CN114295946B - Successive approximation solving method for multi-sample signals of multi-ultrahigh frequency sensor group - Google Patents
Successive approximation solving method for multi-sample signals of multi-ultrahigh frequency sensor group Download PDFInfo
- Publication number
- CN114295946B CN114295946B CN202111651005.7A CN202111651005A CN114295946B CN 114295946 B CN114295946 B CN 114295946B CN 202111651005 A CN202111651005 A CN 202111651005A CN 114295946 B CN114295946 B CN 114295946B
- Authority
- CN
- China
- Prior art keywords
- sample
- sensor
- positioning
- initial value
- successive approximation
- 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.)
- Active
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Position Fixing By Use Of Radio Waves (AREA)
Abstract
The invention provides a multi-sample signal successive approximation solving method of a multi-ultrahigh frequency sensor group, which comprises the following operation steps: s01), arranging and combining the UHF sensor arrays; s02), acquiring N groups of PD sample data at different moments for PD sources at the same position; s03), estimating a positioning initial value corresponding to each PD sample data; s04), respectively giving the midpoint coordinates of each sample initial value point and four sensor position connecting line segments, grouping the initial value points and establishing an optimizing target function based on the length sum of the central lines; s05), carrying out global search on the minimum value of each group of optimizing objective functions, taking the solution obtained in the last step as the initial value of the next optimization, and repeating the operations S04) and S05) until the optimal solution is obtained; s06), repeating the operations of S04) and S05) again to obtain a final globally optimal solution. The invention carries out recursive approximation processing on the optimizing target function established by the sample value obtained by positioning and the sensor array based on the thought of successive approximation, thereby obtaining more accurate PD source position.
Description
Technical Field
The invention relates to the technical field of partial discharge positioning of power equipment, in particular to a multiple sample signal successive approximation solving method of a multiple ultrahigh frequency sensor group, and belongs to recursive approximation processing of a plurality of positioning results obtained by positioning calculation in a transformer partial discharge ultrahigh frequency positioning method.
Background
Partial discharge (Partial Discharge, PD) is a discharge phenomenon generated in a local range when an insulation defect occurs in an electrical device, and how to accurately and reliably locate a fault of a partial discharge signal is one of key problems in the field of monitoring the state of the electrical device. The ultra-high frequency (Ultra High Frequency, UHF) detection technology has the advantages of high sensitivity, strong anti-interference capability and the like, and is widely applied to PD detection and positioning work in recent years. For the traditional positioning method, due to the influence of adverse factors such as field strength electromagnetic interference, a complex multilayer medium structure of a transformer and the like, no matter what anti-interference technology and a time difference extraction algorithm are adopted, the PD signal time difference without delay errors is difficult to accurately obtain, and meanwhile, the problems of local convergence or divergence and the like possibly occurring in the solving process of the positioning algorithm are solved, so that the method for taking the result obtained by single positioning as the PD source space coordinate position is difficult to accurately position the transformer. In the current application, an averaging method is generally adopted to process the multi-sample PD source estimated value obtained by positioning calculation, but the method is only simple to average the positioning result and cannot solve the problem of large positioning error.
Patent document with publication number of CN113092956A discloses a PD source positioning algorithm based on a gradient approximation type dry reactor, which is improved by a traditional grid search Newton iteration method, and the improved algorithm is in a ladder shape and comprises a ladder 1 and a ladder 2; step 1, coarsely positioning a PD source by utilizing a grid search Newton iteration method; step 2 performs distance sum value minimum value approximation optimization on the initial positioning coordinate X, Y, and replaces X, Y after the optimization into a time difference positioning equation to obtain a Z coordinate. The improved algorithm reduces the positioning error of the Z coordinate to a certain extent, improves the PD source positioning accuracy of the dry reactor, however, because electromagnetic wave propagation is carried out at an extremely high speed which is slightly lower than the speed of light, the tiny time delay errors can cause larger positioning errors and lead to positioning failure, the calculation of the positioning equation set has very high dependence on time delay, and the tiny errors can lead to larger positioning errors and even cause the situation that the time difference equation set has no solution and positioning failure.
