CN115047313A - ZPW-2000R track circuit fault diagnosis method and device based on OC-SVM and DNN - Google Patents
ZPW-2000R track circuit fault diagnosis method and device based on OC-SVM and DNN Download PDFInfo
- Publication number
- CN115047313A CN115047313A CN202210621985.4A CN202210621985A CN115047313A CN 115047313 A CN115047313 A CN 115047313A CN 202210621985 A CN202210621985 A CN 202210621985A CN 115047313 A CN115047313 A CN 115047313A
- Authority
- CN
- China
- Prior art keywords
- track circuit
- data
- section
- circuit signal
- signal data
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000003745 diagnosis Methods 0.000 title claims abstract description 29
- 238000012549 training Methods 0.000 claims abstract description 85
- 238000007781 pre-processing Methods 0.000 claims abstract description 22
- 230000006870 function Effects 0.000 claims description 59
- 238000004364 calculation method Methods 0.000 claims description 15
- 230000004913 activation Effects 0.000 claims description 14
- 238000013507 mapping Methods 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 12
- 239000013598 vector Substances 0.000 claims description 12
- 238000013528 artificial neural network Methods 0.000 claims description 8
- 238000005457 optimization Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 102100029469 WD repeat and HMG-box DNA-binding protein 1 Human genes 0.000 description 1
- 101710097421 WD repeat and HMG-box DNA-binding protein 1 Proteins 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/28—Testing of electronic circuits, e.g. by signal tracer
- G01R31/2832—Specific tests of electronic circuits not provided for elsewhere
- G01R31/2836—Fault-finding or characterising
- G01R31/2843—In-circuit-testing
-
- 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)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (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)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Train Traffic Observation, Control, And Security (AREA)
- Digital Transmission Methods That Use Modulated Carrier Waves (AREA)
Abstract
The invention provides a ZPW-2000R track circuit fault diagnosis method and device based on an OC-SVM and a DNN. The method comprises the following steps: acquiring track circuit signal data of a monitored section at the current moment; identifying the type of the section and preprocessing the signal data of the track circuit; predicting the preprocessed track circuit signal data by adopting an OC-SVM prediction model, and judging whether the track circuit signal data belongs to normal data or novel data; if the data belong to normal data, directly sending the track circuit signal data to a DNN prediction model for fault classification prediction; if the new data belongs to the novel data, continuing to detect the track circuit signal data at the next moment, and if each piece of track circuit signal data collected in a certain continuous time period from the current moment is the novel data, adding the track circuit signal data corresponding to the current moment into a training data set, and setting a new fault label for the training data set; and training by using the new training data set to obtain a new OC-SVM prediction model and a new DNN prediction model.
Description
Technical Field
The invention relates to the technical field of track circuits, in particular to a ZPW-2000R track circuit fault diagnosis method and device based on OC-SVM and DNN.
Background
The track circuit is an important part for the safe operation of the railway train. The existing track circuit has various faults due to multiple components and complex external environment factors, and the current diagnosis of the faults of the track circuit still depends on artificial experience, so that the problems of long overhaul time, complex flow, large artificial workload and the like after the faults occur are caused, and the normal operation of a train is greatly influenced.
With the development of artificial intelligence technology, a plurality of machine learning algorithms show good effects in various fields, and some learners introduce the intelligent algorithms into the research of track circuit fault diagnosis. However, most of the current researches on the fault diagnosis of the track circuit adopt a neural network method or other machine learning methods to perform supervised learning on some fault type data samples, the method needs a large amount of labeled fault data to perform algorithm training so as to be operable, and once an unknown or brand-new fault is met, the model cannot be identified, so that the model cannot be applied to an actual working scene.
Disclosure of Invention
In order to solve the problem of automatic identification of unknown or brand-new faults, the invention provides a ZPW-2000R track circuit fault diagnosis method and device based on OC-SVM and DNN.
