CN116702064A - Method, system, storage medium and equipment for estimating operation behavior of electric power tool - Google Patents
Method, system, storage medium and equipment for estimating operation behavior of electric power tool Download PDFInfo
- Publication number
- CN116702064A CN116702064A CN202310609543.2A CN202310609543A CN116702064A CN 116702064 A CN116702064 A CN 116702064A CN 202310609543 A CN202310609543 A CN 202310609543A CN 116702064 A CN116702064 A CN 116702064A
- Authority
- CN
- China
- Prior art keywords
- attitude
- electric power
- power tool
- parameters
- angle
- 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
- 238000000034 method Methods 0.000 title claims abstract description 46
- 230000006399 behavior Effects 0.000 claims abstract description 66
- 239000013598 vector Substances 0.000 claims abstract description 56
- 238000007637 random forest analysis Methods 0.000 claims abstract description 51
- 238000012549 training Methods 0.000 claims abstract description 23
- 230000004927 fusion Effects 0.000 claims description 42
- 239000011159 matrix material Substances 0.000 claims description 29
- 238000012360 testing method Methods 0.000 claims description 16
- 230000003044 adaptive effect Effects 0.000 claims description 14
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 238000010276 construction Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 7
- 238000001914 filtration Methods 0.000 claims description 7
- 238000012952 Resampling Methods 0.000 claims description 6
- 230000007704 transition Effects 0.000 claims description 5
- 230000002349 favourable effect Effects 0.000 abstract description 3
- 238000005259 measurement Methods 0.000 description 9
- 239000004973 liquid crystal related substance Substances 0.000 description 6
- 230000003068 static effect Effects 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000005484 gravity Effects 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- 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
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the application discloses a method for estimating the operation behavior of an electric power tool, which comprises the following steps: acquiring attitude parameters of the electric tool and determining logarithmic energy entropy according to the attitude parameters; acquiring a weighted feature vector according to the attitude parameters and the logarithmic energy entropy of the electric power tool; constructing a feature data set according to the weighted feature vector; training a random forest classification model according to the characteristic data set; the method can acquire the operation behaviors of the current electric power tools, realize judgment of dangerous operation behaviors in the electric power working scene, and is favorable for timely finding and eliminating potential safety hazards of the electric power working scene.
Description
Technical Field
The application relates to the technical field of communication and data processing of the Internet of things, in particular to a method, a system, a storage medium and equipment for estimating operation behaviors of electric tools and appliances.
Background
In a plurality of electric power operation site monitoring means, the operation behavior analysis of the electric power tools is helpful for enhancing electric power operation site monitoring, and timely finding and eliminating potential safety hazards has important value for guaranteeing stable production and safety of operators.
In the prior art, when the attitude of a tool is estimated by using a sensor gyroscope, a magnetometer, an accelerometer and the like, the calculation errors can be accumulated along with the time to cause low reliability of a navigation attitude angle output by the accelerometer and the magnetometer in a combined way under the rapid motion state of the power tool, so that the problem of distortion in the detection of the attitude parameters of the tool easily occurs, the attitude of the tool is estimated inaccurately, and the judgment of dangerous operation behaviors is further influenced.
Disclosure of Invention
In view of the above, it is necessary to provide a method for estimating the operation behavior of an electric power tool.
A method of estimating a working behavior of an electrical power tool, the method comprising the steps of:
acquiring attitude parameters of the electric power tool;
determining a logarithmic energy entropy according to the attitude parameters;
acquiring a weighted feature vector according to the attitude parameters and the logarithmic energy entropy of the electric power tool;
constructing a feature data set according to the weighted feature vector;
training a random forest classification model according to the characteristic data set;
and determining the operation behavior of the current electric power tool according to the random forest classification model.
In the above scheme, the attitude parameters comprise position parameters and attitude angle parameters, wherein the position parameters are collected by a GPS module, and the attitude angle parameters are fused according to a dual-channel self-adaptive weighting fusion filter.
In the above scheme, the attitude angle parameter includes a pitch angle θ 0 Roll angle gamma 0 Heading angle psi 0 The method comprises the steps of carrying out a first treatment on the surface of the The location parameter includes longitude L on Latitude L at And height A lt 。
In the above scheme, the attitude angle parameters are fused according to the two-channel self-adaptive weighting fusion filter, and specifically include:
acquiring a first group of pitch angle, roll angle and course angle according to a gyroscope on an electric power tool;
acquiring a second group of pitch angle, roll angle and course angle according to a magnetometer, an accelerometer and a GPS module on the electric power tool;
inputting two groups of pitch angle, roll angle and course angle parameters into a double-channel self-adaptive weighting fusion filter for fusion;
the dual-channel self-adaptive weighted fusion filter outputs a fusion attitude angle: pitch angle theta 0 Roll angle gamma 0 Heading angle psi 0 。
In the above scheme, the noise variance value of the dual-channel adaptive weighted fusion filter is determined according to the singular value of the square interpolation resampling matrix.