Patent publication No. CN113030660A discloses a GIS partial discharge positioning method based on truncated singular value decomposition, which comprises the following steps: s1: establishing a mathematical model of partial discharge positioning based on the arrival time difference; s2: obtaining a linear positioning equation set PX=Q by a spherical conversion method; s3: carrying out centralized pretreatment on the linear positioning equation set PX=Q to obtain a centralized positioning equation set PX=Q'; s4: decomposing and calculating the matrix P by applying a principal component triangle decomposition method to obtain an equivalent matrix P' of the matrix P; s5: substituting the matrix P 'into the set of positioning equations px=q' to obtain an equivalent set of linear equations: p 'x=q'; s6: singular value decomposition is performed on the matrix P 'to obtain two orthogonal matrices U= (U1, …, un) and V= (V1, …, vn) of the matrix P', and a filter factor fi is designed; s7: and solving the P 'X=Q' by using a truncated singular value decomposition method. According to the method, PD signals at different moments in the GIS are positioned, however, when ultrasonic signals propagate in electrical equipment, the attenuation influence of various acoustic media on the signals is very serious, and the PD signals are difficult to effectively detect due to low sensitivity of an ultrasonic sensor, so that the positioning is inaccurate or cannot be performed.
Disclosure of Invention
The invention provides a multi-sample signal successive approximation solving method of a multi-ultrahigh frequency sensor group, which is used for carrying out recursive approximation processing on an optimizing objective function established by a sample value obtained by positioning and a sensor array based on the thought of successive approximation, so as to obtain a more accurate PD source position.
In order to solve the technical problems, the invention adopts the following technical scheme:
a successive approximation solving method for multi-sample signals of a multi-ultrahigh frequency sensor group comprises the following steps:
s01), UHF sensor arrays S0, S1, …, sk, …, sn (n)>4) Are arranged and combined in groups of 4, and are together
A seed combination form;
s02), acquiring N groups of PD sample data at different moments for a PD source at the same position through UHF sensor arrays S0, S1 …, sk, … and Sn (N > 4);
s03), estimating a positioning initial value P corresponding to each PD sample data by adopting a traditional TDOA (Time Difference of Arrival ) positioning algorithm ij (x ij ,y ij ,z ij ) Wherein: i represents the number of UHF sensor combinations of the i-th UHF sensor combination (i=1, 2, …,
) J represents the j-th set of PD sample data (j=1, 2, …, N) acquired under the specific sensor combination;
s04) respectively giving initial value points P of each sample under each sensor combination ij (x ij ,y ij ,z ij ) Midpoint coordinates of the line segments connected with the four sensor positions are aligned with the initial positions according to the sample sequenceGrouping the value points and establishing an optimizing objective function based on the length sum of the central lines;
s05), performing global search on minimum d of each group of optimizing objective functions by adopting a convolutional neural network method min Repeating S04) and S05) until the optimal solution P under each sensor combination is obtained by taking the solution obtained in the last step as the initial value of the next optimization i (x i ,y i ,z i )，i＝1,2,…,
S06), repeating again the operations of S04) and S05), for the optimal solution P under each sensor combination i (x i ,y i ,z i ) And establishing an optimizing objective function again, and obtaining a final global optimal solution P (x, y, z) by adopting a convolutional neural network method.
Further, step S04) specifically includes: n sample positioning initial value points can be calculated by utilizing time delay information corresponding to N groups of PD signals captured by each sensor combination form at different moments, and the N initial value points are scattered around a real PD source randomly due to the existence of time delay errors.
Further, 4 sensor coordinates S in each combination k (x k ,y k ,z k ) And an initial value point P ij (x ij ,y ij ,z ij ) The midpoint coordinates of the wire segments are:
further, according to the sample sequence, each m (N is more than or equal to m is more than or equal to 2) initial value points are taken as a group, and the distance and the function between the position of the PD source to be solved and the midpoint of 4m straight line segments in the sample group are constructed
The problem of optimizing the set of sample values is to perform minimum value optimization on the above equation.
