CN115879012A - Power line noise analysis method, device, equipment and storage medium - Google Patents
Power line noise analysis method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN115879012A CN115879012A CN202211742169.5A CN202211742169A CN115879012A CN 115879012 A CN115879012 A CN 115879012A CN 202211742169 A CN202211742169 A CN 202211742169A CN 115879012 A CN115879012 A CN 115879012A
- Authority
- CN
- China
- Prior art keywords
- noise
- data
- matching degree
- neural network
- frequency domain
- 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.)
- Pending
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 37
- 238000013528 artificial neural network Methods 0.000 claims abstract description 78
- 230000002776 aggregation Effects 0.000 claims abstract description 54
- 238000004220 aggregation Methods 0.000 claims abstract description 54
- 238000012549 training Methods 0.000 claims abstract description 28
- 238000000034 method Methods 0.000 claims description 22
- 238000004590 computer program Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000000379 polymerizing effect Effects 0.000 claims description 4
- 230000004931 aggregating effect Effects 0.000 claims description 3
- 238000004891 communication Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000012360 testing method Methods 0.000 description 3
- 210000002569 neuron Anatomy 0.000 description 2
- 238000006116 polymerization reaction Methods 0.000 description 2
- 230000008054 signal transmission Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Images
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The application discloses a power line noise analysis method, a device, equipment and a storage medium, wherein noise data generated by various power electronic devices and a noise characteristic data set corresponding to the power electronic devices in a power line noise library are acquired; calculating an aggregate device matching degree between the noise data and the power electronic device based on the noise data and the noise feature data set; training a preset neural network based on the matching degree of the aggregation device, the noise data and the noise characteristic data until the preset neural network reaches a preset convergence condition to obtain an aggregation frequency domain neural network, so that the aggregation frequency domain neural network learns the influence of the matching degree on the noise characteristic analysis, and the prediction precision is improved; and analyzing the target noise data generated by the target power electronic device by utilizing the aggregation frequency domain neural network to obtain the target noise characteristic data of the target power electronic device. The influence of the device matching degree on noise analysis is considered, and high-precision prediction is achieved.
Description
Technical Field
The present disclosure relates to the field of power line communication technologies, and in particular, to a method, an apparatus, a device, and a storage medium for analyzing noise of a power line.
Background
With the construction of new power systems, the scale of power grids is continuously enlarged, and the demand of power services for communication is also continuously increased. Compared with other communication technologies, the power line carrier communication takes a power line as a transmission medium, does not need to erect a line, and has the advantages of wide coverage range, long online time and the like. However, since a large number of power electronic devices are deployed in a low-voltage distribution area, noise interference is particularly complicated, and the error rate of signal transmission is increased due to the characteristics of the power electronic devices, so that the communication quality is reduced.
At present, the noise characteristics can be effectively identified by using a power line noise analysis technology, so that the error rate of signal transmission is reduced, and the communication quality is improved. However, the conventional power line noise analysis technique analyzes the noise characteristics based only on the prior information of the matching device, but ignores the influence of the degree of matching between the noise data and the power electronic device on the noise analysis, resulting in poor prediction accuracy of the noise characteristics.
Disclosure of Invention
The application provides a power line noise analysis method, a device, equipment and a storage medium, which aim to solve the technical problem that the noise prediction precision is poor in the noise interference of power electronic devices in a low-voltage distribution room in the current noise analysis technology.
In order to solve the above technical problem, in a first aspect, the present application provides a power line noise analysis method, including:
acquiring noise data generated by various power electronic devices and a noise characteristic data set corresponding to the power electronic devices in a power line noise library, wherein the noise characteristic data set comprises the noise characteristic data of the various power electronic devices;
calculating an aggregate device match metric between the noise data and the power electronics device based on the noise data and the noise signature dataset;
training a preset neural network based on the aggregation device matching degree, the noise data and the noise characteristic data until the preset neural network reaches a preset convergence condition to obtain an aggregation frequency domain neural network;
and analyzing the target noise data generated by the target power electronic device by using the aggregation frequency domain neural network to obtain the target noise characteristic data of the target power electronic device.