In one aspect, the invention provides a ZPW-2000R track circuit fault diagnosis method based on OC-SVM and DNN, which comprises the following steps:
step 1: acquiring track circuit signal data of a monitored section at the current moment;
step 2: identifying the section type of the monitored section and carrying out data preprocessing on the track circuit signal data according to a data preprocessing mode corresponding to the section type;
and step 3: predicting the preprocessed track circuit signal data by adopting a trained OC-SVM prediction model, and judging whether the track circuit signal data belongs to normal data or novel data according to a prediction result;
and 4, step 4: if the track circuit signal data belong to normal data, directly sending the track circuit signal data to a DNN prediction model for fault classification prediction;
and 5: if the track circuit signal data belong to novel data, continuing to detect the track circuit signal data of the monitored section at the next moment, if each piece of track circuit signal data collected in a certain continuous time period from the current moment is novel data, adding the track circuit signal data corresponding to the current moment into a training data set, and setting a new fault label for the training data set;
step 6: and training by using the new training data set to obtain a new OC-SVM prediction model and a new DNN prediction model so as to detect the signal data of the subsequent track circuit.
Further, the section types include a no rear section, a no front section, a middle section, and a no front no rear section.
Further, step 2 specifically includes:
if the monitored section is a section without a rear section, merging the track circuit signal data of the monitored section and the track circuit signal data of the section in front of the monitored section, and deleting the data item of which the characteristic value is not digital;
if the monitored section is a section without a front section, combining the track circuit signal data of the monitored section with the track circuit signal data of the section behind the monitored section, and deleting the data item of which the characteristic value is not digital;
if the monitored section is the middle section, merging the track circuit signal data of the monitored section with the track circuit signal data of the front and rear sections of the monitored section, and deleting the data item of which the characteristic value is not digital;
and if the monitored section is a section without a front part and a rear part, directly deleting the data item of which the characteristic value is not digital in the track circuit signal data.
Further, the training process of the OC-SVM predictive model comprises:
step A1: loading a training data set, wherein each track circuit signal data in the training data set is used as a sample, and each sample corresponds to a fault label;
step A2: mapping each sample to a high-dimensional feature space through a kernel function, and solving an optimal hypersphere in the feature space by using a formula (1);
s.t.ω Τ φ(χ i )≥ρ-ξ i
ξ i ≥0,i＝1,2,……,l (1)
wherein, ω and ρ represent normal vector and offset of separating hyperplane in feature space, respectively; xi i A relaxation variable representing the ith sample, v is a parameter controlling the upper limit of the number of outliers and the lower limit of the number of all support vectors, l is the number of samples, and phi (χ) is a function mapping the samples x to a feature space;
step A3: obtaining a classification decision function shown in formula (2) by using a Lagrange multiplier method, wherein the classification decision function is an OC-SVM prediction model, and if f (x) is equal to 1, the classification decision function indicates that the sample x is normal data, and if f (x) is equal to-1, the classification decision function indicates that the sample x is novel data;
wherein, kappa (χ) i χ) represents a kernel function; alpha (alpha) ("alpha") i Is a sample x i The corresponding lagrangian coefficient.
Further, the training process of the DNN prediction model includes:
step B1: loading a training data set, wherein each track circuit signal data in the training data set is used as a sample, and each sample corresponds to a fault label;
step B2: constructing a deep neural network framework, setting network parameters, and initializing a weight value w and a bias item b at random; the network parameters comprise the number of hidden layers, the number of nodes in each hidden layer and a learning rate;
step B3: setting a gradient optimization method, a hidden layer activation function and an output layer activation function;
step B4: and performing forward calculation, feedback calculation and parameter updating on all samples, and determining the optimal values of the weight value w and the bias item b so as to obtain the DNN prediction model.
In another aspect, the present invention provides a ZPW-2000R rail circuit fault diagnosis apparatus based on OC-SVM and DNN, comprising:
the data acquisition module is used for acquiring track circuit signal data of the monitored section;
the data preprocessing module is used for identifying the section type of the monitored section and preprocessing the track circuit signal data according to a data preprocessing mode corresponding to the section type;
the novel data prediction module is used for predicting the preprocessed track circuit signal data by adopting a trained OC-SVM prediction model and judging whether the track circuit signal data belongs to normal data or novel data according to a prediction result;
the fault classification module is used for directly sending the track circuit signal data belonging to the normal data to the DNN prediction model for fault classification prediction;
the training data set updating module is used for adding the track circuit signal data at the current moment meeting the preset conditions to a training data set and setting a new fault label for the training data set; the preset conditions include: each track circuit signal data collected in a certain continuous time period from the current moment is novel data;
and the model training module is used for obtaining an OC-SVM prediction model and a DNN prediction model by training through the training data set.