Step S05) specifically comprises the following five steps:
s51) initializing a weight of the network;
s52) forward transmission of input data through a convolution layer, a downsampling layer and a full connection layer to obtain an output value;
s53) obtaining an error between the output value and the target value of the network;
s54) when the error is larger than the expected value, transmitting the error back to the network, sequentially obtaining the errors of the full connection layer, the downsampling layer and the convolution layer, updating the weight according to the obtained errors, and returning to the second step;
s55) when the error is equal to or smaller than the expected value, the operation is ended, and the global optimal solution is output, which is regarded as the sought PD source position.
Due to the diversity and complexity of UHF electromagnetic wave propagation environments in electrical equipment, the acquisition of positioning parameters is easily interfered by various factors in the wireless signal propagation process. Among the most dominant positioning influencing factors are two: one is a complex transformer structure, so that obstacles such as iron cores are diffracted in a discharge signal propagation path, and therefore, the signals received by all sensors cannot be guaranteed to be wave front reflection of the signals arriving from a PD source through a linear propagation path, and even the signals possibly received are signals overlapped by multipath propagation; the other is background noise interference, including complex strong electromagnetic interference existing in the field, so that the sensor receives PD signals mixed with background noise, especially mixed PD signals, so that no matter what anti-interference technology is used, it is difficult to accurately extract the time difference without delay error. In addition, the response characteristics of the sensor, and the different types of insulation defects have different rising edges of the PD pulses, and the like, all of which cause errors in the detected UHF PD signal characteristic measurement value, and the error is amplified in multiple steps, so that the positioning accuracy is seriously affected. Most of the current PD positioning error suppression methods are to perform positioning calculation by taking an average value of a plurality of time delay samples or to perform positioning calculation by taking an average value after positioning respective values of a plurality of samples, which cannot solve the problem of overlarge positioning error well. The result obtained by adopting the average time difference method can not accurately judge the fault point, and the method adopting the average positioning result is better than the average time difference method, but the problem of low positioning precision is not fundamentally solved because the method is only simple operation on the positioning result.
Compared with the related art, the invention has the following beneficial effects:
firstly, arranging and combining UHF sensor arrays to obtain PD sample data; then, a traditional TDOA positioning algorithm is adopted to perform preliminary positioning on the sample data, and a positioning initial value is obtained; establishing an optimizing objective function based on the length sum of the central line by utilizing the positioning initial value and the four sensor positions; and (3) performing global searching for the minimum value of each group of optimizing objective functions by adopting a convolutional neural network method, taking the solution obtained in the last step as the initial value of the next optimization, and repeating the operation until the optimal solution under each sensor combination is obtained, namely the PD source position. Under the condition of larger positioning initial value error, the PD source can be positioned more accurately after the optimizing treatment, and the positioning precision after the optimizing treatment is improved greatly. The invention carries out recursive approximation treatment on the optimizing target function established by the sample value obtained by positioning and the sensor array based on the thought of successive approximation, thereby obtaining more accurate PD source position, solving the technical problems of sensitive delay error, easy local convergence or divergence, low operation speed and the like of the existing positioning algorithm, simultaneously reducing the calculation complexity of the algorithm, improving the positioning efficiency and achieving the effect of accurately and rapidly positioning the PD source.
Drawings
The present invention will be described in further detail with reference to the accompanying drawings.
FIG. 1 is a flow chart of a convolutional neural network algorithm of the present invention;
FIG. 2 is a block diagram of a partial discharge positioning system of a multiple UHF sensor according to the present invention;
fig. 3 is a schematic diagram of the successive approximation positioning principle based on the centerline length and the establishment of the optimizing objective function according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention are clearly and completely described below with reference to fig. 1 to 3. It will be apparent that the described embodiments are some, but not all, embodiments of the invention. All other embodiments, which are obtained by a person skilled in the art based on the described embodiments of the invention, fall within the scope of protection of the invention.