In some implementations, the calculating an aggregate device match between the noise data and the power electronic device based on the noise data and the noise signature dataset includes:
calculating a device matching degree between the noise data and a plurality of types of the power electronic devices based on the noise data and the noise feature data set;
and polymerizing the matching degrees of the devices to obtain the matching degree of the polymerized devices.
In some implementations, the calculating a device match score between the noise data and the plurality of power electronic devices based on the noise data and the noise signature dataset includes:
calculating a distance value between the noise data and the noise characteristic data of each power electronic device based on a preset device matching degree formula, wherein the distance value is the device matching degree, and the preset device matching degree formula is as follows:
wherein, theta n (i) The device matching degree between the ith noise data and the nth power electronic device is shown, x (i) is the ith noise data, u n The noise characteristic data of the nth power electronic device.
In some implementations, the aggregating the plurality of device matching degrees to obtain the aggregated device matching degree includes:
calculating the matching degree of the aggregation device according to the matching degrees of the devices by using a preset aggregation matching degree formula, wherein the preset aggregation matching degree formula is as follows:
wherein,
aggregate device match, w, for ith noisy data n (i) Corresponding to the device matching degree weight of the nth power electronic device for the ith noise data n (i) The device matching degree between the ith noise data and the nth power electronic device is shown.
In some implementations, the training a preset neural network based on the aggregation device matching degree, the noise data, and the noise feature data until the preset neural network reaches a preset convergence condition to obtain an aggregated frequency domain neural network includes:
performing Fourier transform on the aggregation device matching degree, the noise data and the noise characteristic data to obtain frequency domain matching degree, noise frequency domain data and noise frequency domain characteristic data;
taking the frequency domain matching degree and the noise frequency domain data as the input of the preset neural network, and performing iterative training on the preset neural network;
for each iteration training, comparing a network prediction result output by each iteration of the preset neural network with the noise frequency domain characteristic data, and calculating a frequency domain prediction error of each iteration;
updating the device matching degree weight and updating the network parameters of the preset neural network based on the frequency domain prediction error obtained by each iteration until the frequency domain prediction error is smaller than a preset threshold value, and judging that the preset neural network reaches a preset convergence condition to obtain the aggregation frequency domain neural network.
In some implementations, the updating the device match-degree weight includes:
performing Fourier transform on the frequency domain prediction error to obtain a time domain prediction error;
updating the device matching degree weight according to the time domain prediction error by using a preset weight updating formula, wherein the preset weight updating formula is as follows:
wherein w n (i) The ith noise data corresponds to the device matching degree weight of the nth power electronic device, alpha (i) is the update rate of the device matching degree weight, e (i) is the time domain prediction error, and theta (theta) is n (i) The device match between the ith noise data and the nth power electronics device.
In some implementations, after analyzing, by using the aggregated frequency domain neural network, target noise data generated by a target power electronic device to obtain target noise characteristic data of the target power electronic device, the method further includes:
and updating the power line noise library based on the target noise characteristic data.
In a second aspect, the present application further provides a power line noise analysis method, including:
the device comprises an acquisition module, a comparison module and a comparison module, wherein the acquisition module is used for acquiring noise data generated by various power electronic devices and noise characteristic data sets corresponding to the power electronic devices in a power line noise library, and the noise characteristic data sets comprise noise characteristic data of various power electronic devices;
a calculation module to calculate an aggregate device match metric between the noise data and the power electronics device based on the noise data and the noise signature dataset;
the training module is used for training a preset neural network based on the aggregation device matching degree, the noise data and the noise characteristic data until the preset neural network reaches a preset convergence condition, so as to obtain an aggregation frequency domain neural network;
and the analysis module is used for analyzing the target noise data generated by the target power electronic device by utilizing the aggregation frequency domain neural network to obtain the target noise characteristic data of the target power electronic device.