Further, the section types include a no rear section, a no front section, a middle section, and a no front no rear section.
Further, the data preprocessing module is specifically configured to:
if the monitored section is a section without a rear section, merging the track circuit signal data of the monitored section and the track circuit signal data of the section in front of the monitored section, and deleting the data item of which the characteristic value is not digital;
if the monitored section is a section without a front section, merging the track circuit signal data of the monitored section and the track circuit signal data of the section behind the monitored section, and deleting the data item of which the characteristic value is not digital;
if the monitored section is the middle section, merging the track circuit signal data of the monitored section with the track circuit signal data of the front and rear sections of the monitored section, and deleting the data item of which the characteristic value is not digital;
and if the monitored section is a section without a front part and a rear part, directly deleting the data item of which the characteristic value is not digital in the track circuit signal data.
Further, the process of training the model training module to obtain the OC-SVM prediction model comprises:
step A1: loading a training data set, wherein each track circuit signal data in the training data set is used as a sample, and each sample corresponds to a fault label;
step A2: mapping each sample to a high-dimensional feature space through a kernel function, and solving an optimal hypersphere in the feature space by using a formula (1);
s.t.ω Τ φ(χ i )≥ρ-ξ i
ξ i ≥0,i＝1,2,……,l (1)
where, ω and ρRespectively representing a normal vector and an offset of a separating hyperplane in a feature space; xi i A relaxation variable representing the ith sample, v is a parameter controlling the upper limit of the number of outliers and the lower limit of the number of all support vectors, l is the number of samples, and phi (χ) is a function mapping the sample x to a feature space;
step A3: obtaining a classification decision function shown in formula (2) by using a Lagrange multiplier method, wherein the classification decision function is an OC-SVM prediction model, and if f (x) is equal to 1, the classification decision function indicates that the sample x is normal data, and if f (x) is equal to-1, the classification decision function indicates that the sample x is novel data;
wherein, kappa (χ) i χ) represents a kernel function; alpha is alpha i Is a sample x i The corresponding lagrangian coefficient.
Further, the process of training the DNN prediction model by the model training module includes:
step B1: loading a training data set, wherein each track circuit signal data in the training data set is used as a sample, and each sample corresponds to a fault label;
step B2: constructing a deep neural network framework, setting network parameters, and initializing a weight value w and a bias item b at random; the network parameters comprise the number of hidden layers, the number of nodes in each hidden layer and a learning rate;
step B3: setting a gradient optimization method, a hidden layer activation function and an output layer activation function;
step B4: and performing forward calculation, feedback calculation and parameter updating on all samples, and determining the optimal values of the weight value w and the bias term b to obtain the DNN prediction model.
The invention has the beneficial effects that:
aiming at the problems that the current track circuit has various faults and high-quality label data is difficult to obtain, the ZPW-2000R track circuit fault diagnosis method and device based on OC-SVM and DNN provided by the invention utilize the OC-SVM to perform single-classification recognition on 48 monitoring data acquired by the ZPW-2000R track circuit so as to collect signal data of different fault types; and then training the collected fault data by utilizing a DNN method so as to automatically classify the known fault data. A large number of experiments are carried out on the method by using ZPW-2000R track circuit data, and the experimental results show that the fault diagnosis method provided by the invention can effectively and automatically collect novel fault data and can accurately detect the fault type.
Drawings
Fig. 1 is a schematic flow chart of a ZPW-2000R rail circuit fault diagnosis method based on OC-SVM and DNN according to an embodiment of the present invention;
FIG. 2 is a schematic illustration of sector types provided in accordance with an embodiment of the present invention;
FIG. 3 is a schematic diagram of a deep neural network training process provided by an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a ZPW-2000R rail circuit fault diagnosis apparatus based on OC-SVM and DNN according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly described below with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1
As shown in fig. 1, an embodiment of the present invention provides a ZPW-2000R rail circuit fault diagnosis method based on OC-SVM and DNN, including the following steps:
s101: acquiring track circuit signal data of a monitored section at the current moment;
s102: identifying the section type of the monitored section and carrying out data preprocessing on the track circuit signal data according to a data preprocessing mode corresponding to the section type;
specifically, different sections of different types of stations at different positions have different circuit structures due to different equipment adopted beside the outdoor track, so that different section types have different characteristic attributes and need to be subjected to differentiation processing. The section types include four broad categories of no-rear section, no-front section, middle section, and no-front no-rear section. Wherein, (1) there is no rear section: forward non-back zone data (neighbor jurisdiction cannot acquire), such as the a, e, c, g zones in fig. 2; (2) no front section: forward non-predecessor segment data (neighbor jurisdiction cannot acquire), such as d, h, b, f segments in fig. 2; (3) a middle section: the data of the rear section and the front section are complete; (4) no front and no rear section: the forward direction has no forward and no backward sector data, such as the intra-station sector in fig. 2.