A successive approximation solving method for multi-sample signals of a multi-ultrahigh frequency sensor group comprises the following steps:
s01), UHF sensor arrays S0, S1, …, sk, …, sn (n)>4) Are arranged and combined in groups of 4, and are together
A seed combination form;
s02), acquiring N groups of PD sample data at different moments for a PD source at the same position through UHF sensor arrays S0, S1 …, sk, … and Sn (N > 4);
s03), estimating a positioning initial value P corresponding to each PD sample data by adopting a traditional TDOA (Time Difference of Arrival ) positioning algorithm ij (x ij ,y ij ,z ij ) Wherein: i represents the number of UHF sensor combinations of the i-th UHF sensor combination (i=1, 2, …,
) J represents the j-th set of PD sample data (j=1, 2, …, N) acquired under the specific sensor combination;
s04) respectively giving initial value points P of each sample under each sensor combination ij (x ij ,y ij ,z ij ) Grouping initial value points according to sample sequences and establishing an optimizing target function based on the length sum of the central lines by the midpoint coordinates of the connecting line segments of the four sensor positions; n sample positioning initial value points can be calculated by utilizing time delay information corresponding to N groups of PD signals captured by each sensor combination form at different moments, and the N initial value points are scattered randomly around a real PD source due to the existence of time delay errors; the coordinates S of 4 sensors in each combination form k (x k ,y k ,z k ) And an initial value point P ij (x ij ,y ij ,z ij ) The midpoint coordinates of the wire segments are:
according to the sample sequence, every m (N is more than or equal to m is more than or equal to 2) initial value points are taken as a group, and the distance and the function of the position of the PD source to be solved to the midpoint of 4m straight line segments in the sample group are constructed>
The optimizing problem of the group of sample values is to perform minimum value optimizing on the sample values;
s05), performing global search on minimum d of each group of optimizing objective functions by adopting a convolutional neural network method min Repeating S04) and S05) until the optimal solution P under each sensor combination is obtained by taking the solution obtained in the last step as the initial value of the next optimization i (x i ,y i ,z i )，i＝1,2,…,
Step S05) specifically comprises the following five steps:
s51) initializing a weight of the network;
s52) forward transmission of input data through a convolution layer, a downsampling layer and a full connection layer to obtain an output value;
s53) obtaining an error between the output value and the target value of the network;
s54) when the error is larger than the expected value, transmitting the error back to the network, sequentially obtaining the errors of the full connection layer, the downsampling layer and the convolution layer, updating the weight according to the obtained errors, and returning to the second step;
s55) when the error is equal to or smaller than the expected value, ending the operation, and outputting a global optimal solution, wherein the global optimal solution is regarded as the position of the PD source;
s06), repeating again the operations of S04) and S05), for the optimal solution P under each sensor combination i (x i ,y i ,z i ) And establishing an optimizing objective function again, and obtaining a final global optimal solution P (x, y, z) by adopting a convolutional neural network method.
Examples
The embodiment discloses a multi-sample signal successive approximation solving method of a multi-ultrahigh frequency sensor group, which is used for improving ultrahigh frequency positioning accuracy of a partial discharge source of a transformer. In consideration of the problems that the single positioning result is low in precision and the average method of the PD source estimated values of multiple samples is insufficient in precision, the successive approximation solving method for the multiple sample signals of the multiple ultrahigh frequency sensor group is provided.
Setting 6 UHF sensors, carrying out recursive approximation processing on an optimizing objective function established by a sample value obtained by positioning and a sensor array based on the thought of successive approximation, wherein a structure diagram of a partial discharge positioning system of the 6 UHF sensor is shown in fig. 3, and the specific solving steps are as follows:
s01), the UHF sensor arrays S0, S1, … and S5 are arranged and combined in groups of 4, and the two are shared
A seed combination form;
s02), 1 group of PD sample data is acquired at different time points for the PD source at the same position through UHF sensor arrays S0, S1, … and S5;
s03), estimating corresponding positioning initial values under 15 sensor combinations by adopting a traditional TDOA (Time Difference of Arrival, arrival time difference) positioning algorithm;
s04), calculating midpoint coordinates of corresponding sample initial value points and corresponding four sensor position connecting line segments under 15 sensor combinations, and establishing an optimizing target function based on a center line length sum for each 3 initial value points according to an arrangement sequence (a successive approximation positioning principle based on the center line length sum for establishing the optimizing target function is shown in figure 3);
s05), performing global search on minimum d of each group of optimizing objective functions by adopting a convolutional neural network method min Repeating S04) and S05) operations by taking the solution obtained in the last step as an initial value of the next optimization to obtain a final global optimal solution (less than 3 data are self-organized in a group during grouping operation, and the grouping calculation process is shown in the following table 1);
table 1 6 grouping calculation procedure under multiple UHF sensors (15 grouping combinations)
Finally, it should be noted that the foregoing description is only an embodiment of the present invention, and is not intended to limit the scope of the present invention, and all equivalent structures or equivalent processes using the descriptions of the present invention and the accompanying drawings, or direct or indirect application in other related technical fields, are included in the scope of the present invention.