In the above scheme, the dual-channel adaptive weighted fusion filter comprises two parts: and establishing a double-channel third-order discrete random system and estimating attitude angle parameters by utilizing an optimal fusion algorithm.
In the above scheme, the establishing a dual-channel third-order discrete random system specifically includes:
x 0 (k+1)＝Φx 0 (k)+w 0 (k)
y 1 (k)＝H 1 x 0 (k)+v 1 (k)
y 2 (k)＝H 2 x 0 (k)+v 2 (k)
wherein x is 0 (k) Is the estimated result of the two groups of fused attitude angle parameters, x 0 (k+1) is the estimation result of two groups of fused attitude angle parameters at time k+1, phi is a state transition matrix and H 1 And H 2 For observing matrix, w 0 (k) V is the system noise vector 1 (k) And v 2 (k) Respectively two groups of measured noise vectors, y 1 And y 2 And respectively fusing the estimated attitude angle parameters of the first two groups.
In the above scheme, the estimating of the attitude angle parameter by using the optimal fusion algorithm specifically includes: determining predicted values of the two groups of attitude angles at the moment K-1 according to the measured values of the two groups of attitude angles at the moment K-1; acquiring prediction errors of two groups of attitude angles at the moment K-1, and determining the filtering gain of the attitude angles according to the prediction errors; acquiring predicted values of the two groups of attitude angles at the moment K according to the predicted values of the two groups of attitude angles at the moment K-1 and the filtering gain; determining the prediction errors of the two groups of attitude angles at the K moment and a covariance matrix of the prediction errors according to the prediction errors of the two groups of attitude angles at the K-1 moment; determining weight indexes of the predicted values of the two groups of attitude angles at the moment K according to the predicted errors of the two groups of attitude angles at the moment K-1 and a covariance matrix of the predicted errors; and determining final estimated parameters of the attitude angles at the moment K according to the predicted values and the weight indexes of the two groups of attitude angles at the moment K.
In the above scheme, the determining the logarithmic energy entropy according to the gesture parameter specifically includes: the logarithmic energy entropy of the pose parameters is obtained using the following formula:
L oe ＝log(θ 0 ) 2 +log(γ 0 ) 2 +log(ψ 0 ) 2 +log(L on ) 2 +log(L at ) 2 +log(A lt ) 2 the method comprises the steps of carrying out a first treatment on the surface of the Wherein L is oe Is the logarithmic energy entropy.
In the above scheme, the obtaining the weighted feature vector according to the gesture parameter and the logarithmic energy entropy of the electric tool specifically includes:
wherein F is V Is a weighted feature vector.
In the above scheme, the training random forest classification model according to the feature data set specifically includes:
inputting the weighted feature vector as training data into a random forest classification model;
identifying the training data using the random forest classification model;
and classifying the training data until the accuracy of the random forest classification model is more than or equal to 0.95.
In the above scheme, the determining the operation behavior of the current power tool according to the random forest classification model further includes: the operation behavior of the electric power tool is as follows: lifting, raising, lowering and lowering.
In the above scheme, the determining the operation behavior of the current electric power tool according to the random forest classification model specifically includes:
inputting the weighted feature vector as test data into a random forest classification model;
identifying the test data by utilizing the random forest classification model, and acquiring electric power operation behavior elements represented by the test data;
classifying the test data according to the random forest classification model to obtain a classification result;
and outputting the test data type label, namely the operation behavior of the electric power tool according to the classification result.
The application also provides a system for estimating the operation behavior of the electric power tool, which is characterized by comprising: the system comprises a gesture parameter acquisition unit, a logarithmic energy entropy acquisition unit, a weighted feature vector acquisition unit, a data set construction unit, a model construction unit and a behavior estimation unit;
the gesture parameter acquisition unit is used for acquiring gesture parameters of the electric power tool;
the logarithmic energy entropy acquisition unit is used for acquiring logarithmic energy entropy;
the weighted feature vector acquisition unit is used for acquiring weighted feature vectors;
the data set construction unit is used for acquiring a characteristic data set;
the model construction unit is used for constructing a random forest classification model;
the behavior estimation unit is used for determining the operation behavior of the current power tool.
The application also proposes a readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the steps of:
acquiring attitude parameters of the electric power tool;
determining a logarithmic energy entropy according to the attitude parameters;
acquiring a weighted feature vector according to the attitude parameters and the logarithmic energy entropy of the electric power tool;
constructing a feature data set according to the weighted feature vector;
training a random forest classification model according to the characteristic data set;
and determining the operation behavior of the current electric power tool according to the random forest classification model.
The application also proposes a computer device comprising a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to perform the steps of: acquiring attitude parameters of the electric power tool;
determining a logarithmic energy entropy according to the attitude parameters;
acquiring a weighted feature vector according to the attitude parameters and the logarithmic energy entropy of the electric power tool;
constructing a feature data set according to the weighted feature vector;
training a random forest classification model according to the characteristic data set;
and determining the operation behavior of the current electric power tool according to the random forest classification model.