As an implementation manner, the data preprocessing of the track circuit signal data according to the data preprocessing manner corresponding to the section type specifically includes:
if the monitored section is a section without a rear section, merging the track circuit signal data of the monitored section and the track circuit signal data of the section in front of the monitored section, and deleting the data item of which the characteristic value is not digital;
for example, if the monitored segment is 1485BG of fig. 1, it is necessary to combine the track circuit signal data of 1485BG and 1485AG, and then delete the data item whose characteristic value is not digital, so as to input the data item into the OC-SVM prediction model for processing.
If the monitored section is a section without a front section, merging the track circuit signal data of the monitored section and the track circuit signal data of the section behind the monitored section, and deleting the data item of which the characteristic value is not digital;
for example, if the monitored sector is 1510AG sector in fig. 1, it is necessary to combine the track circuit signal data of 1510AG and 1510BG sectors, and then delete the data item whose characteristic value is not a number for subsequent input into the OC-SVM prediction model for processing.
If the monitored section is the middle section, merging the track circuit signal data of the monitored section with the track circuit signal data of the front and rear sections of the monitored section, and deleting the data item of which the characteristic value is not digital;
for example, if the monitored segment is the segment 1505AG in fig. 1, it is necessary to combine the track circuit signal data of the three segments 1505AG, 1505BG and 1525BG, and then delete the data item whose characteristic value is not digital, so as to input the data item into the OC-SVM prediction model for processing.
And if the monitored section is a section without a front part and a rear part, directly deleting the data item of which the characteristic value is not digital in the track circuit signal data.
S103: predicting the preprocessed track circuit signal data by adopting a trained OC-SVM prediction model, and judging whether the track circuit signal data belongs to normal data or novel data according to a prediction result;
specifically, the prediction result of the OC-SVM prediction model includes two decision values (e.g., 1 and-1), and it is assumed that the prediction result corresponding to the decision value 1 indicates that the track circuit signal data belongs to normal data, and the prediction result corresponding to the decision value-1 indicates that the track circuit signal data belongs to novel data, so that if the decision value after the track circuit signal data preprocessed at the current time is input to the OC-SVM prediction model is 1, the track circuit signal data is represented as normal data, and otherwise, the track circuit signal data is represented as novel data.
S104: if the track circuit signal data belong to normal data, directly sending the track circuit signal data to a DNN prediction model for fault classification prediction;
specifically, the track circuit signal data is input to the DNN prediction model to perform forward calculation, and then the corresponding fault category can be obtained.
S105: if the track circuit signal data belong to novel data, continuing to detect the track circuit signal data of the monitored section at the next moment, if each piece of track circuit signal data collected in a certain continuous time period from the current moment is novel data, adding the track circuit signal data corresponding to the current moment into a training data set, and setting a new fault label for the training data set;
as an implementation manner, if each piece of track circuit signal data acquired at ten consecutive sampling times from the current time is novel data, the track circuit signal data corresponding to the current time is added to a training data set, and a new fault label is set for the track circuit signal data.
For example, from time t, if the decision values of 10 times continuously output by the OC-SVM prediction model are all-1, it is considered that the track circuit signal data corresponding to time t is added to the training data set, and a new fault label is set for the track circuit signal data.
S106: and training by using the new training data set to obtain a new OC-SVM prediction model and a new DNN prediction model so as to detect the subsequent track circuit signal data.
