CN115563475A - Pressure soft sensor of excavator hydraulic system - Google Patents
Pressure soft sensor of excavator hydraulic system Download PDFInfo
- Publication number
- CN115563475A CN115563475A CN202211308682.3A CN202211308682A CN115563475A CN 115563475 A CN115563475 A CN 115563475A CN 202211308682 A CN202211308682 A CN 202211308682A CN 115563475 A CN115563475 A CN 115563475A
- Authority
- CN
- China
- Prior art keywords
- pressure
- sensor
- excavator
- gate
- hydraulic system
- 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 claims abstract description 22
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 15
- 230000008054 signal transmission Effects 0.000 claims abstract description 7
- 238000010219 correlation analysis Methods 0.000 claims abstract description 4
- 238000012216 screening Methods 0.000 claims abstract description 4
- 230000006870 function Effects 0.000 claims description 21
- 239000011159 matrix material Substances 0.000 claims description 12
- 238000005457 optimization Methods 0.000 claims description 11
- 230000004913 activation Effects 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000015654 memory Effects 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 4
- 230000001186 cumulative effect Effects 0.000 claims description 3
- 238000005315 distribution function Methods 0.000 claims description 3
- 238000003745 diagnosis Methods 0.000 abstract 1
- 238000013528 artificial neural network Methods 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 2
- 238000009412 basement excavation Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 229910002092 carbon dioxide Inorganic materials 0.000 description 1
- 239000001569 carbon dioxide Substances 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 239000000306 component Substances 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Operation Control Of Excavators (AREA)
Abstract
The invention provides a pressure soft sensor of an excavator hydraulic system, which comprises: acquiring pressure signals of each pressure sensor when the excavator works normally; performing correlation analysis on the pressure signals, and screening out signal characteristics associated with the target pressure sensor; constructing an LSTM model according to the screened signal characteristics, and updating the cell state by utilizing a forgetting gate, an input gate and an output gate; determining optimal LSTM network parameters by using a BO algorithm; obtaining a BO-LSTM pressure prediction model, and predicting a pressure signal of a target pressure sensor on the excavator through the prediction model; comparing the actual value with the predicted value of the target pressure sensor, and judging whether the pressure sensor fails or not; if the fault is not valid, the signal transmission channel is switched by an active fault-tolerant control method, and the predicted value is compensated to the controller. The pressure soft sensor is high in precision, and can be effectively used for fault diagnosis and control application of an excavator hydraulic system.
Description
Technical Field
The invention relates to a pressure soft sensor, in particular to a pressure soft sensor of an excavator hydraulic system.
Background
To improve work efficiency and reduce carbon dioxide emissions, engineering machinery speeds up the process of electromotion and intelligence, and the use of numerous sensors is the basis for digitization and informatization. The excavator is used as the king of engineering machinery, the operation environment is severe, the working conditions are complex and changeable, and the strict requirement is provided for the reliability of the parts of the excavator, particularly the high-precision high-frequency response sensor participating in the control.
In the working process, the control failure of the whole excavator caused by the failure of the sensor not only reduces the working efficiency, but also can cause inestimable personal injury even due to false operation. Due to the failure of the sensor of the traditional excavator, a driver can stop and repair the excavator at any time, and the 'unmanned excavator' cannot be automatically stopped.
In recent years, artificial intelligence models of machine learning and deep learning have become very popular in data predictive analysis. In particular, the LSTM (Long Short-Term Memory artificial neural network) network has a wide adaptability in data-driven time series prediction models, and has been successfully used to solve problems in many engineering fields (refer to li linfang, shi dazi, cheng, exploration of Long-Term Memory neural networks in mid-Term earthquake prediction-take the yunnan area as an example [ J ]. Geophysical press, 2022,65 (01): 12-25.).
Although prediction algorithms based on LSTM have worked well, the determination of network hyper-parameters is largely empirically optimized and inefficient. The selection of the hyper-parameters directly influences the performance of the prediction algorithm, so that the LSTM network parameter optimization is very important. Typical parameter optimization algorithms include grid search, ant colony optimization, particle swarm optimization and the like. The above optimization algorithm does not take into account the past evaluation results.
The problems of high failure rate and short service life of a pressure sensor on the traditional excavator are urgently needed to be solved.
Disclosure of Invention
The purpose of the invention is as follows: the invention aims to solve the technical problem of the prior art and provides a pressure soft sensor of an excavator hydraulic system.
In order to solve the technical problem, the invention discloses a pressure soft sensor of an excavator hydraulic system, which comprises the following steps:
step 2, carrying out correlation analysis on the pressure signals, and screening out signal characteristics associated with a target pressure sensor on the excavator;
step 3, constructing an LSTM (Long Short-Term Memory-Artificial neural network) model according to the screened signal characteristics, and updating the cell state by utilizing a forgetting gate, an input gate and an output gate in the LSTM model; the input gate comprises an input gate first channel and an input gate second channel;
step 4, determining the optimal LSTM model parameters by using a BO (Bayesian Optimization, BO) algorithm, and constructing a pressure prediction model of the BO-LSTM (Bayesian Optimization-based long-short term memory artificial neural network);
step 5, predicting a pressure signal of a target pressure sensor on the excavator through the pressure prediction model to obtain a predicted value;
step 6, comparing the actual measurement value with the predicted value of the target pressure sensor, and judging whether the pressure sensor fails or not;
and 7, if the target pressure sensor fails, switching a pressure signal transmission channel by an active fault-tolerant control method, and compensating a predicted value to the controller.
In the present invention, the signal characteristics in step 2 include: a first pump outlet pressure, a second pump outlet pressure, a boom lift pressure, a boom lower pressure, an arm dig pressure, an arm unload pressure, a bucket dig pressure, a bucket unload pressure, a left swing pressure, and a right swing pressure.
In the invention, the forgetting door f in the step 3 t The following:
f t ＝σ(W f ·[h t-1 ，x t ]+b f )
where σ represents an activation function; w f A weight matrix representing a forgetting gate; h is t-1 Indicating the cell status at time t-1A variable; x is the number of t An input representing time t; b f A bias term representing a forgetting gate.
Forget door f t Information to be discarded in the input sequence of the LSTM model is determined.
In the present invention, the input gate i in step 3 t The following are:
i t ＝σ(W i ·[h t-1 ，x t ]+b i )
wherein, W i A weight matrix representing a first channel of the input gate; b i An offset term representing a first channel of the input gate;
input gate i t The information to be memorized of the cell state is determined.
In the present invention, the method for updating the cell state in step 3 comprises:
inputting candidate state value C 'of cell at time t' t The following were used:
C′ t ＝tanh(W c ·[h t-1 ，x t ]+b c )
wherein, tanh represents an activation function; w c A weight matrix representing a second channel of the input gate; b c An offset term representing a second channel of the input gate.
Calculating a cell state update value C at the time t according to the candidate state values of the forgetting gate, the input gate and the input cell t The method comprises the following steps:
C t ＝f t ·C t-1 +i t ·C′ t
in the present invention, the value O of the output gate described in step 3 t And h t The calculation method of (2) is as follows:
O t ＝σ(W o ·[h t-1 ，x t ]+b o )
h t ＝O t ·tanh(C t )
wherein, W o A weight matrix representing the output gates; b o A bias term representing an output gate;
value h of the output gate t Enter LThe output layer of STM network calculates the total output at time t and updates the cell state to a cell state update value C t Proceed to the next memory cell.
In the present invention, the BO algorithm in step 4 includes: a probability agent model and an acquisition function; the specific method obtains a solution which minimizes the objective function through calculation, and the calculation method is as follows:
wherein x is an unknown hyper-parameter combination; x represents a hyper-parametric combinatorial space; f (x) is an objective function; PI (x) is an acquisition function; α is a hyperparameter, α =0 converges the value to f (x) + ) Local optimization is avoided; x is the number of + Representing an optimal hyper-parameter combination; phi (#) is a normal cumulative distribution function; μ (x) and σ (x) represent the expectation and variance, respectively, obtained from the posterior model, x next Representing the next hyper-parametric combination selected by the acquisition function.
In the present invention, the method for constructing a pressure prediction model for BO-LSTM described in step 4 comprises: and (3) constructing the LSTM network by taking any signal characteristic in the step (2) as the output of the pressure prediction model and other signal characteristics as the input of the pressure prediction model.
In the present invention, the method for determining whether the pressure sensor fails in step 6 includes: and defining a threshold value according to an actual application scene, and if the difference value between the actual measurement value and the predicted value of the target pressure sensor exceeds the threshold value, judging that the pressure sensor fails.
In the invention, the active fault-tolerant control method in the step 7 is used for switching the pressure signal transmission channel, and the pressure soft sensor is used for replacing a failure sensor to continue working after a target sensor on the excavator fails.
Has the advantages that:
1. the invention uses the pressure soft measurer based on the neural network for software redundancy control or replaces the traditional pressure sensor, thereby effectively improving the reliability of the excavator control system.
2. The invention utilizes the LSTM network model to predict the pressure signal of the pressure sensor of the excavator.
3. The invention determines the optimal parameters of the LSTM network model through the BO algorithm, thereby reducing the LSTM training cost.
Drawings
The foregoing and/or other advantages of the invention will become further apparent from the following detailed description of the invention when taken in conjunction with the accompanying drawings.
FIG. 1 is a schematic diagram of the general architecture of the present invention.
FIG. 2 is a schematic representation of the LSTM model of the invention.
FIG. 3 is a schematic diagram of the main oil circuit of the hydraulic system of the excavator in the embodiment of the invention.
FIG. 4 is a schematic flow chart of the present invention.
FIG. 5 is a diagram illustrating the predicted effect of a conventional LSTM model in an embodiment of the present invention.
FIG. 6 is a schematic diagram of the BO-LSTM model prediction effect in the embodiment of the present invention.
Detailed Description
In order that the advantages of the invention will be readily understood, a more particular description of the invention briefly described above will be rendered by reference to specific embodiments that are illustrated in the appended drawings, which are provided to illustrate, but not to limit the scope of the invention.
A pressure soft sensor of an excavator hydraulic system trains an LSTM prediction model by using data under the normal work of an excavator, and then determines the optimal parameters of the LSTM prediction model by using a BO algorithm so as to improve the prediction accuracy.
The invention specifically adopts the following scheme:
a pressure soft sensor of a hydraulic system of an excavator is shown in figure 4, and comprises the following steps:
step 2, carrying out correlation analysis on the pressure signals, and screening out signal characteristics associated with the target pressure sensor;
the screened pressure signal characteristics include: the first pump outlet pressure, the second pump outlet pressure, the boom raising pressure, the boom lowering pressure, the arm digging pressure, the arm unloading pressure, the bucket digging pressure, the bucket unloading pressure, the left swing pressure, and the right swing pressure are used as input or output of the LSTM network in subsequent training.
Step 3, constructing an LSTM model according to the screened signal characteristics, wherein an LSTM memory cell is shown in fig. 2, and the states of the cell are updated by utilizing a forgetting gate, an input gate and an output gate;
forget door f t The information to be discarded in the input sequence is determined as follows (1):
f t ＝σ(W f ·[h t-1 ，x t ]+b f )......(1)
in formula (1), σ represents an activation function; w f A weight matrix representing the channel; h is t-1 Represents a cell state variable at the t-1 moment; x is the number of t An input representing time t; b is a mixture of f Representing the bias term for that channel.
Input door i t The information to be memorized for determining the cell state is represented by the following formula (2):
i t ＝σ(W i ·[h t-1 ，x t ]+b i )......(2)
in the formula (2), W i A weight matrix representing the channel; b is a mixture of i Representing the bias term for that channel.
Inputting candidate state value C 'of cell at time t' t The following formula (3) is calculated:
C′ t ＝tanh(W c ·[h t-1 ，x t ]+b c )......(3)
in formula (3), tanh represents an activation function; w is a group of c A weight matrix representing the channel; b c Representing the bias term for that channel.
The cell state update value C at time t can be calculated from the expressions (1), (2) and (3) t The following formula (4):
C t ＝f t ·C t-1 +i t ·C′ t ......(4)
finally, the value O of the output gate is calculated t And h t ，h t The output layer entering the LSTM network calculates the total output at time t, and simultaneously with the cell state C t Entering the next memory unit as shown in the following formulas (5) and (6):
O t ＝σ(W o ·[h t-1 ，x t ]+b o )......(5)
h t ＝O t ·tanh(C t )......(6)
in the formula (5), W o A weight matrix representing the channel; b o Representing the bias term for that channel.
Step 4, determining the optimal LSTM network parameters by using a BO algorithm;
the BO algorithm consists of two core components (probabilistic proxy model and acquisition function) whose goal is to compute an optimum point that minimizes the objective function as calculated by the following equations (7) to (9):
in the formulas (7) to (9), x is an unknown hyper-parameter combination; f (x) is the objective function. PI (x) is an acquisition function; α is a hyperparameter, α =0So that the value converges on f (x) + ) Local optimization is avoided; Φ (—) is a normal cumulative distribution function.
Step 5, obtaining a BO-LSTM pressure prediction model, and predicting a pressure signal of a target pressure sensor on the excavator through the prediction model;
and taking the bucket arm excavation pressure signal as an output, and taking other pressure signals as inputs to construct an LSTM network.
Step 6, comparing the actual value with the predicted value of the target pressure sensor, and judging whether the pressure sensor fails or not;
and defining a threshold value according to an actual application scene, and if the difference value between the actual value and the predicted value exceeds the threshold value, judging that the pressure sensor fails.
And 7, if the pressure sensor fails, compensating the predicted value to the controller, as shown in fig. 1. The soft pressure sensor functions as a "fault observer". Under normal conditions, the signal transmission of the pressure sensor is in a channel 1; when the target pressure sensor is observed to be invalid, the signal transmission is switched to the channel 2 by an active fault-tolerant control method, and finally the predicted value is compensated to the controller.
Examples
In order to verify the superiority of the pressure soft sensor of the excavator hydraulic system, after data collected when the excavator normally works under the 90-degree swing working condition is processed, relevant pressure signal characteristics are screened out by analyzing a main oil way (shown in figure 3) of the excavator hydraulic system. In the invention, the core of the pressure soft sensor is a BO-LSTM prediction model. And (3) obtaining an optimal network model through mutual training of a BO algorithm and the LSTM network, thereby predicting the pressure value of the failed sensor. The soft measurement method enables the excavator to form a closed-loop pressure control system, if the difference value between the actual value and the predicted value of the pressure sensor exceeds a threshold value, the pressure sensor is judged to be invalid, and the predicted value is transmitted to the controller through active fault-tolerant control.
In an embodiment, any 1 set of pressure signals may be taken as output, and the other 9 sets of pressure signals may be taken as input, and the data set may be divided into a training set, a validation set, and a test set in a ratio of 7: 2: 1. Since the boom system structure and the operation during work are the most complicated, the boom excavation pressure is used as an output.
12000 groups of data (sampling frequency 50 Hz) are selected for prediction experiments by the method, the predicted values and the actual values of the traditional LSTM model (shown in figure 5) and the BO-LSTM model (shown in figure 6) are subjected to curve fitting respectively, and the higher the degree of curve fitting, the better the prediction effect. Through comparison, the predicted value of the LSTM model after parameter optimization by using the BO algorithm to the bucket rod excavating pressure is closer to an actual value compared with the traditional LSTM model, the Root Mean Square Error (RMSE) and the average absolute error (MAE) are smaller than those of the traditional LSTM model, and the experimental results are shown in table 1.
TABLE 1 Experimental results Table
According to practical application, the sudden misoperation of the working device needs to be avoided for the continuous control of the excavator, so that the prediction error needs to be controlled within 30bar, and more continuous prediction outliers need to be avoided.
In specific implementation, the application provides a computer storage medium and a corresponding data processing unit, wherein the computer storage medium can store a computer program, and the computer program can run the inventive content of the pressure soft sensor of the excavator hydraulic system and some or all steps in each embodiment provided by the invention when being executed by the data processing unit. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), a Random Access Memory (RAM), or the like.
It is clear to those skilled in the art that the technical solutions in the embodiments of the present invention can be implemented by means of a computer program and its corresponding general-purpose hardware platform. Based on such understanding, the technical solutions in the embodiments of the present invention may be embodied in the form of a computer program, that is, a software product, which may be stored in a storage medium and includes several instructions to enable a device (which may be a personal computer, a server, a single chip, an MUU, or a network device) including a data processing unit to execute the method in each embodiment or some parts of the embodiments of the present invention.
While the present invention provides a method and system for a pressure soft sensor of a hydraulic system of an excavator, and the method and system for implementing the technical solution are numerous, the above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, a plurality of modifications and embellishments can be made without departing from the principle of the present invention, and these modifications and embellishments should also be regarded as the protection scope of the present invention. All the components not specified in the present embodiment can be realized by the prior art.
Claims (10)
1. A pressure soft sensor of a hydraulic system of an excavator is characterized by comprising the following steps:
step 1, acquiring pressure signals of pressure sensors when an excavator works normally;
step 2, carrying out correlation analysis on the pressure signals, and screening out signal characteristics associated with a target pressure sensor on the excavator;
step 3, constructing an LSTM model according to the screened signal characteristics, and updating the cell state by utilizing a forgetting gate, an input gate and an output gate in the LSTM model; the input gate comprises an input gate first channel and an input gate second channel;
step 4, determining the optimal parameters of the LSTM model by using a BO algorithm, and constructing a pressure prediction model of the BO-LSTM;
step 5, predicting a pressure signal of a target pressure sensor on the excavator through the pressure prediction model to obtain a predicted value;
step 6, comparing the actual measurement value with the predicted value of the target pressure sensor, and judging whether the pressure sensor fails or not;
and 7, if the target pressure sensor fails, switching a pressure signal transmission channel by an active fault-tolerant control method, and compensating a predicted value to the controller.
2. The pressure soft sensor of the hydraulic system of the excavator according to claim 1, wherein the signal characteristic in the step 2 comprises: a first pump outlet pressure, a second pump outlet pressure, a boom lift pressure, a boom lower pressure, an arm dig pressure, an arm unload pressure, a bucket dig pressure, a bucket unload pressure, a left swing pressure, and a right swing pressure.
3. The pressure soft sensor of the hydraulic system of the excavator according to claim 2, wherein the forgetting door f in the step 3 t The following are:
f t ＝σ(W f ·[h t-1 ,x t ]+b f )
where σ represents an activation function; w f A weight matrix representing a forgetting gate; h is t-1 Represents a cell state variable at the t-1 moment; x is the number of t An input representing time t; b f A bias term representing a forgetting gate;
forget door f t Information to be discarded in the input sequence of the LSTM model is determined.
4. The pressure soft sensor of hydraulic system of excavator according to claim 3, wherein the input gate i in step 3 t The following are:
i t ＝σ(W i ·[h t-1 ,x t ]+b i )
wherein, W i A weight matrix representing a first channel of the input gate; b i An offset term representing a first channel of the input gate;
input door i t The information to be memorized of the cell state is determined.
5. The pressure soft sensor of the hydraulic system of the excavator according to claim 4, wherein the method for updating the cell state in the step 3 comprises:
inputting candidate state value C 'of cell at time t' t The following:
C′ t ＝tanh(W c ·[h t-1 ,x t ]+b c )
wherein, tanh represents an activation function; w c A weight matrix representing a second channel of the input gate; b c An offset term representing a second channel of the input gate;
calculating a cell state update value C at the time t according to the forgetting gate, the input gate and the candidate state value of the input cell t The method comprises the following steps:
C t ＝f t ·C t-1 +i t ·C′ t 。
6. the pressure soft sensor of hydraulic system of excavator according to claim 5, wherein the value O of output gate in step 3 t And h t The calculation method of (2) is as follows:
O t ＝σ(W o ·[h t-1 ,x t ]+b o )
h t ＝O t ·tanh(C t )
wherein, W o A weight matrix representing the output gates; b o A bias term representing an output gate;
value h of the output gate t Entering the output layer of the LSTM network, calculating the total output at the time t, and simultaneously updating the cell state, namely the cell state t Proceed to the next memory cell.
7. The soft pressure sensor for an excavator hydraulic system as claimed in claim 6, wherein the BO algorithm in the step 4 comprises: a probabilistic proxy model and an acquisition function; the specific method obtains a solution which minimizes the objective function through calculation, and the calculation method is as follows:
wherein x is an unknown hyper-parameter combination; x represents a hyper-parametric combinatorial space; f (x) is an objective function; PI (x) is an acquisition function; α is a hyperparameter, α =0 makes the value converge to f (x) + ) Local optimization is avoided; x is the number of + Representing an optimal hyper-parameter combination; phi (#) is a normal cumulative distribution function; μ (x) and σ (x) represent the expectation and variance, respectively, obtained from the posterior model, x next Representing the next hyper-parametric combination selected by the acquisition function.
8. The soft pressure sensor for hydraulic system of excavator according to claim 7, wherein the method for constructing pressure prediction model of BO-LSTM in step 4 comprises: and (3) constructing an LSTM network by taking any signal characteristic in the step (2) as the output of the pressure prediction model and other signal characteristics as the input of the pressure prediction model.
9. The pressure soft sensor of the hydraulic system of the excavator according to claim 8, wherein the method for determining whether the pressure sensor fails in step 6 comprises: and defining a threshold value according to an actual application scene, and if the difference value between the actual measurement value and the predicted value of the target pressure sensor exceeds the threshold value, judging that the pressure sensor fails.
10. The pressure soft sensor of the hydraulic system of the excavator as claimed in claim 9, wherein the active fault-tolerant control method in step 7 is used for switching the pressure signal transmission channel to ensure that the pressure soft sensor is used for replacing a failed sensor to continue working after a target sensor on the excavator fails.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211308682.3A CN115563475A (en) | 2022-10-25 | 2022-10-25 | Pressure soft sensor of excavator hydraulic system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211308682.3A CN115563475A (en) | 2022-10-25 | 2022-10-25 | Pressure soft sensor of excavator hydraulic system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115563475A true CN115563475A (en) | 2023-01-03 |
Family
ID=84746359
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211308682.3A Pending CN115563475A (en) | 2022-10-25 | 2022-10-25 | Pressure soft sensor of excavator hydraulic system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115563475A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117150415A (en) * | 2023-10-25 | 2023-12-01 | 智隆(广州)网络科技有限公司 | Communication equipment state monitoring method and system based on artificial intelligence |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016145850A1 (en) * | 2015-03-19 | 2016-09-22 | 清华大学 | Construction method for deep long short-term memory recurrent neural network acoustic model based on selective attention principle |
| CN109084931A (en) * | 2018-07-19 | 2018-12-25 | 电子科技大学 | A kind of sensor mistuning calibration function method |
| CN113176022A (en) * | 2021-05-12 | 2021-07-27 | 南京邮电大学 | Segmented neural network pressure sensor pressure detection method and system |
| WO2021232655A1 (en) * | 2020-05-20 | 2021-11-25 | 国网河北省电力有限公司电力科学研究院 | Method for evaluating mechanical state of high-voltage shunt reactor on the basis of vibration feature |
| CN114065667A (en) * | 2021-11-19 | 2022-02-18 | 国网上海市电力公司 | SF6 equipment gas pressure prediction method based on Prophet-LSTM model |
| CN115081279A (en) * | 2022-06-16 | 2022-09-20 | 西安理工大学 | PCCP pipeline structure damage evaluation method under rockfall impact |
-
2022
- 2022-10-25 CN CN202211308682.3A patent/CN115563475A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016145850A1 (en) * | 2015-03-19 | 2016-09-22 | 清华大学 | Construction method for deep long short-term memory recurrent neural network acoustic model based on selective attention principle |
| CN109084931A (en) * | 2018-07-19 | 2018-12-25 | 电子科技大学 | A kind of sensor mistuning calibration function method |
| WO2021232655A1 (en) * | 2020-05-20 | 2021-11-25 | 国网河北省电力有限公司电力科学研究院 | Method for evaluating mechanical state of high-voltage shunt reactor on the basis of vibration feature |
| CN113176022A (en) * | 2021-05-12 | 2021-07-27 | 南京邮电大学 | Segmented neural network pressure sensor pressure detection method and system |
| CN114065667A (en) * | 2021-11-19 | 2022-02-18 | 国网上海市电力公司 | SF6 equipment gas pressure prediction method based on Prophet-LSTM model |
| CN115081279A (en) * | 2022-06-16 | 2022-09-20 | 西安理工大学 | PCCP pipeline structure damage evaluation method under rockfall impact |
Non-Patent Citations (3)
| Title |
|---|
| 李兆廷等: "《Python数据分析与应用案例教程》", 31 August 2022, pages: 110 - 111 * |
| 闫万梓: ""基于多源信息的综采工作面智能来压预警模型研究"", 《中国优秀硕士学位论文全文数据库 工程科技I辑》, 15 March 2022 (2022-03-15), pages 31 - 42 * |
| 闻新等: "《控制系统的故障诊断和容错控制》", 29 February 1988, pages: 214 - 215 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117150415A (en) * | 2023-10-25 | 2023-12-01 | 智隆(广州)网络科技有限公司 | Communication equipment state monitoring method and system based on artificial intelligence |
| CN117150415B (en) * | 2023-10-25 | 2024-02-06 | 智隆(广州)网络科技有限公司 | Communication equipment state monitoring method and system based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111639430B (en) | Natural gas pipeline leakage identification system driven by digital twinning | |
| CN112926273B (en) | Method for predicting residual life of multivariate degradation equipment | |
| CN111339712A (en) | Method for predicting residual life of proton exchange membrane fuel cell | |
| CN110288046B (en) | Fault prediction method based on wavelet neural network and hidden Markov model | |
| CN114925623B (en) | Oil and gas reservoir yield prediction method and system | |
| Fu et al. | A hybrid deep learning approach for dynamic attitude and position prediction in tunnel construction considering spatio-temporal patterns | |
| CN112784336A (en) | Bridge static displacement prediction technology based on deep learning LSTM network | |
| CN114186475B (en) | Slewing bearing service life prediction method based on Attention-MGRU | |
| CN111638707B (en) | Intermittent process fault monitoring method based on SOM clustering and MPCA | |
| CN113420509A (en) | Wind turbine state evaluation method and device and storage medium | |
| CN115563475A (en) | Pressure soft sensor of excavator hydraulic system | |
| CN115952590B (en) | BO-RF-MOMPA-based shield tunnel optimization design method and equipment | |
| CN115759409A (en) | Water gate deformation prediction method for optimizing LSTM (least Square TM) model by multi-time mode attention mechanism | |
| CN116629136A (en) | Method, device, equipment and storage medium for updating digital twin model | |
| CN116305614A (en) | Rolling bearing residual service life prediction method based on residual convolution neural network and LSTM | |
| CN115660198B (en) | Method for predicting residual service life of rolling bearing | |
| CN109610549B (en) | Frequency conversion intelligent control system and control method for hydraulic excavator electric appliance | |
| CN116933441A (en) | Digital twin data driving model modeling method for autonomous shoveling process of loader | |
| CN116147917A (en) | Construction method, device and equipment for predicting residual service life label of bearing | |
| CN112911532B (en) | Foundation pit big data detection system | |
| CN112232557B (en) | Short-term prediction method for health degree of switch machine based on long-short-term memory network | |
| CN115660091A (en) | Method for bit quantum overturning early warning and threshold setting of quantum computer | |
| CN114841063A (en) | Aero-engine residual life prediction method based on deep learning | |
| CN114692490A (en) | Method for diagnosing faults of rotating machinery by using clear double spectrograms and improved convolutional neural network | |
| CN114201914A (en) | Solid engine test data analysis method based on BP neural network |
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 |