Claims (5)
1. The successive approximation solving method for the multi-sample signals of the multi-ultrahigh frequency sensor group is characterized by comprising the following operation steps of:
s01), UHF sensor arrays S0, S1, & gt, sk, & gt, sn, n >4 are arranged and combined in groups of 4, and the two are combined together
A seed combination form;
s02), acquiring N sets of PD sample data at different times for a same location PD source through the UHF sensor array S0, S1., sk,., sn, N > 4;
s03), estimating a positioning initial value P corresponding to each PD sample data by adopting a traditional TDOA positioning algorithm ij (x ij ，y ij ，z ij ) Wherein: i represents an ith UHF sensor combination format,
j represents the j-th set of PD sample data acquired under a specific sensor combination, j=1, 2, …, N;
s04) respectively giving initial value points P of each sample under each sensor combination ij (x ij ,y ij ,z ij ) Grouping initial value points according to sample sequences and establishing an optimizing target function based on the length sum of the central lines by the midpoint coordinates of the connecting line segments of the four sensor positions;
s05), adopting a convolutional neural network methodGlobal search of minimum d for each set of optimizing objective functions min Repeating S04) and S05) until the optimal solution P under each sensor combination is obtained by taking the solution obtained in the last step as the initial value of the next optimization i (x i ,y i ,z i )，
S06), repeating again the operations of S04) and S05), for the optimal solution P under each sensor combination i (x i ,y i ,z i ) And establishing an optimizing objective function again, and obtaining a final global optimal solution P (x, y, z) by adopting a convolutional neural network method.
2. The multiple ultrahigh frequency sensor group multiple sample signal successive approximation solving method according to claim 1, wherein: in step S04), N sample positioning initial points P can be calculated by using the time delay information corresponding to N sets of PD signals captured at different times by each sensor combination 1 ,P 2 ,...,P N The presence of delay errors will cause the N initial points to be randomly dispersed around the true PD source.
3. The multiple ultrahigh frequency sensor group multiple sample signal successive approximation solving method according to claim 1, wherein: in step S04), 4 sensor coordinates S for each combination k (x k ,y k ,z k ) And locating the calculated sample initial value point P ij (x ij ,y ij ,z ij ) The midpoint coordinates of the wire segments are:
4. the multiple uhf sensor set multiple sample signal successive approximation solution method of claim 3, wherein: in the step S04), every m initial value points are a group according to the sample sequence, N is more than or equal to m is more than or equal to 2, and the real position P (x, y, z) of the PD source to be solved is constructed until the samples are dividedDistance sum function of midpoints of 4m straight line segments in group
The problem of optimizing the set of sample values is to perform minimum value optimization on the above equation.