On the basis of the above embodiment, the training process of the OC-SVM predictive model includes:
step A1: loading a training data set, wherein each track circuit signal data in the training data set is used as a sample, and each sample corresponds to a fault label;
step A2: mapping each sample to a high-dimensional feature space through a kernel function, and solving an optimal hypersphere in the feature space by using a formula (1);
s.t.ω Τ φ(χ i )≥ρ-ξ i
ξ i ≥0,i＝1,2,……,l (1)
wherein, ω and ρ represent normal vector and offset of separating hyperplane in feature space, respectively; xi i A relaxation variable representing the ith sample, v is a parameter controlling the upper limit of the number of outliers and the lower limit of the number of all support vectors, l is the number of samples, and phi (χ) is a function mapping the samples x to a feature space;
step A3: obtaining a classification decision function shown in formula (2) by using a Lagrange multiplier method, wherein the classification decision function is an OC-SVM prediction model, and if f (x) is equal to 1, the classification decision function indicates that the sample x is normal data, and if f (x) is equal to-1, the classification decision function indicates that the sample x is novel data;
wherein, kappa (χ) i χ) represents a kernel function, and replaces inner product calculation in a feature space; alpha is alpha i Is a sample x i The corresponding lagrangian coefficient.
As an implementable embodiment, as shown in fig. 3, the training process of the DNN prediction model includes:
step B1: loading a training data set, wherein each track circuit signal data in the training data set is used as a sample, and each sample corresponds to a fault label;
step B2: constructing a deep neural network framework, setting network parameters, and initializing a weight value w and a bias item b at random; the network parameters comprise the number of hidden layers, the number of nodes in each hidden layer and a learning rate;
for example, the number of hidden layers is 2, the number of hidden layer nodes is 20, and the learning rate is 0.01.
Step B3: setting a gradient optimization method, a hidden layer activation function and an output layer activation function;
for example, the gradient optimization method is an Admin algorithm, the hidden layer activation function is a tanh function, and the output layer activation function is a sigmoid function.
Step B4: and performing forward calculation, feedback calculation and parameter updating on all samples, and determining the optimal values of the weight value w and the bias item b so as to obtain the DNN prediction model.
Example 2
As shown in fig. 4, an embodiment of the present invention provides a ZPW-2000R rail circuit fault diagnosis apparatus based on OC-SVM and DNN, including: the system comprises a data acquisition module, a data preprocessing module, a novel data prediction module, a fault classification module, a training data set updating module and a model training module. Wherein:
the data acquisition module is used for acquiring track circuit signal data of the monitored section; the data preprocessing module is used for identifying the section type of the monitored section and preprocessing the track circuit signal data according to a data preprocessing mode corresponding to the section type; the novel data prediction module is used for predicting the preprocessed track circuit signal data by adopting a trained OC-SVM prediction model and judging whether the track circuit signal data belongs to normal data or novel data according to a prediction result; the fault classification module is used for directly sending the track circuit signal data belonging to the normal data to the DNN prediction model for fault classification prediction; the training data set updating module is used for adding the track circuit signal data at the current moment meeting the preset conditions to a training data set and setting a new fault label for the training data set; the preset conditions include: each track circuit signal data collected in a certain continuous time period from the current moment is novel data; the model training module is used for training by using a training data set to obtain an OC-SVM prediction model and a DNN prediction model.
As one possible implementation, the section types include a no rear section, a no front section, a middle section, and a no front no rear section.
As an implementation manner, the data preprocessing module is specifically configured to:
if the monitored section is a section without a rear section, merging the track circuit signal data of the monitored section and the track circuit signal data of the section in front of the monitored section, and deleting the data item of which the characteristic value is not digital;
if the monitored section is a section without a front section, combining the track circuit signal data of the monitored section with the track circuit signal data of the section behind the monitored section, and deleting the data item of which the characteristic value is not digital;
if the monitored section is the middle section, merging the track circuit signal data of the monitored section with the track circuit signal data of the front and rear sections of the monitored section, and deleting the data item of which the characteristic value is not digital;
and if the monitored section is a section without a front part and a rear part, directly deleting the data item of which the characteristic value is not digital in the track circuit signal data.