In a third aspect, the present application further provides a computer device comprising a processor and a memory for storing a computer program, which when executed by the processor implements the power line noise analysis method according to the first aspect.
In a fourth aspect, the present application further provides a computer-readable storage medium storing a computer program, which when executed by a processor implements the power line noise analysis method according to the first aspect.
Compared with the prior art, the application has the following beneficial effects at least:
calculating an aggregate device matching degree between noise data and a power electronic device based on the noise data and a noise characteristic data set corresponding to the power electronic device in a power line noise library by acquiring the noise data generated by various power electronic devices and the noise characteristic data set corresponding to the power electronic device in the power line noise library so as to consider the influence of the prior information matching degree on noise analysis; training a preset neural network based on the aggregation device matching degree, the noise data and the noise characteristic data until the preset neural network reaches a preset convergence condition to obtain an aggregation frequency domain neural network, so that the aggregation frequency domain neural network learns the influence of the matching degree on noise characteristic analysis, and the prediction precision is improved; and analyzing the target noise data generated by the target power electronic device by using the aggregation frequency domain neural network to obtain the target noise characteristic data of the target power electronic device, thereby realizing high-precision prediction of the noise characteristic of the power line.
Drawings
Fig. 1 is a schematic flowchart of a power line noise analysis method according to an embodiment of the present disclosure;
fig. 2 is a schematic structural diagram of a power line noise analysis apparatus according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of a computer device according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, 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 application.
Referring to fig. 1, fig. 1 is a schematic flow chart illustrating a method for analyzing power line noise according to an embodiment of the present disclosure. The power line noise analysis method can be applied to computer equipment including, but not limited to, smart phones, laptops, tablet computers, desktop computers, physical servers, cloud servers and other equipment. As shown in fig. 1, the power line noise analysis method of the present embodiment includes steps S101 to S104, which are detailed as follows:
step S101, acquiring noise data generated by various power electronic devices and a noise characteristic data set corresponding to the power electronic devices in a power line noise library, wherein the noise characteristic data set comprises the noise characteristic data of the various power electronic devices.
In this step, noise data generated when the power electronic devices facing the low-voltage platform area operate is collected, and the power electronic devices generating the noise data are classified into N types according to the difference of the noise type of the noise data. Obtaining noise characteristic data of each type (each type) of power electronic devices by searching prior information of a power line noise library, taking the noise characteristic data as initial central data, and defining a central data set (set noise characteristic data set) omega = { u = (a noise characteristic data set of a power line noise) 1 ,u 2 ,...,u n ,...,u N H, where u n And represents noise characteristic data of the nth power electronic device.
Step S102, calculating an aggregate device matching degree between the noise data and the power electronic device based on the noise data and the noise characteristic data set.
In this step, for each kind of noise data, the device matching degree between the noise data and the noise characteristic data of each kind of power electronic device is calculated, and then all the device matching degrees of the noise data are aggregated to obtain an aggregated device matching degree, so that the influence of the device matching degree between the noise data and the power electronic device on noise analysis is considered.
In some embodiments, the step S101 includes:
calculating a device matching degree between the noise data and a plurality of types of the power electronic devices based on the noise data and the noise feature data set;
and polymerizing the matching degrees of the devices to obtain the matching degree of the polymerized devices.
In this embodiment, the noise data is traversed through all the center data in step S101, the distances between the noise data and the center data are compared, so as to obtain the device matching degrees between the noise data and the power electronic devices, and then the device matching degrees are aggregated, so as to obtain the aggregated device matching degree.
Optionally, a distance value between the noise data and the noise characteristic data of each of the power electronic devices is calculated based on a preset device matching degree formula, where the distance value is the device matching degree, and the preset device matching degree formula is:
wherein, theta n (i) The device matching degree between the ith noise data and the nth power electronic device is shown, x (i) is the ith noise data, u n The noise characteristic data of the nth power electronic device.
Optionally, a preset aggregate matching degree formula is used to calculate the aggregate device matching degree according to the device matching degrees, where the preset aggregate matching degree formula is:
wherein,
aggregate device match, w, for ith noisy data n (i) Corresponding to the device matching degree weight of the nth power electronic device for the ith noise data n (i) The device matching degree between the ith noise data and the nth power electronic device is shown.
Step S103, training a preset neural network based on the aggregation device matching degree, the noise data and the noise characteristic data until the preset neural network reaches a preset convergence condition, and obtaining an aggregation frequency domain neural network.
In this step, the aggregated frequency domain neural network comprises an input layer, a hidden layer and an output layer, the number of neurons in the input layer is P, and the number of neurons in each layer of the hidden layer is PDefining the input of the aggregation frequency domain neural network suitable for power line noise analysis as noise frequency domain data x f (i) And frequency domain sum matching degree
And outputting the data as actual noise frequency domain characteristic data.
In some embodiments, the step S103 includes:
performing Fourier transform on the aggregation device matching degree, the noise data and the noise characteristic data to obtain frequency domain matching degree, noise frequency domain data and noise frequency domain characteristic data;
taking the frequency domain matching degree and the noise frequency domain data as the input of the preset neural network, and performing iterative training on the preset neural network;
for each iteration training, comparing a network prediction result output by each iteration of the preset neural network with the noise frequency domain characteristic data, and calculating a frequency domain prediction error of each iteration;
updating the device matching degree weight and updating the network parameters of the preset neural network based on the frequency domain prediction error obtained by each iteration until the frequency domain prediction error is smaller than a preset threshold value, and judging that the preset neural network reaches a preset convergence condition to obtain the aggregation frequency domain neural network.
In the present embodiment, the noise data x (i), the aggregate device matching degree
And carrying out Fourier transform on the noise characteristic data to obtain noise frequency domain data x f (i) The frequency domain matching degree->
And noise frequency domain feature data, and dividing the noise frequency domain feature data into a training set and a test set. The training set is used for training the network, and the testing set is used for calculating posterior information of the noise frequency domain characteristics.
Inputting training sets into aggregationsThe frequency domain neural network carries out iterative training and defines the ith noise data to be output at the r th output of the aggregation frequency domain neural network
Expressed as:
wherein,
for aggregating the p-th input of the input layer of the frequency-domain neural network>
For weights between an input layer and a hidden layer>
For hiding a layer threshold value>
For weights between hidden layer and output layer>
Is the output layer threshold.
The network actually outputs (namely actual noise frequency domain characteristic data)
Based on the desired output (i.e., the noise frequency domain feature data)>
Calculating the frequency domain prediction error e f (i) It is represented as:
according toFrequency domain prediction error e f (i) The aggregated frequency domain network parameters (weights and thresholds) are updated in the negative gradient direction of (1).
Optionally, updating the device matching degree weight includes: performing Fourier transform on the frequency domain prediction error to obtain a time domain prediction error; updating the device matching degree weight according to the time domain prediction error by using a preset weight updating formula, wherein the preset weight updating formula is as follows:
wherein, w n (i) The ith noise data corresponds to the device matching degree weight of the nth power electronic device, alpha (i) is the update rate of the device matching degree weight, e (i) is the time domain prediction error, and theta (theta) is n (i) The device matching degree between the ith noise data and the nth power electronic device is shown.
In the embodiment, the device matching degree weight is updated in a self-adaptive manner by utilizing the time domain prediction error, when the error does not meet the precision, the matching degree weight of the power electronic device type with the maximum matching degree is increased, and the matching degree weights of other types of power electronic devices are reduced, so that the prediction precision of the neural network is improved.
When the time domain prediction error is smaller than the preset time domain prediction error
And judging that the network converges, namely finishing the training.
In some embodiments, after step S103, the method further includes:
and updating the power line noise library based on the target noise characteristic data.
In this embodiment, the test set is input to the aggregation frequency domain neural network after training to obtain power line noise frequency domain feature data, and the power line noise frequency domain feature data is subjected to inverse laplace transform to obtain power line noise time domain feature data, which is stored in the noise library to update the noise feature data of the corresponding power electronic device in the noise library.
And step S104, analyzing the target noise data generated by the target power electronic device by using the aggregation frequency domain neural network to obtain the target noise characteristic data of the target power electronic device.
In the step, the target noise data is input into the aggregation frequency domain neural network, and target noise frequency domain characteristic data is output; and performing inverse Laplace transform on the target noise frequency domain characteristic data to obtain the target noise characteristic data.
It should be noted that, compared with the related art, the method and the device for predicting the noise characteristics of the power line noise device realize the matching of the noise data and the power electronic device based on the prior information in the noise library, update the prior information in the noise library in real time by utilizing the output of the aggregation frequency domain neural network, improve the information reliability of the noise library, and realize the high-precision matching of the power line noise device and the high-precision prediction analysis of the noise characteristics.
The method and the device for predicting the power line noise characteristics of the power line are based on matching degree and noise frequency domain data, aggregation frequency domain neural network prediction power line noise characteristics are constructed, the influence of device matching degree on aggregation frequency domain neural network prediction is considered, device matching degree weight is updated in real time by utilizing aggregation frequency domain neural network prediction errors, when the prediction errors are too large, the weight of the maximum matching degree is improved when the matching degree is aggregated, the weights of other matching degrees are reduced, high-precision aggregation of the device matching degree is achieved, and the power line noise characteristic prediction accuracy is improved.
In order to implement the power line noise analysis method corresponding to the above method embodiment, corresponding functions and technical effects are achieved. Referring to fig. 2, fig. 2 shows a block diagram of a power line noise analysis apparatus according to an embodiment of the present application. For convenience of explanation, only the parts related to the present embodiment are shown, and the power line noise analysis apparatus provided in the embodiment of the present application includes:
an obtaining module 201, configured to obtain noise data generated by a plurality of power electronic devices and a noise feature data set corresponding to the power electronic device in a power line noise library, where the noise feature data set includes noise feature data of the plurality of power electronic devices;
a calculation module 202 for calculating an aggregate device match metric between the noise data and the power electronics device based on the noise data and the noise signature dataset;
the training module 203 is configured to train a preset neural network based on the aggregation device matching degree, the noise data, and the noise feature data until the preset neural network reaches a preset convergence condition, so as to obtain an aggregation frequency domain neural network;
the analysis module 204 is configured to analyze the target noise data generated by the target power electronic device by using the aggregation frequency domain neural network, so as to obtain target noise characteristic data of the target power electronic device.
In some embodiments, the calculation module 202 includes:
a calculation unit configured to calculate a device matching degree between the noise data and a plurality of kinds of the power electronic devices based on the noise data and the noise feature data set;
and the polymerization unit is used for polymerizing the device matching degrees to obtain the polymerized device matching degree.
In some embodiments, the computing unit is specifically configured to:
calculating a distance value between the noise data and the noise characteristic data of each power electronic device based on a preset device matching degree formula, wherein the distance value is the device matching degree, and the preset device matching degree formula is as follows:
wherein, theta n (i) The device matching degree between the ith noise data and the nth power electronic device is shown, x (i) is the ith noise data, u n The noise characteristic data of the nth power electronic device.
In some embodiments, the polymerization unit is specifically for:
calculating the matching degree of the aggregation device according to the matching degrees of the devices by using a preset aggregation matching degree formula, wherein the preset aggregation matching degree formula is as follows:
wherein,
aggregate device match, w, for ith noisy data n (i) The ith noise data corresponds to the device matching degree weight of the nth power electronic device n (i) The device matching degree between the ith noise data and the nth power electronic device is shown.
In some embodiments, the training module 203 comprises:
the transformation unit is used for carrying out Fourier transformation on the aggregation device matching degree, the noise data and the noise characteristic data to obtain frequency domain matching degree, noise frequency domain data and noise frequency domain characteristic data;
the training unit is used for performing iterative training on the preset neural network by taking the frequency domain matching degree and the noise frequency domain data as the input of the preset neural network;
the comparison unit is used for comparing the network prediction result output by each iteration of the preset neural network with the noise frequency domain characteristic data and calculating the frequency domain prediction error of each iteration for each iteration;
and the updating unit is used for updating the device matching degree weight and the network parameters of the preset neural network based on the frequency domain prediction error obtained by each iteration until the frequency domain prediction error is smaller than a preset threshold value, and judging that the preset neural network reaches a preset convergence condition to obtain the aggregation frequency domain neural network.
In some embodiments, the updating unit is further specifically configured to:
performing Fourier transform on the frequency domain prediction error to obtain a time domain prediction error;
updating the device matching degree weight according to the time domain prediction error by using a preset weight updating formula, wherein the preset weight updating formula is as follows:
wherein, w n (i) The ith noise data corresponds to the device matching degree weight of the nth power electronic device, alpha (i) is the update rate of the device matching degree weight, e (i) is the time domain prediction error, and theta (theta) is n (i) The device matching degree between the ith noise data and the nth power electronic device is shown.
In some embodiments, the analysis device further comprises:
and the updating module is used for updating the power line noise library based on the target noise characteristic data.
The power line noise analysis device can implement the power line noise analysis method of the method embodiment. The alternatives in the above-described method embodiments are also applicable to this embodiment and will not be described in detail here. The rest of the embodiments of the present application may refer to the contents of the above method embodiments, and in this embodiment, details are not described again.
Fig. 3 is a schematic structural diagram of a computer device according to an embodiment of the present application. As shown in fig. 3, the computer device 3 of this embodiment includes: at least one processor 30 (only one shown in fig. 3), a memory 31, and a computer program 32 stored in the memory 31 and executable on the at least one processor 30, the processor 30 implementing the steps in any of the above-described method embodiments when executing the computer program 32.
The computer device 3 may be a computing device such as a smart phone, a tablet computer, a desktop computer, and a cloud server. The computer device may include, but is not limited to, a processor 30, a memory 31. Those skilled in the art will appreciate that fig. 3 is merely an example of the computer device 3, and does not constitute a limitation of the computer device 3, and may include more or less components than those shown, or combine some of the components, or different components, such as input output devices, network access devices, etc.
The processor 30 may be a Central Processing Unit (CPU), and the processor 30 may be other general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, a discrete hardware component, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory 31 may in some embodiments be an internal storage unit of the computer device 3, such as a hard disk or a memory of the computer device 3. The memory 31 may also be an external storage device of the computer device 3 in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like provided on the computer device 3. Further, the memory 31 may also include both an internal storage unit and an external storage device of the computer device 3. The memory 31 is used for storing an operating system, an application program, a BootLoader (BootLoader), data, and other programs, such as program codes of the computer program. The memory 31 may also be used to temporarily store data that has been output or is to be output.
In addition, an embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the computer program implements the steps in any of the method embodiments described above.
The embodiments of the present application provide a computer program product, which when executed on a computer device, enables the computer device to implement the steps in the above method embodiments.
In several embodiments provided herein, it will be understood that each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application, or portions thereof, which substantially or partially contribute to the prior art, may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above-mentioned embodiments are provided to further explain the objects, technical solutions and advantages of the present application in detail, and it should be understood that the above-mentioned embodiments are only examples of the present application and are not intended to limit the scope of the present application. It should be understood that any modifications, equivalents, improvements and the like, which come within the spirit and principle of the present application, may occur to those skilled in the art and are intended to be included within the scope of the present application.
Claims (10)
1. A method for analyzing power line noise, comprising:
acquiring noise data generated by various power electronic devices and a noise characteristic data set corresponding to the power electronic devices in a power line noise library, wherein the noise characteristic data set comprises the noise characteristic data of the various power electronic devices;
calculating an aggregate device match metric between the noise data and the power electronics device based on the noise data and the noise signature dataset;
training a preset neural network based on the aggregation device matching degree, the noise data and the noise characteristic data until the preset neural network reaches a preset convergence condition to obtain an aggregation frequency domain neural network;
and analyzing the target noise data generated by the target power electronic device by using the aggregation frequency domain neural network to obtain the target noise characteristic data of the target power electronic device.
2. The power line noise analysis method of claim 1, wherein said calculating an aggregate device match between the noise data and the power electronics device based on the noise data and the noise signature data set comprises:
calculating a device matching degree between the noise data and a plurality of types of the power electronic devices based on the noise data and the noise feature data set;
and polymerizing the matching degrees of the devices to obtain the matching degree of the polymerized devices.
3. The power line noise analysis method according to claim 2, wherein said calculating a device matching degree between the noise data and a plurality of kinds of the power electronic devices based on the noise data and the noise feature data set includes:
calculating a distance value between the noise data and the noise characteristic data of each power electronic device based on a preset device matching degree formula, wherein the distance value is the device matching degree, and the preset device matching degree formula is as follows:
wherein, theta n (i) The device matching degree between the ith noise data and the nth power electronic device is shown, x (i) is the ith noise data, u n The noise characteristic data of the nth power electronic device.
4. The method for analyzing power line noise according to claim 2, wherein the aggregating the plurality of device matching degrees to obtain the aggregated device matching degree comprises:
calculating the matching degree of the aggregation device according to the matching degrees of the devices by using a preset aggregation matching degree formula, wherein the preset aggregation matching degree formula is as follows:
wherein,
aggregate device match, w, for ith noisy data n (i) The ith noise data corresponds to the device matching degree weight of the nth power electronic device n (i) The device match between the ith noise data and the nth power electronics device.
5. The method for analyzing power line noise according to claim 1, wherein the training a predetermined neural network based on the aggregation device matching degree, the noise data, and the noise feature data until the predetermined neural network reaches a predetermined convergence condition to obtain an aggregated frequency domain neural network comprises:
performing Fourier transform on the aggregation device matching degree, the noise data and the noise characteristic data to obtain frequency domain matching degree, noise frequency domain data and noise frequency domain characteristic data;
taking the frequency domain matching degree and the noise frequency domain data as the input of the preset neural network, and performing iterative training on the preset neural network;
for each iteration training, comparing a network prediction result output by the preset neural network in each iteration with the noise frequency domain characteristic data, and calculating a frequency domain prediction error of each iteration;
updating the device matching degree weight and updating the network parameters of the preset neural network based on the frequency domain prediction error obtained by each iteration until the frequency domain prediction error is smaller than a preset threshold value, and judging that the preset neural network reaches a preset convergence condition to obtain the aggregation frequency domain neural network.
6. The method for power line noise analysis according to claim 5, wherein the updating the device matching degree weights comprises:
performing Fourier transform on the frequency domain prediction error to obtain a time domain prediction error;
updating the device matching degree weight according to the time domain prediction error by using a preset weight updating formula, wherein the preset weight updating formula is as follows:
wherein, w n (i) The ith noise data corresponds to the device matching degree weight of the nth power electronic device, alpha (i) is the update rate of the device matching degree weight, e (i) is the time domain prediction error, and theta (theta) is n (i) The device matching degree between the ith noise data and the nth power electronic device is shown.
7. The method for analyzing power line noise according to claim 1, wherein after analyzing the target noise data generated by the target power electronic device by using the aggregated frequency domain neural network to obtain the target noise characteristic data of the target power electronic device, the method further comprises:
and updating the power line noise library based on the target noise characteristic data.
8. A method for analyzing noise in a power line, comprising:
the noise characteristic data set comprises noise characteristic data of various power electronic devices;
a calculation module to calculate an aggregate device match metric between the noise data and the power electronics device based on the noise data and the noise signature dataset;
the training module is used for training a preset neural network based on the aggregation device matching degree, the noise data and the noise characteristic data until the preset neural network reaches a preset convergence condition, so as to obtain an aggregation frequency domain neural network;
and the analysis module is used for analyzing the target noise data generated by the target power electronic device by utilizing the aggregation frequency domain neural network to obtain the target noise characteristic data of the target power electronic device.
9. A computer device comprising a processor and a memory for storing a computer program which, when executed by the processor, implements a power line noise analysis method as claimed in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that it stores a computer program which, when executed by a processor, implements the power line noise analysis method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211742169.5A CN115879012A (en) | 2022-12-29 | 2022-12-29 | Power line noise analysis method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211742169.5A CN115879012A (en) | 2022-12-29 | 2022-12-29 | Power line noise analysis method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115879012A true CN115879012A (en) | 2023-03-31 |
Family
ID=85757847
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211742169.5A Pending CN115879012A (en) | 2022-12-29 | 2022-12-29 | Power line noise analysis method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115879012A (en) |
-
2022
- 2022-12-29 CN CN202211742169.5A patent/CN115879012A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108197652B (en) | Method and apparatus for generating information | |
| CN113221183B (en) | Method, device and system for realizing privacy protection of multi-party collaborative update model | |
| US8204714B2 (en) | Method and computer program product for finding statistical bounds, corresponding parameter corners, and a probability density function of a performance target for a circuit | |
| JP6870508B2 (en) | Learning programs, learning methods and learning devices | |
| CN113298152B (en) | Model training method, device, terminal equipment and computer readable storage medium | |
| WO2023020307A1 (en) | Quick simulation method and apparatus for integrated circuit, and storage medium | |
| Zhao et al. | High-dimensional linear regression via implicit regularization | |
| CN111680642A (en) | Terrain classification method and device | |
| Nepomuceno et al. | On the use of interval extensions to estimate the largest Lyapunov exponent from chaotic data | |
| Pfaff et al. | Multimodal circular filtering using Fourier series | |
| CN112488297A (en) | Neural network pruning method, model generation method and device | |
| CN108228959A (en) | Using the method for Random censorship estimating system virtual condition and using its wave filter | |
| WO2022197615A1 (en) | Techniques for adaptive generation and visualization of quantized neural networks | |
| JP2019036112A (en) | Abnormal sound detector, abnormality detector, and program | |
| CN113158550A (en) | Method and device for federated learning, electronic equipment and storage medium | |
| CN111565065B (en) | Unmanned aerial vehicle base station deployment method and device and electronic equipment | |
| CN115035017A (en) | Cell density grouping method, device, electronic apparatus and storage medium | |
| CN116365511A (en) | Active power distribution network model construction method, device, terminal and storage medium | |
| CN115879012A (en) | Power line noise analysis method, device, equipment and storage medium | |
| CN111401569A (en) | Hyper-parameter optimization method and device and electronic equipment | |
| CN109388784A (en) | Minimum entropy Density Estimator device generation method, device and computer readable storage medium | |
| CN114970732A (en) | Posterior calibration method and device for classification model, computer equipment and medium | |
| CN110647805B (en) | Reticulate pattern image recognition method and device and terminal equipment | |
| CN115688873A (en) | Graph data processing method, device and computer program product | |
| CN113341373A (en) | Positioning method and device and electronic equipment |
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 |