5. The successive approximation solving method for the multi-sample signals of the multi-ultrahigh frequency sensor group according to claim 4, wherein the method comprises the following steps of: step S05) comprises the steps of:
s51) initializing a weight of the network;
s52) forward transmission of input data through a convolution layer, a downsampling layer and a full connection layer to obtain an output value;
s53) obtaining an error between the output value and the target value of the network;
s54) when the error is larger than the expected value, transmitting the error back to the network, sequentially obtaining the errors of the full connection layer, the downsampling layer and the convolution layer, updating the weight according to the obtained errors, and returning to the second step;
s55) when the error is equal to or smaller than the expected value, the operation is ended, and the global optimal solution is output, which is regarded as the sought PD source position.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111651005.7A CN114295946B (en) | 2021-12-30 | 2021-12-30 | Successive approximation solving method for multi-sample signals of multi-ultrahigh frequency sensor group |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111651005.7A CN114295946B (en) | 2021-12-30 | 2021-12-30 | Successive approximation solving method for multi-sample signals of multi-ultrahigh frequency sensor group |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114295946A CN114295946A (en) | 2022-04-08 |
| CN114295946B true CN114295946B (en) | 2023-07-04 |
Family
ID=80972646
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111651005.7A Active CN114295946B (en) | 2021-12-30 | 2021-12-30 | Successive approximation solving method for multi-sample signals of multi-ultrahigh frequency sensor group |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114295946B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105676085A (en) * | 2016-01-31 | 2016-06-15 | 国家电网公司 | Extra-high voltage GIS partial discharge detection method based on multi-sensor information fusion |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108536648B (en) * | 2018-03-30 | 2021-07-06 | 武汉大学 | Partial discharge nonlinear model conversion solving and optimizing method based on multiple ultrasonic sensors |
-
2021
- 2021-12-30 CN CN202111651005.7A patent/CN114295946B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105676085A (en) * | 2016-01-31 | 2016-06-15 | 国家电网公司 | Extra-high voltage GIS partial discharge detection method based on multi-sensor information fusion |
Non-Patent Citations (1)
| Title |
|---|
| 一种多样本信息的局部放电源逐次逼近定位方法;唐炬;黄亮;曾福平;张晓星;;电工技术学报(10);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114295946A (en) | 2022-04-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108536648B (en) | Partial discharge nonlinear model conversion solving and optimizing method based on multiple ultrasonic sensors | |
| CN112147573A (en) | Passive positioning method based on amplitude and phase information of CSI (channel State information) | |
| CN107124762B (en) | Wireless positioning method for efficiently eliminating non-line-of-sight errors | |
| CN109991519B (en) | Partial discharge direction-finding method and system based on neural network and wireless sensor array | |
| CN111126318A (en) | Parameter-adjustable double-subspace signal detection method under signal mismatch | |
| CN109490728B (en) | Regularization-based transformer substation partial discharge positioning method | |
| CN111157943A (en) | TOA-based sensor position error suppression method in asynchronous network | |
| CN114241272B (en) | Heterogeneous information fusion positioning method based on deep learning | |
| CN112526602A (en) | P wave arrival time picking method based on long and short time windows and AR model variance surge effect | |
| CN108761384B (en) | Target positioning method for robust sensor network | |
| CN106093925B (en) | Multi-radar joint detection method | |
| CN114295946B (en) | Successive approximation solving method for multi-sample signals of multi-ultrahigh frequency sensor group | |
| CN111770528B (en) | Visual distance and non-visual distance identification method and device based on channel parameter extraction method | |
| CN109752633B (en) | Method and system for positioning partial discharge signal of transformer substation | |
| CN111859241A (en) | Unsupervised sound source orientation method based on sound transfer function learning | |
| Gadhgadhi et al. | Distance estimation using polynomial approximation and neural network based on rssi technique | |
| CN112261578B (en) | Indoor fingerprint positioning method based on mode filtering | |
| CN112986967A (en) | Self-adaptive detection method and device for reverberation edge | |
| Wei et al. | Angle–of–Arrival (AoA) Factorization in Multipath Channels | |
| CN111929671A (en) | Method and system for estimating receiving amplitude of radio detection based on normal matching | |
| CN112399366A (en) | Indoor positioning method based on Hankel matrix and WKNN variance extraction | |
| CN110988800A (en) | Semi-positive relaxation positioning method based on acoustic energy | |
| CN113687306B (en) | Multi-frequency synchronous two-dimensional off-grid compressed beam forming sound source identification method | |
| Guofeng et al. | Partial Discharge Positioning Based on Statistical Analysis of UHF RSSI Values | |
| CN115226202B (en) | Positioning base station screening method based on maximum mutual information |
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 |