The embodiment of the application has the following beneficial effects: acquiring attitude parameters of the electric tool and determining logarithmic energy entropy according to the attitude parameters; acquiring a weighted feature vector according to the attitude parameters and the logarithmic energy entropy of the electric power tool; constructing a feature data set according to the weighted feature vector; training a random forest classification model according to the characteristic data set; the method can acquire the operation behaviors of the current electric power tools, realize judgment of dangerous operation behaviors in the electric power working scene, and is favorable for timely finding and eliminating potential safety hazards of the electric power working scene.
Drawings
In order to more clearly illustrate the embodiments of the application or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Wherein:
FIG. 1 is a flow chart of a method for estimating operation behavior of an electrical tool according to one embodiment;
FIG. 2 is a flow chart of a method for acquiring attitude angle parameters based on a dual-channel adaptive weighted fusion filter in one embodiment;
FIG. 3 is a schematic flow chart of estimating attitude angle parameters by using an optimal fusion algorithm in one embodiment;
FIG. 4 is a flow diagram of training a random forest classification model based on a feature data set in one embodiment;
FIG. 5 is a flow diagram of determining the operational behavior of a current power tool based on a random forest classification model in one embodiment.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments.
All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
In the following description, numerous specific details are set forth in order to provide a more thorough understanding of the present application. It will be apparent, however, to one skilled in the art that the application may be practiced without one or more of these details.
In other instances, well-known features have not been described in detail in order to avoid obscuring the application. It should be understood that the present application may be embodied in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the application to those skilled in the art.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used herein, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise, and it is to be understood that the terms "comprises" and/or "comprising" when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. As used herein, the term "and/or" includes any and all combinations of the associated listed items.
In order to provide a thorough understanding of the present application, detailed structures will be presented in the following description in order to illustrate the technical solutions presented by the present application. Alternative embodiments of the application are described in detail below, however, the application may have other implementations in addition to these detailed descriptions.
As shown in fig. 1, in one embodiment, there is provided a method of estimating operation behavior of an electric power tool, the method of estimating operation behavior of an electric power tool including steps S101 to S106, which are described in detail as follows:
s101, acquiring attitude parameters of an electric tool;
in some embodiments, the attitude parameters include a position parameter and an attitude angle parameter, wherein the position parameter is acquired by a GPS module and the attitude angle parameter is fused according to a dual-channel adaptive weighted fusion filter.
Preferably, the attitude angle parameter includes a pitch angle θ 0 Roll angle gamma 0 Heading angle psi 0 The location parameter includes longitude L on Latitude L at And height A lt 。
Because the electric power tool does not move rapidly in the use scene, the change speed of longitude, latitude and altitude parameters is slow, so that the longitude L of the electric power tool can be directly acquired according to the GPS acquisition module on Latitude L at And height A lt And outputting as an estimation result.
The method disclosed by the application also provides longitude, latitude and altitude position information of the electric power tool on the basis of the known attitude angle parameters, and can provide more detailed and more accurate motion state parameters for analysis of electric power operation behavior elements.
Under the existing working scene of the electric power tool, if only a gyroscope is used for estimating the attitude angle parameters, the estimated attitude angle drift accuracy is lower due to the drift characteristic of the gyroscope; if only the accelerometer and the magnetometer are used for solving the attitude angle parameters, the accuracy of the static parameters obtained by the method is high, but the static parameters are easily interfered by external vibration, and the accuracy of the dynamic parameters is difficult to guarantee.
The single-channel sensors are easy to interfere by the working measurement environment of the electric power tools and the motion state of the electric power tools in the working process, the problem of distortion in the detection of the gesture parameters of the tools and the problem of missing position information can occur, and both factors can cause the deviation of the motion gesture information and the position information in the analysis process of the behavior elements of the electric power tools.
The dual-channel self-adaptive weighted fusion filter provided by the application can carry out weighted fusion on the attitude angle estimated by the gyroscope and the attitude angles estimated by the accelerometer, the magnetometer and the GPS module, and can accurately estimate the attitude angle parameters of the electric power tool in static and dynamic states at the same time.
As shown in fig. 2, in some embodiments, the attitude angle parameters are fused according to a dual-channel adaptive weighted fusion filter, which specifically includes:
s1010, acquiring a first group of pitch angle, roll angle and course angle according to a gyroscope on an electric tool;
s1011, acquiring a second group of pitch angle, roll angle and course angle according to a magnetometer, an accelerometer and a GPS module on the electric tool;
s1012, inputting two groups of pitch angle, roll angle and course angle parameters into a double-channel self-adaptive weighting fusion filter for fusion;
s1013, outputting a fusion attitude angle by a double-channel self-adaptive weighted fusion filter: pitch angle theta 0 Roll angle gamma 0 Heading angle psi 0 。
The algorithm used by the two-channel adaptive weighted fusion filter is an algorithm for performing adaptive weighted optimal estimation on the system state by using a linear system state equation and through two groups of system input and output observation data.
In some embodiments, before performing kalman filtering on two groups of multiple attitude angle measurement values, redundant processing is needed to be performed on the two groups of data, so that target measurement data with smaller loss is obtained, large-jump point data can be removed, expandability is good, redundant information is eliminated, communication complexity is reduced on a large scale, meanwhile, the defect that accuracy of measuring attitude angles in a single mode is not high enough is avoided, and robustness is good.
In some embodiments, the dual-channel adaptive weighted fusion filter can adaptively calculate a set of weight coefficients in real time according to the attitude angle estimation errors of different channels, and then the two channel corresponding weights are multiplied by the estimation results of the corresponding channels and added, so that the specific gravity of the estimation result of the channel with small error in the final estimation result can be improved, the specific gravity of the estimation result of the channel with large error in the final estimation result is reduced, and the accuracy of the attitude angle estimation result of the electric power tool is improved.
Specifically, the two-channel adaptive weighted fusion filter comprises two parts: and establishing a double-channel third-order discrete random system and estimating attitude angle parameters by utilizing an optimal fusion algorithm.
The method for establishing the double-channel third-order discrete random system comprises the following steps:
x 0 (k+1)＝Φx 0 (k)+w 0 (k)
y 1 (k)＝H 1 x 0 (k)+v 1 (k)
y 2 (k)＝H 2 x 0 (k)+v 2 (k)
wherein x is 0 (k)＝[θ 0,k γ 0,k ψ 0,k ] T For the two sets of the fused attitude angle parameter estimation results,for state transition matrix>For the observation matrices of group 1 and group 2,>is a systematic noise vector>For the first group of measuring noise vectors, +.>Measuring noise vector, y for group 2 1 Three attitude angle parameters, y, estimated for the first set before fusion 2 Three attitude angle parameters, X, estimated for the second set before fusion 0 (K+1) is the estimation result of the two groups of fused attitude angle parameters at the moment K+1;
as shown in fig. 3, the estimating of the attitude angle parameters by using the optimal fusion algorithm specifically includes:
s110, determining predicted values of the two groups of attitude angles at the moment K-1 according to the measured values of the two groups of attitude angles at the moment K-1;
s111, obtaining prediction errors of two groups of attitude angles at the moment K-1, and determining the filtering gain of the attitude angles according to the prediction errors;
s112, obtaining predicted values of the two groups of attitude angles at the moment K according to the predicted values of the two groups of attitude angles at the moment K-1 and the filter gain;
s113, determining the prediction errors of the two groups of attitude angles at the moment K and a covariance matrix of the prediction errors according to the prediction errors of the two groups of attitude angles at the moment K-1;
s114, determining weight indexes of predicted values of the two groups of attitude angles at the moment K according to the predicted errors of the two groups of attitude angles at the moment K-1 and a covariance matrix of the predicted errors;
s115, determining final estimated parameters of the attitude angles at the moment K according to the predicted values and the weight indexes of the two groups of attitude angles at the moment K.
In one embodiment, the algorithm flow of S110 to S115 described above is shown using a specific formula.
Calculating attitude angle predicted values of the 1 st group and the 2 nd group at the moment of K-1:
wherein, the liquid crystal display device comprises a liquid crystal display device,predicted attitude angle values for group 1, +.>Predicted attitude angle values for group 2, Φ is transfer matrix, < >>Predicted attitude angle for group 1 at time K-1 and +.>The predicted value of the attitude angle of the 2 nd group at the moment K-1;
calculating the prediction errors of the attitude angles of the 1 st group and the 2 nd group at the moment K-1:
P 1 (k|k-1)＝Φ(k)P 1 (k-1|k-1)Φ T +w 0 (k-1)w 0 (k-1) T
P 2 (k|k-1)＝Φ(k)P 2 (k-1|k-1)Φ T +w 0 (k-1)w 0 (k-1) T
wherein P is 1 (k|k-1) is the attitude angle prediction error, P of group 1 2 (k|k-1) is the 2 nd set of attitude angle prediction error, Φ is the transition matrix, w 0 (k-1) is a 3 x 1 system noise vector;
calculating the filter gains of the 1 st group and the 2 nd group:
K 1 (k)＝P 1 (k|k-1)H 1 T [H 1 P 1 (k|k-1)H 1 T +v 1 (k-1)v 1 (k-1) T ]
K 2 (k)＝P 2 (k|k-1)H 2 T [H 2 P 2 (k|k-1)H 2 T +v 2 (k-1)v 2 (k-1) T ]
wherein K is 1 (k) Filter gain, K, for group 1 attitude angles 2 (k) Filter gain, P for group 2 attitude angles 1 (k|k-1) is the 1 st group attitude angle prediction error, P 2 (k|k-1) is the 2 nd group attitude angle prediction error, H 1 For group 1 observation matrix, H 2 For group 2 observation matrix, v 1 (k-1) is the corresponding 3×1 measurement noise vector of group 1, v 2 (k-1) is a 3×1 measurement noise vector corresponding to group 2;
calculating predicted values of attitude angles of the 1 st group and the 2 nd group at the moment K (current):
wherein, the liquid crystal display device comprises a liquid crystal display device,predicted values of attitude angles of group 1 at K moment, < >>A predicted value of the attitude angle of the 2 nd group at the K moment,Predicted values of attitude angle of group 1 at time K-1,>predicted values of attitude angles of the 2 nd group at the moment K-1 are transition matrixes, H 1 For group 1 observation matrix, H 2 For group 2 observation matrix, K 1 (k) Filter gain, K, for group 1 attitude angles 2 (k) Filter gain, y of group 2 attitude angles 1 (k) K moment 1 st group attitude angle observation value, y 2 (k) The observation value of the attitude angle of the 2 nd group at the moment K;
calculating the attitude angle prediction errors of the 1 st group and the 2 nd group at the moment K and covariance of the two groups of attitude angle prediction errors:
P 1 (k|k)＝[I-K 1 (k)H 1 ]P 1 (k|k-1)
P 2 (k|k)＝[I-K 2 (k)H 2 ]P 2 (k|k-1)
P 12 (k|k)＝[I-K 1 (k)H 1 ]×[Φ(k)P 12 (k-1|k-1)Φ T +w 0 (k-1)w 0 (k-1) T ]×[I-K 1 (k)H 1 ]
wherein P is 12 (k|k) is the covariance matrix of the two-channel prediction error, P 1 (k|k) is the attitude angle prediction error, P of the 1 st group at the K moment 2 (k|k) is the attitude angle prediction error, P of the 2 nd group at the K moment 1 Attitude angle prediction error, P, of group 1 at time (k|k-1) being K-1 2 (k|k-1) is the attitude angle prediction error of the 2 nd group at time K-1, H 1 For group 1 observation matrix, H 2 Is the 2 nd group observation matrix, I isThe identity matrix phi is a transfer matrix, w 0 (k-1) is a 3 x 1 system noise vector;
calculating weight indexes of attitude angle predicted values of the 1 st group and the 2 nd group at the K moment:
wherein beta is 1 Weight, beta, when fused for group 1 attitude angle predictions 2 Weight, tr [. Cndot.]Is a trace of the matrix;
calculating final estimated parameters of the attitude angle of the electric power tool:
wherein, the liquid crystal display device comprises a liquid crystal display device,predicted values of attitude angles of group 1 at K moment, < >>Predicted value of attitude angle of group 2 at K moment beta 1 The weight value and beta corresponding to the fusion of the predicted values of the attitude angles of the 1 st group 2 The weight value corresponding to the fusion of the predicted values of the attitude angles of the group 2.
Therefore, the two sets of attitude angle data are fused by adopting the two-channel self-adaptive weighted fusion filtering algorithm, the motion attitude angle in the use process of the power tool can be quickly and accurately obtained, the fused attitude angle data can be better adapted to dynamic and static detection, the effects of low measurement deviation, good uniformity and small fluctuation degree can be achieved, the performance is superior to any single-channel measurement result, and the practical effect is very good.
In addition, as the measurement background and noise interference in practical application can change along with the change of the operation site of the electric power tool, the accuracy of the acquired attitude angle parameters can greatly fluctuate.
In some embodiments, the determining of the noise variance value of the dual-channel adaptive weighted fusion filter from the singular values of the square interpolation resampling matrix comprises: the measured noise variance value of the filter is adaptively and dynamically updated in the whole estimation process, and is obtained by calculating singular values of a square interpolation resampling matrix constructed by an input sequence, wherein the calculation process is as follows:
A. for any parameter to be estimated, a (2L-1) × (2L-1) square interpolation resampling matrix is constructed with an estimated parameter input sequence [ y (1), y (2),..y (N) ] comprising N measurement points as follows:
wherein N is odd and L is even, and l= (n+1)/2 is satisfied;
B. singular value decomposition is carried out on the square interpolation resampling matrix Y to obtain N singular values So= [ So (1), so (2) ], so (N), and then root mean square is respectively calculated on LN singular values according to a group of two adjacent singular values to obtain LN/2 synthesized singular values Sr= [ Sr (1), sr (2), sr (N/2) ];
finally, the noise level estimation of the used dual-channel self-adaptive weighting fusion filter is obtained
Process noise level estimate is +.>Wherein mean (s r ) To give a singular valueResults after the line average smoothing processing, media (s r ) The result of median smoothing of the singular values is obtained.
S102, determining a logarithmic energy entropy according to the gesture parameters;
in some embodiments, determining the logarithmic energy entropy from the pose parameters specifically includes: the logarithmic energy entropy of the pose parameters is obtained using the following formula:
L oe ＝log(θ 0 ) 2 +log(γ 0 ) 2 +log(ψ 0 ) 2 +log(L on ) 2 +log(L at ) 2 +log(A lt ) 2 ；
wherein L is oe Is logarithmic energy entropy and theta 0 Is pitch angle, gamma 0 Is a roll angle phi 0 Is course angle L on Is longitude, L at Is the latitude sum A lt Is high.
In fact, the traditional classifier only considers the gesture parameters of the tools and the tools, and does not consider the energy entropy information of the gesture parameters, so that the recognition accuracy of the behavior elements of the power operation of the tools and the tools is low, false alarm occurs frequently, and the monitoring effect of the power operation site is affected.
The application defines the logarithmic energy entropy index of the gesture parameters, combines other gesture parameters to form a novel weighted feature vector, and is beneficial to improving the success probability of the random forest classification algorithm on the gesture recognition of tools and instruments.
S103, acquiring a weighted feature vector according to the attitude parameters and the logarithmic energy entropy of the electric power tool;
in some embodiments, the method for obtaining the weighted feature vector according to the gesture parameter and the logarithmic energy entropy of the electric power tool specifically comprises the following steps:
wherein F is V For weighting feature vectors, theta 0 Is pitch angle, gamma 0 Is a roll angle phi 0 Is course angle L on Is a longitude,L at Is the latitude sum A lt Is high.
S104, constructing a feature data set according to the weighted feature vector;
in some examples, screening conditions may be set to screen out a large number of weighted feature vectors, where data with a certain age meaning and a higher quality is used as the weighted feature vector, which is also beneficial to ensure the credibility and accuracy of the data contained in the feature data set.
S105, training a random forest classification model according to the characteristic data set;
as shown in fig. 3, in some embodiments, training the random forest classification model based on the feature data set specifically includes:
s501, taking the weighted feature vector as training data and inputting the training data into a random forest classification model;
s502, recognizing training data by using a random forest classification model;
s503, classifying the training data until the accuracy of the random forest classification model is more than or equal to 0.95;
s106, determining the operation behavior of the current electric power tool according to the random forest classification model.
In particular, the above-described operational activities of the power tool include power tool lifting, power tool lowering, and power tool lowering.
The random forest classification model based on the logarithmic energy entropy and the feature vector provided by the application has high-precision classification and identification capability on lifting, descending and putting down elements of the current power operation behavior of the tool.
As shown in fig. 4, in some embodiments, determining the operation behavior of the current power tool according to the random forest classification model specifically includes:
s601, taking the weighted feature vector as test data and inputting the test data into a random forest classification model;
s602, identifying test data by using a random forest classification model, and acquiring electric power operation behavior elements represented by the test data;
s603, classifying the test data according to a random forest classification model to obtain a classification result;
s604, outputting the test data type label, namely the operation behavior of the electric power tool according to the classification result.
In summary, the random forest classification model can acquire the current behavior parameter information (pitch angle, roll angle, course angle, longitude, latitude and altitude) of the electric power tool according to the input weighted feature vector information, so that the operation behavior of the current electric power tool can be determined according to the information, the judgment of the operation behavior of the electric power tool can be realized, and the method is beneficial to timely finding and eliminating the potential safety hazard of an electric power working scene.
The application also provides a system for estimating the operation behavior of the electric power tool, which comprises: the system comprises a gesture parameter acquisition unit, a logarithmic energy entropy acquisition unit, a weighted feature vector acquisition unit, a data set construction unit, a model construction unit and a behavior estimation unit;
the gesture parameter acquisition unit is used for acquiring six gesture parameters of the electric power tool;
the logarithmic energy entropy acquisition unit is used for acquiring logarithmic energy entropy;
a weighted feature vector acquisition unit configured to acquire a weighted feature vector;
the data set construction unit is used for acquiring a characteristic data set;
the model building unit is used for building a random forest classification model;
and the behavior estimation unit is used for determining the operation behavior of the current power tool.
The application also proposes a readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the steps of:
acquiring attitude parameters of the electric power tool;
determining a logarithmic energy entropy according to the attitude parameters;
acquiring a weighted feature vector according to the attitude parameters and the logarithmic energy entropy of the electric power tool;
constructing a feature data set according to the weighted feature vector;
training a random forest classification model according to the characteristic data set;
and determining the operation behavior of the current power tool according to the random forest classification model.
The application also proposes a computer device comprising a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to perform the steps of: acquiring attitude parameters of the electric power tool;
determining a logarithmic energy entropy according to the attitude parameters;
acquiring a weighted feature vector according to the attitude parameters and the logarithmic energy entropy of the electric power tool;
constructing a feature data set according to the weighted feature vector;
training a random forest classification model according to the characteristic data set;
and determining the operation behavior of the current power tool according to the random forest classification model.
According to the application, a random forest classification algorithm based on attitude angle parameters, logarithmic energy entropy and weight feature vectors is constructed, so that the accuracy of classification of the tool and instrument attitude is improved, and the comprehensive and accurate monitoring of the electric power operation site is facilitated; the method can acquire the operation behaviors of the electric power tools, realize the judgment of dangerous operation behaviors in the electric power working scene, and is favorable for timely finding and eliminating the potential safety hazards of the electric power working scene.
Those skilled in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments can be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, can comprise the steps of the above-described embodiments of the methods.
Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory.
By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above embodiments can be arbitrarily combined, and for brevity, all of the possible combinations of the technical features of the above embodiments are not described, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope described in the present specification, the above embodiments only express several embodiments of the present application, the description of which is more specific and detailed, but not construed as limiting the scope of the patent claims.
It should be noted that it is possible for those skilled in the art to make several variations and modifications without departing from the spirit of the application, and these are all the preferred embodiments of the application, and it is needless to say that the scope of the claims of the application shall not be limited thereto, and therefore equivalent variations according to the claims of the application shall still fall within the scope of the application.
Claims (16)
1. A method of estimating operation behavior of an electrical power tool, the method comprising the steps of:
acquiring attitude parameters of the electric power tool;
determining a logarithmic energy entropy according to the attitude parameters;
acquiring a weighted feature vector according to the attitude parameters and the logarithmic energy entropy of the electric power tool;
constructing a feature data set according to the weighted feature vector;
training a random forest classification model according to the characteristic data set;
and determining the operation behavior of the current electric power tool according to the random forest classification model.
2. The method for estimating a working behavior of an electric power tool according to claim 1, wherein the attitude parameters include a position parameter and an attitude angle parameter, wherein the position parameter is acquired by a GPS module, and the attitude angle parameter is fused according to a two-channel adaptive weighted fusion filter.
3. The method of estimating a working behavior of an electric power tool according to claim 2, wherein the attitude angle parameter includes a pitch angle θ 0 Roll angle gamma 0 Heading angle psi 0 The method comprises the steps of carrying out a first treatment on the surface of the The location parameter includes longitude L on Latitude L at And height A lt 。
4. The method for estimating operation behavior of an electric power tool according to claim 2, wherein the attitude angle parameters are fused according to a two-channel adaptive weighted fusion filter, specifically comprising:
acquiring a first set of attitude angles according to a gyroscope on the electrical tool: pitch angle, roll angle, heading angle;
acquiring a second set of attitude angles according to magnetometers, accelerometers and GPS modules on the power tool: pitch angle, roll angle, heading angle;
inputting two groups of pitch angle, roll angle and course angle parameters into a double-channel self-adaptive weighting fusion filter for fusion;
the dual-channel self-adaptive weighted fusion filter outputs a fusion attitude angle: pitch angle theta 0 Roll angle gamma 0 Heading angle psi 0 。
5. The method according to any one of claims 2 to 4, wherein the noise variance of the two-channel adaptive weighted fusion filter is determined from singular values of a square interpolation resampling matrix.
6. The method of claim 5, wherein the two-channel adaptive weighted fusion filter comprises two parts: and establishing a double-channel third-order discrete random system and estimating attitude angle parameters by utilizing an optimal fusion algorithm.
7. The method for estimating a working behavior of an electric power tool according to claim 6, wherein the establishing a two-channel third-order discrete random system specifically comprises:
x 0 (k+1)＝Φx 0 (k)+w 0 (k)
y 1 (k)＝H 1 x 0 (k)+v 1 (k)
y 2 (k)＝H 2 x 0 (k)+v 2 (k)
wherein x is 0 (k) Is the estimated result of the two groups of fused attitude angle parameters, x 0 (k+1) is the estimation result of two groups of fused attitude angle parameters at time k+1, phi is a state transition matrix and H 1 And H 2 For observing matrix, w 0 (k) V is the system noise vector 1 (k) And v 2 (k) Respectively two groups of measured noise vectors, y 1 And y 2 And respectively fusing the estimated attitude angle parameters of the first two groups.
8. The method for estimating operation behavior of an electric power tool according to claim 6, wherein the estimating of attitude angle parameters by using an optimal fusion algorithm specifically comprises:
determining predicted values of the two groups of attitude angles at the moment K-1 according to the measured values of the two groups of attitude angles at the moment K-1;
acquiring prediction errors of two groups of attitude angles at the moment K-1, and determining the filtering gain of the attitude angles according to the prediction errors;
acquiring predicted values of the two groups of attitude angles at the moment K according to the predicted values of the two groups of attitude angles at the moment K-1 and the filtering gain;
determining the prediction errors of the two groups of attitude angles at the K moment and a covariance matrix of the prediction errors according to the prediction errors of the two groups of attitude angles at the K-1 moment;
determining weight indexes of the predicted values of the two groups of attitude angles at the moment K according to the predicted errors of the two groups of attitude angles at the moment K-1 and a covariance matrix of the predicted errors;
and determining final estimated parameters of the attitude angles at the moment K according to the predicted values and the weight indexes of the two groups of attitude angles at the moment K.
9. The method for estimating the operation behavior of the electric power tool according to claim 1, wherein the determining the logarithmic energy entropy according to the attitude parameter specifically includes: the logarithmic energy entropy of the pose parameters is obtained using the following formula:
L oe ＝log(θ 0 ) 2 +log(γ 0 ) 2 +log(ψ 0 ) 2 +log(L on ) 2 +log(L at ) 2 +log(A lt ) 2 ；
wherein L is oe Is the logarithmic energy entropy.
10. The method for estimating operation behavior of an electric power tool according to claim 1, wherein the step of obtaining a weighted feature vector according to the attitude parameter and the logarithmic energy entropy of the electric power tool specifically comprises:
wherein F is V Is a weighted feature vector.
11. The method for estimating the operation behavior of an electric power tool according to claim 1, wherein training a random forest classification model according to the feature data set specifically includes:
inputting the weighted feature vector as training data into a random forest classification model;
identifying the training data using the random forest classification model;
and classifying the training data until the accuracy of the random forest classification model is more than or equal to 0.95.
12. The method of claim 1, wherein determining the current power tool's operational behavior based on the random forest classification model further comprises: the operation behavior of the electric power tool is as follows: lifting, raising, lowering and lowering.
13. The method for estimating the operation behavior of the electric power tool according to claim 12, wherein the determining the operation behavior of the current electric power tool according to the random forest classification model specifically comprises:
inputting the weighted feature vector as test data into a random forest classification model;
identifying the test data by utilizing the random forest classification model, and acquiring electric power operation behavior elements represented by the test data;
classifying the test data according to the random forest classification model to obtain a classification result;
and outputting the test data type label, namely the operation behavior of the electric power tool according to the classification result.
14. A system for estimating the operation behavior of an electrical power tool, the system comprising: the system comprises a gesture parameter acquisition unit, a logarithmic energy entropy acquisition unit, a weighted feature vector acquisition unit, a data set construction unit, a model construction unit and a behavior estimation unit;
the gesture parameter acquisition unit is used for acquiring gesture parameters of the electric power tool;
the logarithmic energy entropy acquisition unit is used for acquiring logarithmic energy entropy;
the weighted feature vector acquisition unit is used for acquiring weighted feature vectors;
the data set construction unit is used for acquiring a characteristic data set;
the model construction unit is used for constructing a random forest classification model;
the behavior estimation unit is used for determining the operation behavior of the current power tool.
15. A readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the steps of the method of any one of claims 1 to 13.
16. A computer device comprising a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the method as claimed in any one of claims 1 to 13.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310609543.2A CN116702064A (en) | 2023-05-26 | 2023-05-26 | Method, system, storage medium and equipment for estimating operation behavior of electric power tool |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310609543.2A CN116702064A (en) | 2023-05-26 | 2023-05-26 | Method, system, storage medium and equipment for estimating operation behavior of electric power tool |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116702064A true CN116702064A (en) | 2023-09-05 |
Family
ID=87828539
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310609543.2A Pending CN116702064A (en) | 2023-05-26 | 2023-05-26 | Method, system, storage medium and equipment for estimating operation behavior of electric power tool |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116702064A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117034201A (en) * | 2023-10-08 | 2023-11-10 | 东营航空产业技术研究院 | Multi-source real-time data fusion method |
-
2023
- 2023-05-26 CN CN202310609543.2A patent/CN116702064A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117034201A (en) * | 2023-10-08 | 2023-11-10 | 东营航空产业技术研究院 | Multi-source real-time data fusion method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10310020B2 (en) | Method, device and system for estimating the charge state of a battery | |
| CN110334680B (en) | Ship water gauge identification method, system and device based on wall-climbing robot | |
| CN116702064A (en) | Method, system, storage medium and equipment for estimating operation behavior of electric power tool | |
| CN110780164B (en) | Insulator infrared fault positioning diagnosis method and device based on YOLO | |
| CN105974273A (en) | Power distribution network fault positioning system | |
| JP6776319B2 (en) | Detecting temporary errors in body fluid samples | |
| CN114707632A (en) | Sensor network sensor fault positioning method, system, equipment and medium | |
| CN112816881A (en) | Battery differential pressure abnormality detection method, battery differential pressure abnormality detection device and computer storage medium | |
| JP2020042669A (en) | Inspection apparatus and machine learning method | |
| CN112379268A (en) | Lithium battery SOC estimation method and device based on SVM _ EKF algorithm and storage medium | |
| CN114463932A (en) | Non-contact construction safety distance active dynamic recognition early warning system and method | |
| JP2020129233A (en) | Abnormality detection device | |
| CN111444833B (en) | Fruit actual measurement production method, device, computer equipment and storage medium | |
| CN112101313A (en) | Machine room robot inspection method and system | |
| JP2020042668A (en) | Inspection apparatus and machine learning method | |
| CN106767773B (en) | Indoor geomagnetic reference map construction method and device | |
| CN112419402A (en) | Positioning method and system based on multispectral image and laser point cloud | |
| CN115266583B (en) | Ambient light filtering method, system, computer equipment and computer readable storage medium | |
| CN109657907B (en) | Quality control method and device for geographical national condition monitoring data and terminal equipment | |
| CN114279311A (en) | GNSS deformation monitoring method and system based on inertia | |
| CN113419502A (en) | Process monitoring method and system for shell manufacturing production process based on artificial intelligence | |
| CN114004138A (en) | Building monitoring method and system based on big data artificial intelligence and storage medium | |
| CN112966762A (en) | Wild animal detection method and device, storage medium and electronic equipment | |
| CN111401123A (en) | S L AM loop detection method and system based on deep learning | |
| KR101847113B1 (en) | Estimation method and apparatus for information corresponding camera orientation by using image |
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 |