder automation adaptation for a 5G power distribution network according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310077578.6A CN115800274B (en) | 2023-02-08 | 2023-02-08 | 5G distribution network feeder automation self-adaptation method, device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310077578.6A CN115800274B (en) | 2023-02-08 | 2023-02-08 | 5G distribution network feeder automation self-adaptation method, device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115800274A true CN115800274A (en) | 2023-03-14 |
| CN115800274B CN115800274B (en) | 2023-06-30 |
Family
ID=85430368
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310077578.6A Active CN115800274B (en) | 2023-02-08 | 2023-02-08 | 5G distribution network feeder automation self-adaptation method, device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115800274B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117791597A (en) * | 2024-02-23 | 2024-03-29 | 广东电网有限责任公司广州供电局 | Power distribution network fault self-healing method and system based on machine learning |
| CN117791597B (en) * | 2024-02-23 | 2024-05-28 | 广东电网有限责任公司广州供电局 | Power distribution network fault self-healing method and system based on machine learning |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113036731A (en) * | 2021-04-25 | 2021-06-25 | 武汉大学 | Point-to-point communication power distribution network feeder fault isolation method and self-healing system |
| CN114566964A (en) * | 2022-04-29 | 2022-05-31 | 国网天津市电力公司电力科学研究院 | Power distribution network feeder automation control method, device, equipment and storage medium |
| CN114726105A (en) * | 2022-06-02 | 2022-07-08 | 国网天津市电力公司电力科学研究院 | Power distribution network feeder automatic switching method, device and medium based on communication time delay |
-
2023
- 2023-02-08 CN CN202310077578.6A patent/CN115800274B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113036731A (en) * | 2021-04-25 | 2021-06-25 | 武汉大学 | Point-to-point communication power distribution network feeder fault isolation method and self-healing system |
| CN114566964A (en) * | 2022-04-29 | 2022-05-31 | 国网天津市电力公司电力科学研究院 | Power distribution network feeder automation control method, device, equipment and storage medium |
| CN114726105A (en) * | 2022-06-02 | 2022-07-08 | 国网天津市电力公司电力科学研究院 | Power distribution network feeder automatic switching method, device and medium based on communication time delay |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117791597A (en) * | 2024-02-23 | 2024-03-29 | 广东电网有限责任公司广州供电局 | Power distribution network fault self-healing method and system based on machine learning |
| CN117791597B (en) * | 2024-02-23 | 2024-05-28 | 广东电网有限责任公司广州供电局 | Power distribution network fault self-healing method and system based on machine learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115800274B (en) | 2023-06-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101414366B (en) | Method for forecasting electric power system short-term load based on method for improving uttermost learning machine | |
| CN102622418B (en) | Prediction device and equipment based on BP (Back Propagation) nerve network | |
| CN112116144B (en) | Regional power distribution network short-term load prediction method | |
| CN111144552B (en) | Multi-index grain quality prediction method and device | |
| CN108879732B (en) | Transient stability evaluation method and device for power system | |
| CN111027772A (en) | Multi-factor short-term load prediction method based on PCA-DBILSTM | |
| CN112990444B (en) | Hybrid neural network training method, system, equipment and storage medium | |
| CN114791571B (en) | Lithium ion battery life prediction method and device based on improved CSO-LSTM network | |
| CN112651519A (en) | Secondary equipment fault positioning method and system based on deep learning theory | |
| Bi et al. | A novel ANN fault diagnosis system for power systems using dual GA loops in ANN training | |
| CN113722980A (en) | Ocean wave height prediction method, system, computer equipment, storage medium and terminal | |
| CN112272074A (en) | Information transmission rate control method and system based on neural network | |
| CN115824261A (en) | Control moment gyroscope fault detection method and device and related storage medium | |
| CN108665001B (en) | Cross-tested idle state detection method based on deep belief network | |
| CN114048811A (en) | Wireless sensor node fault diagnosis method and device based on deep learning | |
| Skorpil et al. | Back-propagation and k-means algorithms comparison | |
| CN115800274A (en) | Automatic self-adaption method and device for feeder of 5G power distribution network and storage medium | |
| Alam | Recurrent neural networks in electricity load forecasting | |
| CN116542701A (en) | Carbon price prediction method and system based on CNN-LSTM combination model | |
| CN113239547B (en) | Core backbone network frame construction method based on TLBO algorithm | |
| CN114330688A (en) | Model online migration training method, device and chip based on resistive random access memory | |
| CN114662658A (en) | On-chip optical network hot spot prediction method based on LSTM neural network | |
| CN114169416A (en) | Short-term load prediction method under small sample set based on transfer learning | |
| CN113033695A (en) | Method for predicting faults of electronic device | |
| CN113821974A (en) | Engine residual life prediction method based on multiple failure modes |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |