CN116894213A - Method for identifying working condition of excavator and working condition identifying device of excavator - Google Patents
Method for identifying working condition of excavator and working condition identifying device of excavator Download PDFInfo
- Publication number
- CN116894213A CN116894213A CN202310716264.6A CN202310716264A CN116894213A CN 116894213 A CN116894213 A CN 116894213A CN 202310716264 A CN202310716264 A CN 202310716264A CN 116894213 A CN116894213 A CN 116894213A
- Authority
- CN
- China
- Prior art keywords
- data
- excavator
- feature vector
- working condition
- signal
- 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 54
- 239000013598 vector Substances 0.000 claims abstract description 101
- 230000009467 reduction Effects 0.000 claims abstract description 37
- 238000013145 classification model Methods 0.000 claims abstract description 32
- 238000007781 pre-processing Methods 0.000 claims abstract description 9
- 238000012545 processing Methods 0.000 claims description 21
- 238000010276 construction Methods 0.000 claims description 17
- 238000007599 discharging Methods 0.000 claims description 12
- 238000003860 storage Methods 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 11
- 230000002068 genetic effect Effects 0.000 claims description 10
- 238000007637 random forest analysis Methods 0.000 claims description 10
- 238000004064 recycling Methods 0.000 claims description 8
- 230000009977 dual effect Effects 0.000 claims description 2
- 239000002131 composite material Substances 0.000 abstract description 4
- 238000005065 mining Methods 0.000 abstract description 4
- 238000004422 calculation algorithm Methods 0.000 description 17
- 238000010586 diagram Methods 0.000 description 12
- 238000004590 computer program Methods 0.000 description 7
- 239000002689 soil Substances 0.000 description 6
- 238000001914 filtration Methods 0.000 description 5
- 230000035772 mutation Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 4
- 238000012706 support-vector machine Methods 0.000 description 4
- 238000009412 basement excavation Methods 0.000 description 3
- 238000002790 cross-validation Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000001174 ascending effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008094 contradictory effect Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000009440 infrastructure construction Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
- 238000002922 simulated annealing Methods 0.000 description 1
- 230000003068 static 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
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
- G06F18/15—Statistical pre-processing, e.g. techniques for normalisation or restoring missing data
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Genetics & Genomics (AREA)
- Physiology (AREA)
- Probability & Statistics with Applications (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Operation Control Of Excavators (AREA)
Abstract
The application discloses a method for identifying working conditions of an excavator and an excavator working condition identification device. The method comprises the following steps: acquiring pilot control signal data and double-pump pressure signal data of an operating handle of the excavator; preprocessing the pilot control signal data of the operating handle and the double pump pressure signal data to obtain preprocessed data; constructing a feature vector according to the preprocessed data; performing feature dimension reduction on the feature vector to obtain a dimension-reduced feature vector; inputting the feature vector subjected to dimension reduction into a classification model to obtain a first working condition mode of the excavator; and checking the first working condition mode and outputting a working condition identification result detail table. According to the application, the feature vector is constructed by combining the basic time domain features of the composite signal and the actual conditions of mining, the feature vector is subjected to dimension reduction, the feature factors with smaller influence on the recognition result are removed, and after the model recognition is finished, the classification result is checked to correct the model recognition result, so that the classification precision of the model is improved.
Description
Technical Field
The application relates to the technical field of excavator mode identification, in particular to a method for identifying excavator work and an excavator work identification device.
Background
The excavator is an important construction machine essential for infrastructure construction, and is widely used in mines, buildings, hydraulic engineering, and the like. At present, in the use of the excavator, the problems of low working efficiency and unreasonable task planning exist, the service life of the excavator is seriously influenced, the working mode of each excavator needs to be identified for better embodying the multifunctional and efficient operation of the excavator, the oil consumption of the excavator is monitored, the aims of energy saving control and energy saving and emission reduction are achieved, meanwhile, the construction task is reasonably planned, and the working capacity of the excavator is fully exerted.
At present, two main methods are available for identifying the working condition of the hydraulic excavator, the first method is mainly image identification, the method needs to install image or video acquisition equipment on a construction site, and image processing algorithms such as target detection and the like are used for identifying and metering the working condition of the excavator, so that the method has high cost. The second mode is to construct a classification model for identifying the main pump pressure and the pilot signal, but at present, the statistical characteristic values of the basic extraction samples are modeled, such as the mean value, the variance and other widely used characteristics, but the modes of excavation, land leveling and the like cannot be well reflected, and the model precision is low. Meanwhile, when a machine learning algorithm is used, few methods are used for explaining how algorithm parameters are selected, and the problems of low model learning capacity or over fitting are easily caused.
Disclosure of Invention
The embodiment of the application aims to provide a method for identifying working conditions of an excavator and an excavator working condition identification device, which are used for solving the problem of low accuracy of identifying working condition modes of the excavator in the prior art.
In order to achieve the above object, a first aspect of the present application provides a method for identifying working conditions of an excavator, which is applied to an excavator working condition identifying device, the method comprising:
acquiring pilot control signal data and double-pump pressure signal data of an operating handle of the excavator;
preprocessing the pilot control signal data of the operating handle and the double pump pressure signal data to obtain preprocessed data;
constructing a feature vector according to the preprocessed data;
performing feature dimension reduction on the feature vector to obtain a dimension-reduced feature vector;
inputting the feature vector subjected to dimension reduction into a classification model to obtain a first working condition mode of the excavator;
and checking the first working condition mode and outputting a working condition identification result detail table.
In an embodiment of the present application, the method further includes:
acquiring crushing working mode state, machine construction state, machine walking state and engine rotating speed data of an excavator;
and determining a second working condition mode of the excavator according to the crushing working mode state, the machine construction state, the machine walking state and the engine rotating speed data, and outputting a working condition identification result detail table.
In the embodiment of the application, the classification model is a libsvm model, and the method further comprises:
target parameters of the libsvm model are determined to construct a classification model.
In an embodiment of the present application, determining target parameters of the libsvm model to construct the classification model includes:
encoding initial parameters of the libsvm model to form an initial population, wherein the initial population comprises a plurality of individuals;
respectively determining the fitness of each individual in the initial population;
judging whether an individual with the fitness meeting the termination condition exists or not;
if it is determined that an individual whose fitness satisfies the termination condition exists, determining a parameter corresponding to the individual whose fitness satisfies the termination condition as a target parameter of the libsvm model;
in the case that no individual meeting the termination condition is judged, the initial population is updated through selection, crossing and mutation by genetic operators until the individual meeting the termination condition is judged.
In an embodiment of the present application, preprocessing the pilot control signal data of the operating handle and the double pump pressure signal data to obtain preprocessed data includes:
performing data filling processing on the pilot control signal data of the operating handle and the double pump pressure signal data to obtain filled data;
carrying out data alignment processing on the filled data to obtain aligned data;
and denoising the aligned data to obtain preprocessed data.
In an embodiment of the present application, the feature vector includes a first feature vector and a second feature vector, and constructing the feature vector according to the preprocessed data includes:
acquiring a pressure wave band of the excavator;
intercepting a pressure wave band of a preset size window;
extracting time domain features of a pressure wave band of a preset size window to obtain a first feature vector;
and extracting the pressure ratio of the digging signal, the unloading signal and the rotation signal of the pressure wave band of the preset size window to obtain a second characteristic vector.
In the embodiment of the present application, performing feature dimension reduction on the feature vector to obtain a feature vector after dimension reduction includes:
inputting the feature vectors and the working condition mode labels into a random forest function to obtain the contribution rate of each feature vector;
sorting the feature vectors according to the contribution rate and a preset sequence;
and determining the feature vectors of the preset quantity as feature vectors after the dimension reduction.
In an embodiment of the present application, verifying the first working condition mode includes:
acquiring the wave crest and wave trough type vibration frequency of an excavating pilot signal, a discharging pilot signal, a rotary pilot signal, an arm stretching pilot signal and an arm recycling pilot signal of the excavator;
and verifying the working condition recognition result according to the wave crest and wave trough type vibration frequency of the excavating pilot signal, the discharging pilot signal, the rotary pilot signal, the bucket rod stretching pilot signal and the bucket rod recycling pilot signal.
A second aspect of the present application provides an excavator work identification device comprising:
a memory configured to store instructions; and
a processor configured to invoke the instructions from the memory and when executing the instructions is capable of implementing the method for excavator work identification according to the above.
A third aspect of the application provides a machine-readable storage medium having stored thereon instructions for causing a machine to perform a method for excavator work identification according to the above.
According to the technical scheme, firstly, operation handle pilot control signal data and double pump pressure signal data of the excavator are obtained, the operation handle pilot control signal data and the double pump pressure signal data are preprocessed to obtain preprocessed data, then feature vectors are built according to the preprocessed data, feature dimension reduction is conducted on the feature vectors to obtain feature vectors after dimension reduction, finally the feature vectors after dimension reduction are input into a classification model to obtain a first working condition mode of the excavator, the first working condition mode is verified, and a working condition identification result detail table is output. According to the application, the feature vector is constructed by combining the basic time domain features of the composite signal and the actual conditions of mining, the feature vector is subjected to dimension reduction, the feature factors with smaller influence on the recognition result are removed, and after the model recognition is finished, the classification result is checked to correct the model recognition result, so that the classification precision of the model is improved.
Additional features and advantages of embodiments of the application will be set forth in the detailed description which follows.
Drawings
The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain, without limitation, the embodiments of the application. In the drawings:
FIG. 1 schematically illustrates a flow chart of a method for excavator work identification in accordance with an embodiment of the present application;
FIG. 2 (a) schematically illustrates a bulldozer blade operating mode and corresponding pressure band diagram according to an embodiment of the present application;
FIG. 2 (b) schematically illustrates a colluded, flat ground operating condition and corresponding pressure band diagram according to an embodiment of the present application;
fig. 3 schematically shows a block diagram of a construction of an excavator work recognition device according to an embodiment of the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it should be understood that the detailed description described herein is merely for illustrating and explaining the embodiments of the present application, and is not intended to limit the embodiments of the present application. 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.
It should be noted that, if directional indications (such as up, down, left, right, front, and rear … …) are included in the embodiments of the present application, the directional indications are merely used to explain the relative positional relationship, movement conditions, etc. between the components in a specific posture (as shown in the drawings), and if the specific posture is changed, the directional indications are correspondingly changed.
In addition, if there is a description of "first", "second", etc. in the embodiments of the present application, the description of "first", "second", etc. is for descriptive purposes only and is not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In addition, the technical solutions of the embodiments may be combined with each other, but it is necessary to base that the technical solutions can be realized by those skilled in the art, and when the technical solutions are contradictory or cannot be realized, the combination of the technical solutions should be considered to be absent and not within the scope of protection claimed in the present application.
Fig. 1 schematically shows a flow chart of a method for excavator work identification according to an embodiment of the present application. As shown in fig. 1, an embodiment of the present application provides a method for identifying working conditions of an excavator, which is applied to an excavator working condition identification device, and the method may include the following steps.
Step 101, acquiring pilot control signal data and double pump pressure signal data of an operating handle of the excavator;
102, preprocessing pilot control signal data of an operating handle and double pump pressure signal data to obtain preprocessed data;
step 103, constructing a feature vector according to the preprocessed data;
104, carrying out feature dimension reduction on the feature vector to obtain a feature vector after dimension reduction;
step 105, inputting the feature vector subjected to dimension reduction into a classification model to obtain a first working condition mode of the excavator;
and 106, checking the first working condition mode and outputting a working condition identification result detail table.
With the development of the electromechanical integration technology, the working parameters of the excavator can be acquired through a controller local area network bus, the requirements of the current excavator detection can be objectively reflected, and the parameters of the excavator become important bases for identifying the working stage of the excavator. In the embodiment of the application, the data such as the double-pump pressure, the pilot control signal of the operating handle and the like are processed, analyzed and modeled in a combined way, so that the method has higher recognition precision and better persuasion compared with a single signal.
Specifically, the excavator work recognition device may first acquire operation handle pilot control signal data of the excavator, including stick-out pilot pressure, stick-back pilot pressure, swing pilot pressure, boom-down pilot pressure, boom-up pilot pressure, bucket-digging pilot pressure, bucket-unloading pilot pressure, left-travel pilot pressure, and right-travel pilot pressure, and dual pump pressure signal data, including front pump pressure and rear pump pressure. After the pilot control signal data of the operating handle and the double pump pressure signal data are obtained, the excavator work identification device can preprocess all the data to obtain preprocessed data, and due to the fact that the positions of the sensors are different, various data can have the problem of time misalignment, filling processing can be carried out on the data firstly, namely forward filling and backward filling null values are carried out on the data, and then data alignment processing is carried out on the filled data to obtain aligned data. In addition, since the acquisition of the data signal is affected by vibration and interference of other factors, noise is often mixed in the signal, and thus, it is also necessary to perform filtering processing on the data signal. After preprocessing the data, the feature vector can be constructed according to the processed data, and the high-dimensional feature quantity is easy to interfere with the classification model, so that the accuracy of the classification model is affected, and therefore, the high-dimensional data is required to be subjected to dimension reduction processing. And finally, inputting the feature vector subjected to dimension reduction into a classification model to identify a first working condition mode of the excavator. The first working condition mode of the excavator refers to an excavating working condition, a land leveling working condition and other working condition modes except an idling working condition, a walking working condition and a crushing working condition. The embodiment of the application can also verify the first working condition mode after the working conditions are identified through the classification model so as to ensure the accuracy of identifying the working condition of the excavator, and then output a working condition identification result detail table.
According to the technical scheme, firstly, operation handle pilot control signal data and double pump pressure signal data of the excavator are obtained, the operation handle pilot control signal data and the double pump pressure signal data are preprocessed to obtain preprocessed data, then feature vectors are built according to the preprocessed data, feature dimension reduction is conducted on the feature vectors to obtain feature vectors after dimension reduction, finally the feature vectors after dimension reduction are input into a classification model to obtain a first working condition mode of the excavator, the first working condition mode is verified, and a working condition identification result detail table is output. According to the application, the feature vector is constructed by combining the basic time domain features of the composite signal and the actual conditions of mining, the feature vector is subjected to dimension reduction, the feature factors with smaller influence on the recognition result are removed, and after the model recognition is finished, the classification result is checked to correct the model recognition result, so that the classification precision of the model is improved.
In an embodiment of the present application, the method may further include:
acquiring crushing working mode state, machine construction state, machine walking state and engine rotating speed data of an excavator;
and determining a second working condition mode of the excavator according to the crushing working mode state, the machine construction state, the machine walking state and the engine rotating speed data, and outputting a working condition identification result detail table.
In an embodiment of the application, the second operating mode of the excavator includes an idle operating mode, a walking operating mode and a crushing operating mode. In practical application, when the excavator work recognition is carried out, the excavator work recognition device can also acquire crushing work mode state, machine construction state, machine walking state and engine rotating speed data of the excavator. Whether the excavator is in an idle working condition, a walking working condition and a crushing working condition can be judged according to the crushing working mode state, the machine construction state, the machine walking state and the engine rotating speed data of the excavator. Specifically, the excavator work operation recognition device determines that the excavator is in the idle operation state when the machine construction state is 0 or False, the machine traveling state is 0 or False, and the engine speed is in a preset interval, for example, [600, 1100 ]. When the machine walking state parameter value is not 0 or False, the excavator working condition identification device judges that the excavator is walking. When the state of the crushing working mode is 1 or True, the excavator working condition identification device judges that the excavator is in the crushing working condition.
In the embodiment of the present application, the classification model is a libsvm model, and the method may further include:
target parameters of the libsvm model are determined to construct a classification model.
In an embodiment of the present application, determining the target parameters of the libsvm model to construct the classification model may include:
encoding initial parameters of the libsvm model to form an initial population, wherein the initial population comprises a plurality of individuals;
respectively determining the fitness of each individual in the initial population;
judging whether an individual with the fitness meeting the termination condition exists or not;
if it is determined that an individual whose fitness satisfies the termination condition exists, determining a parameter corresponding to the individual whose fitness satisfies the termination condition as a target parameter of the libsvm model;
in the case that no individual meeting the termination condition is judged, the initial population is updated through selection, crossing and mutation by genetic operators until the individual meeting the termination condition is judged.
In the embodiment of the application, the support vector machine (support vector machine, SVM) has better robustness and generalization capability, so that the libsvm model is used as a classification model to identify the working condition of the excavator. When the classification model is constructed, penalty parameters c and kernel parameters g which have important influence on the learning ability of the model are difficult to determine, the penalty parameters c are selected to balance the complexity of the model and training errors, the value of the penalty parameters c is generally 1, the parameters g of the kernel function mainly reflect the range characteristics of training sample data and directly influence the learning ability of the support vector machine model, and 1/k is generally defaulted to be the total class number. The values of the two parameters and the learning ability of the support vector machine model have great influence, if the values of the punishment parameter c and the nuclear parameter g are too small, the learning ability of the model is poor, and if the values of the punishment parameter c and the nuclear parameter g are too large, the model can be over-fitted, so that in order to improve the learning ability of the model and the accuracy of identification, the genetic algorithm is adopted for parameter optimization in a range. And taking the cross verification precision as an objective function, solving the optimal parameter value of the punishment parameter c and the nuclear parameter g, and determining the objective parameter, namely the optimal parameter, of the libsvm model to construct the classification model.
Specifically, the genetic algorithm encodes parameters c and g, trains a libsvm model by using the parameter values obtained by decoding, and then obtains the optimal accuracy by adopting cross-validation. Cross validation (Cross validation) is a method commonly used in machine learning for selecting models and evaluating the quality of models, and the main idea is that: in a given training sample, a model is built by using most of the samples, the rest of the samples are taken for predicting the built model, the prediction error of the prediction result of the small part is obtained, and finally the model with small prediction error is selected as the optimal model. The genetic algorithm uses the accuracy of cross verification as a fitness value, and evolutionary iteration is carried out through selection, crossing and mutation of genetic operators, so that the optimal parameters of the libsvm model are obtained. In practical application, initial parameters of the libsvm model are firstly encoded to form an initial population, the initial population comprises a plurality of individuals, then the fitness of each individual in the initial population is respectively determined, and whether the individuals with the fitness meeting the termination condition exist or not is judged. When it is determined that there is an individual whose fitness satisfies the termination condition, a parameter corresponding to the individual whose fitness satisfies the termination condition is determined as a target parameter of the libsvm model. And under the condition that no individual with the fitness meeting the termination condition exists, the initial population is updated through selection, crossing and mutation by a genetic operator until the individual with the fitness meeting the termination condition exists, and the parameters corresponding to the individual with the fitness meeting the termination condition are determined as target parameters of the libsvm model. It should be noted that, the technical scheme of the present application may use not only a genetic algorithm, but also a simulated annealing algorithm or a tabu search algorithm.
In an embodiment of the present application, preprocessing the pilot control signal data of the operating handle and the double pump pressure signal data to obtain preprocessed data may include:
performing data filling processing on the pilot control signal data of the operating handle and the double pump pressure signal data to obtain filled data;
carrying out data alignment processing on the filled data to obtain aligned data;
and denoising the aligned data to obtain preprocessed data.
In the embodiment of the application, the operating handle pilot control signal data comprise a bucket arm extending pilot pressure, a bucket arm retracting pilot pressure, a turning pilot pressure, a boom descending pilot pressure, a boom ascending pilot pressure, a bucket excavating pilot pressure, a bucket unloading pilot pressure, a left walking pilot pressure and a right walking pilot pressure, the double-pump pressure signal data comprise a front pump pressure and a rear pump pressure, and various data can have the problem of time misalignment due to different positions of sensors, so that the data can be filled firstly, namely, the data can be filled forwards and backwards with empty values, and then the filled data can be subjected to data alignment processing to obtain aligned data. In addition, since the acquisition of the data signal is affected by vibration and interference of other factors, noise is often mixed in the signal, and thus, it is also necessary to perform filtering processing on the data signal. In the embodiment of the application, the data is smoothed by adopting an SG (Savitzky-Golay) algorithm. The Sav itzky-Golay algorithm is a filtering method based on local polynomial least square fitting in a time domain, and the core idea is to carry out weighted filtering on data in a window, wherein the weighted weight is obtained by carrying out least square fitting on a given higher-order polynomial. The SG algorithm is used for smoothing the data, so that smoothness of a data curve can be improved, noise interference can be reduced, and compared with other denoising algorithms, the SG algorithm can better keep characteristics such as shape and width of a signal when filtering and smoothing, so that change information of the signal can be more effectively kept, and further, follow-up working condition identification is more accurate.
FIG. 2 (a) schematically illustrates a bulldozer blade operating mode and corresponding pressure band diagram according to an embodiment of the present application; fig. 2 (b) schematically illustrates a hooked-up-soil land leveling regime and a corresponding pressure band diagram according to an embodiment of the present application. As shown in fig. 2 (a) and fig. 2 (b), in an embodiment of the present application, the feature vector includes a first feature vector and a second feature vector, and constructing the feature vector according to the preprocessed data may include:
acquiring a pressure wave band of the excavator;
intercepting a pressure wave band of a preset size window;
extracting time domain features of a pressure wave band of a preset size window to obtain a first feature vector;
and extracting the pressure ratio of the digging signal, the unloading signal and the rotation signal of the pressure wave band of the preset size window to obtain a second characteristic vector.
In the embodiment of the application, the characteristic vector comprises a first characteristic vector and a second characteristic vector, wherein the first characteristic vector is a time domain characteristic value of the pressure wave band of the excavator, and the second characteristic vector is the pressure duty ratio of the excavating signal, the unloading signal and the rotary signal in the pressure wave band of the excavator. Compared with the prior art, the embodiment of the application increases the pressure ratio of the excavation signal, the unloading signal and the rotation signal in the pressure wave band of the excavator to construct the feature vector, which is beneficial to improving the accuracy of model identification. Specifically, a fixed window method may be adopted to intercept a pressure band when the excavator works, and the window size is set to win at first, and in practical application, is generally set to 15s of one excavating working period, i.e. win=150. And then extracting time domain characteristic parameters of any window, wherein the time domain characteristic parameters comprise dimensionality-containing time domain characteristics and dimensionless time domain characteristics, and the time domain characteristic parameters respectively comprise maximum values, minimum values, average values, standard deviations, root mean square plants, skewness, kurtosis factors, waveform factors, margin factors and the like. In addition, the video data and the signal data of the land leveling can find that two conditions are mainly found when the excavator is in the land leveling working condition, namely a bulldozing land leveling condition and a soil hooking land leveling condition. As shown in fig. 2 (a), the bulldozer type level ground, i.e. no soil in the bucket, is to push out the soil with the anti-bucket, in which the excavating and dumping signals are small; as shown in fig. 2 (b), the other type is a hooked soil type level ground, namely, a bucket with soil, but since the level ground is stretched and retracted in one time, the swing pressure signal is weaker, and meanwhile, the stretching and retracting pressure of the bucket rod is obvious. Therefore, besides basic time domain characteristics, the embodiment of the application also increases the pressure duty ratio of the excavation signal, the unloading signal and the rotation signal in the excavator pressure wave band so as to construct a characteristic vector, and improves the accuracy of model identification.
In the embodiment of the present application, performing feature dimension reduction on the feature vector to obtain the feature vector after dimension reduction may include:
inputting the feature vectors and the working condition mode labels into a random forest function to obtain the contribution rate of each feature vector;
sorting the feature vectors according to the contribution rate and a preset sequence;
and determining the feature vectors of the preset quantity as feature vectors after the dimension reduction.
In the embodiment of the application, more than one hundred feature vectors are constructed according to the preprocessed data, and the high-dimensional feature quantity easily interferes with the classification model, so that the accuracy of the classification model is affected, and therefore, the high-dimensional feature quantity needs to be subjected to dimension reduction processing. Specifically, the feature vector can be subjected to dimension reduction processing by adopting a random forest algorithm, wherein the random forest algorithm is one of classical data mining algorithms, and the core of the random forest algorithm is the construction of decision trees, and the feature recognition and extraction of sample data are completed through the combination of a plurality of decision trees. According to the technical scheme, the feature vectors and the working condition mode labels are input into a random forest (random forest) function to obtain the contribution rate of each feature vector, and then a preset number of feature vectors with the contribution rate ranked at the front are input into a classification model to perform working condition recognition, wherein the preset number can be set according to actual conditions, and preferably can be 20. Therefore, the accuracy of classification model identification can be improved by reducing the dimension of the feature vector through a random forest method.
In an embodiment of the present application, verifying the first working condition mode may include:
acquiring the wave crest and wave trough type vibration frequency of an excavating pilot signal, a discharging pilot signal, a rotary pilot signal, an arm stretching pilot signal and an arm recycling pilot signal of the excavator;
and verifying the working condition recognition result according to the wave crest and wave trough type vibration frequency of the excavating pilot signal, the discharging pilot signal, the rotary pilot signal, the bucket rod stretching pilot signal and the bucket rod recycling pilot signal.
In the embodiment of the application, after the first working condition mode identification is obtained through the classification model, the identification result can be further checked and verified. And verifying the working condition recognition result through the vibration frequency of the wave crest and the wave trough of the excavating pilot signal, the discharging pilot signal, the rotary pilot signal, the bucket rod stretching pilot signal and the bucket rod recycling pilot signal. In one example, first, a maximum value peak_upper and a minimum value peak_low are set for the excavating pilot signal and the discharging pilot signal respectively, two points p1 and p2 are selected on the pressure wave bands of the excavating pilot signal and the discharging pilot signal respectively according to actual conditions, when the trough between the points p1 and p2 is recognized to be smaller than peak_low, p2 is taken as an effective vibration point, and since the excavating pilot signal and the bucket discharging pilot signal have obvious regular characteristics when the excavating work is carried out by the excavator, when more than a plurality of excavating vibration points are recognized in one period, namely the effective vibration points, and the arm stretching pilot pressure and the arm retracting pilot pressure signal are obvious, the excavator is indicated to be in a ground working condition at the moment.
Fig. 3 schematically shows a block diagram of a construction of an excavator work recognition device according to an embodiment of the present application. As shown in fig. 3, an embodiment of the present application provides an excavator working condition identifying device, which may include:
a memory 310 configured to store instructions; and
processor 320 is configured to invoke instructions from memory 310 and when executing the instructions, to implement the method for controlling a boom described above.
Specifically, in an embodiment of the present application, processor 320 may be configured to:
acquiring pilot control signal data and double-pump pressure signal data of an operating handle of the excavator;
preprocessing the pilot control signal data of the operating handle and the double pump pressure signal data to obtain preprocessed data;
constructing a feature vector according to the preprocessed data;
performing feature dimension reduction on the feature vector to obtain a dimension-reduced feature vector;
inputting the feature vector subjected to dimension reduction into a classification model to obtain a first working condition mode of the excavator;
and checking the first working condition mode and outputting a working condition identification result detail table.
Further, the processor 320 may be further configured to:
acquiring crushing working mode state, machine construction state, machine walking state and engine rotating speed data of an excavator;
and determining a second working condition mode of the excavator according to the crushing working mode state, the machine construction state, the machine walking state and the engine rotating speed data, and outputting a working condition identification result detail table.
In the embodiment of the application, the classification model is a libsvm model, and the method further comprises:
target parameters of the libsvm model are determined to construct a classification model.
Further, the processor 320 may be further configured to:
encoding initial parameters of the libsvm model to form an initial population, wherein the initial population comprises a plurality of individuals;
respectively determining the fitness of each individual in the initial population;
judging whether an individual with the fitness meeting the termination condition exists or not;
if it is determined that an individual whose fitness satisfies the termination condition exists, determining a parameter corresponding to the individual whose fitness satisfies the termination condition as a target parameter of the libsvm model;
in the case that no individual meeting the termination condition is judged, the initial population is updated through selection, crossing and mutation by genetic operators until the individual meeting the termination condition is judged.
Further, the processor 320 may be further configured to:
performing data filling processing on the pilot control signal data of the operating handle and the double pump pressure signal data to obtain filled data;
carrying out data alignment processing on the filled data to obtain aligned data;
and denoising the aligned data to obtain preprocessed data.
Further, the processor 320 may be further configured to:
acquiring a pressure wave band of the excavator;
intercepting a pressure wave band of a preset size window;
extracting time domain features of a pressure wave band of a preset size window to obtain a first feature vector;
and extracting the pressure ratio of the digging signal, the unloading signal and the rotation signal of the pressure wave band of the preset size window to obtain a second characteristic vector.
Further, the processor 320 may be further configured to:
inputting the feature vectors and the working condition mode labels into a random forest function to obtain the contribution rate of each feature vector;
sorting the feature vectors according to the contribution rate and a preset sequence;
and determining the feature vectors of the preset quantity as feature vectors after the dimension reduction.
Further, the processor 320 may be further configured to:
acquiring the wave crest and wave trough type vibration frequency of an excavating pilot signal, a discharging pilot signal, a rotary pilot signal, an arm stretching pilot signal and an arm recycling pilot signal of the excavator;
and verifying the working condition recognition result according to the wave crest and wave trough type vibration frequency of the excavating pilot signal, the discharging pilot signal, the rotary pilot signal, the bucket rod stretching pilot signal and the bucket rod recycling pilot signal.
According to the technical scheme, firstly, operation handle pilot control signal data and double pump pressure signal data of the excavator are obtained, the operation handle pilot control signal data and the double pump pressure signal data are preprocessed to obtain preprocessed data, then feature vectors are built according to the preprocessed data, feature dimension reduction is conducted on the feature vectors to obtain feature vectors after dimension reduction, finally the feature vectors after dimension reduction are input into a classification model to obtain a first working condition mode of the excavator, the first working condition mode is verified, and a working condition identification result detail table is output. According to the application, the feature vector is constructed by combining the basic time domain features of the composite signal and the actual conditions of mining, the feature vector is subjected to dimension reduction, the feature factors with smaller influence on the recognition result are removed, and after the model recognition is finished, the classification result is checked to correct the model recognition result, so that the classification precision of the model is improved.
The embodiment of the application also provides a machine-readable storage medium, wherein the machine-readable storage medium is stored with instructions for causing a machine to execute the method for identifying the working condition of the excavator.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, etc., such as Read Only Memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.
Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises an element.
The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and variations of the present application will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the application are to be included in the scope of the claims of the present application.
Claims (10)
1. A method for identifying an excavator work, the method comprising:
acquiring pilot control signal data and double-pump pressure signal data of an operating handle of the excavator;
preprocessing the pilot control signal data of the operating handle and the double pump pressure signal data to obtain preprocessed data;
constructing a feature vector according to the preprocessed data;
performing feature dimension reduction on the feature vector to obtain a dimension-reduced feature vector;
inputting the feature vector subjected to dimension reduction into a classification model to obtain a first working condition mode of the excavator;
and checking the first working condition mode and outputting a working condition identification result detail table.
2. The method according to claim 1, wherein the method further comprises:
acquiring crushing working mode state, machine construction state, machine walking state and engine rotating speed data of the excavator;
and determining a second working condition mode of the excavator according to the crushing working mode state, the machine construction state, the machine walking state and the engine rotating speed data, and outputting a working condition identification result detail table.
3. The method of claim 1, wherein the classification model is a libsvm model, the method further comprising:
target parameters of the libsvm model are determined to construct a classification model.
4. A method according to claim 3, wherein said determining target parameters of the libsvm model to construct a classification model comprises:
encoding initial parameters of the libsvm model to form an initial population, wherein the initial population comprises a plurality of individuals;
respectively determining the fitness of each individual in the initial population;
judging whether an individual with the fitness meeting a termination condition exists or not;
if it is determined that there is an individual whose fitness meets the termination condition, determining a parameter corresponding to the individual whose fitness meets the termination condition as a target parameter of the libsvm model;
and in the case that no individual with the fitness meeting the termination condition exists, selecting, crossing and mutating the initial population through genetic operators until the individual with the fitness meeting the termination condition exists.
5. The method of claim 1, wherein the pre-processing the operating handle pilot control signal data and the dual pump pressure signal data to obtain pre-processed data comprises:
performing data filling processing on the pilot control signal data of the operating handle and the double-pump pressure signal data to obtain filled data;
carrying out data alignment processing on the filled data to obtain aligned data;
and carrying out data denoising on the aligned data to obtain preprocessed data.
6. The method of claim 1, wherein the feature vector comprises a first feature vector and a second feature vector, and wherein constructing a feature vector from the preprocessed data comprises:
acquiring a pressure wave band of the excavator;
intercepting a pressure wave band of a preset size window;
extracting time domain features of the pressure wave band of the preset size window to obtain a first feature vector;
and extracting the pressure ratio of the digging signal, the unloading signal and the rotation signal of the pressure wave band of the preset size window to obtain a second characteristic vector.
7. The method of claim 1, wherein feature-reducing the feature vector to obtain a reduced feature vector comprises:
inputting the feature vectors and the working condition mode labels into a random forest function to obtain the contribution rate of each feature vector;
sorting the feature vectors according to the contribution rate and a preset sequence;
and determining the feature vectors of the preset quantity as feature vectors after the dimension reduction.
8. The method of claim 1, wherein the verifying the first operating mode comprises:
acquiring the wave crest and wave trough type vibration frequency of an excavating pilot signal, a discharging pilot signal, a rotary pilot signal, an arm stretching pilot signal and an arm recovering pilot signal of the excavator;
and verifying the working condition identification result according to the wave crest and wave trough type vibration frequency of the excavating pilot signal, the discharging pilot signal, the rotary pilot signal, the bucket rod stretching pilot signal and the bucket rod recycling pilot signal.
9. An excavator work identification device, comprising:
a memory configured to store instructions; and
a processor configured to invoke the instructions from the memory and when executing the instructions is capable of implementing the method for excavator work identification according to any one of claims 1 to 8.
10. A machine-readable storage medium having stored thereon instructions for causing a machine to perform the method for excavator work identification according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310716264.6A CN116894213A (en) | 2023-06-15 | 2023-06-15 | Method for identifying working condition of excavator and working condition identifying device of excavator |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310716264.6A CN116894213A (en) | 2023-06-15 | 2023-06-15 | Method for identifying working condition of excavator and working condition identifying device of excavator |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116894213A true CN116894213A (en) | 2023-10-17 |
Family
ID=88312849
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310716264.6A Pending CN116894213A (en) | 2023-06-15 | 2023-06-15 | Method for identifying working condition of excavator and working condition identifying device of excavator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116894213A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117807443A (en) * | 2024-02-29 | 2024-04-02 | 江苏海平面数据科技有限公司 | Training method of tractor working condition identification model and tractor working condition identification method |
-
2023
- 2023-06-15 CN CN202310716264.6A patent/CN116894213A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117807443A (en) * | 2024-02-29 | 2024-04-02 | 江苏海平面数据科技有限公司 | Training method of tractor working condition identification model and tractor working condition identification method |
| CN117807443B (en) * | 2024-02-29 | 2024-05-14 | 江苏海平面数据科技有限公司 | Training method of tractor working condition identification model and tractor working condition identification method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107153843B (en) | Ground settlement prediction system and method based on support vector machine | |
| US10495544B1 (en) | Failure detection device for detecting an issue with a part of a machine | |
| KR101967978B1 (en) | Apparatus for predicting net penetration rate of shield tunnel boring machine and method thereof | |
| CN116894213A (en) | Method for identifying working condition of excavator and working condition identifying device of excavator | |
| US20200409823A1 (en) | Method and apparatus for optimal distribution of test cases among different testing platforms | |
| CN111382472A (en) | Method and device for predicting shield-induced proximity structure deformation by random forest fusion SVM (support vector machine) | |
| CN115329657B (en) | Drilling parameter optimization method and device | |
| CN115358494B (en) | Danger early warning method for subway shield underpass construction | |
| CN106650446A (en) | Identification method and system of malicious program behavior, based on system call | |
| KR102631470B1 (en) | System for predicting leakage and failure of bank | |
| CN115423167A (en) | Subway deep foundation pit construction safety early warning and decision-making assisting method and system | |
| KR20200058258A (en) | System and method for predicting ground layer information, and a recording medium having computer readable program for executing the method | |
| Saadallah et al. | Active learning for accurate settlement prediction using numerical simulations in mechanized tunneling | |
| Sağlam et al. | Estimation of duration of earthwork with backhoe excavator by Monte Carlo Simulation | |
| CN114164878A (en) | Method, processor, system and excavator for identifying working conditions | |
| CN115186936A (en) | Optimal well pattern construction method for oil field based on GNN model | |
| CN114219917B (en) | Vulnerability restoration method for laser radar point cloud data | |
| US7054799B1 (en) | Method and system for reduction of a network topology-based system having automated optimization features | |
| CN113313304A (en) | Power grid accident abnormity analysis method and system based on big data decision tree | |
| CN116861237A (en) | Identification method and device for excavator working state, excavator and processor | |
| KR102612959B1 (en) | System and method for predicting shale gas production based on deep learning | |
| CN116834678A (en) | Engineering machinery working condition identification method and device, storage medium and processor | |
| CN115311574A (en) | Building monitoring method, equipment and medium | |
| CN115577619A (en) | Method for predicting residual shear strength of liquefied soil based on machine learning | |
| CN115169412A (en) | Method and device for determining operation mode, controller and engineering vehicle |
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 |