As an implementation mode, the process of training the model training module to obtain the OC-SVM forecasting model comprises the following steps:
step A1: loading a training data set, wherein each track circuit signal data in the training data set is used as a sample, and each sample corresponds to a fault label;
step A2: mapping each sample to a high-dimensional feature space through a kernel function, and solving an optimal hypersphere in the feature space by using a formula (1) to realize the maximum separation of training data and a coordinate origin (namely novel data);
s.t.ω Τ φ(χ i )≥ρ-ξ i
ξ i ≥0,i＝1,2,……,l (1)
wherein, ω and ρ represent normal vector and offset of separating hyperplane in feature space, respectively; xi shape i A relaxation variable representing the ith sample, v is a parameter controlling the upper limit of the number of outliers and the lower limit of the number of all support vectors, l is the number of samples, and phi (χ) is a function mapping the samples x to a feature space;
step A3: obtaining a classification decision function shown in formula (2) by using a Lagrange multiplier method, wherein the classification decision function is an OC-SVM prediction model, and if f (x) is equal to 1, the classification decision function indicates that the sample x is normal data, and if f (x) is equal to-1, the classification decision function indicates that the sample x is novel data;
wherein, kappa (χ) i χ) represents a kernel function, and replaces inner product calculation in a feature space; alpha (alpha) ("alpha") i Is a sample x i The corresponding lagrangian coefficient.
As an implementation manner, the process of training the model training module to obtain the DNN prediction model includes:
step B1: loading a training data set, wherein each track circuit signal data in the training data set is used as a sample, and each sample corresponds to a fault label;
step B2: constructing a deep neural network framework, setting network parameters, and initializing a weight value w and a bias item b at random; the network parameters comprise the number of hidden layers, the number of nodes in each hidden layer and a learning rate;
step B3: setting a gradient optimization method, a hidden layer activation function and an output layer activation function;
step B4: and performing forward calculation, feedback calculation and parameter updating on all samples, and determining the optimal values of the weight value w and the bias item b so as to obtain the DNN prediction model.
It should be noted that the track circuit fault diagnosis device provided in the embodiment of the present invention is for implementing the above method embodiment, and the functions thereof may specifically refer to the above method embodiment, and are not described herein again.
Aiming at the problems that the current track circuit has various faults and high-quality label data is difficult to obtain, the ZPW-2000R track circuit fault diagnosis method and device based on OC-SVM and DNN provided by the invention utilize OC-SVM to perform single-classification recognition on 48 monitoring data acquired by the ZPW-2000R track circuit so as to collect signal data of different fault types; and then training the collected fault data by utilizing a DNN method so as to automatically classify the known fault data.
In addition, a large number of experiments are carried out on the method by using ZPW-2000R track circuit data, and the experimental results show that the fault diagnosis method provided by the invention can effectively and automatically collect novel fault data and can accurately detect the fault type.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A ZPW-2000R track circuit fault diagnosis method based on OC-SVM and DNN is characterized by comprising the following steps:
step 1: acquiring track circuit signal data of a monitored section at the current moment;
step 2: identifying the section type of the monitored section and carrying out data preprocessing on the track circuit signal data according to a data preprocessing mode corresponding to the section type;
and 3, step 3: predicting the preprocessed track circuit signal data by adopting a trained OC-SVM prediction model, and judging whether the track circuit signal data belongs to normal data or novel data according to a prediction result;
and 4, step 4: if the track circuit signal data belong to normal data, directly sending the track circuit signal data to a DNN prediction model for fault classification prediction;
and 5: if the track circuit signal data belong to novel data, continuing to detect the track circuit signal data of the monitored section at the next moment, if each piece of track circuit signal data collected in a certain continuous time period from the current moment is novel data, adding the track circuit signal data corresponding to the current moment into a training data set, and setting a new fault label for the training data set;
step 6: and training by using the new training data set to obtain a new OC-SVM prediction model and a new DNN prediction model so as to detect the signal data of the subsequent track circuit.
2. A ZPW-2000R track circuit fault diagnosis method based on OC-SVM and DNN according to claim 1, wherein the zone types include rear-less zone, front-less zone, middle zone and front-less rear zone.
3. The ZPW-2000R track circuit fault diagnosis method based on OC-SVM and DNN as claimed in claim 2, wherein step 2 specifically comprises:
if the monitored section is a section without a rear section, merging the track circuit signal data of the monitored section and the track circuit signal data of the section in front of the monitored section, and deleting the data item of which the characteristic value is not digital;
if the monitored section is a section without a front section, combining the track circuit signal data of the monitored section with the track circuit signal data of the section behind the monitored section, and deleting the data item of which the characteristic value is not digital;
if the monitored section is the middle section, merging the track circuit signal data of the monitored section with the track circuit signal data of the front and rear sections of the monitored section, and deleting the data item of which the characteristic value is not digital;
and if the monitored section is a section without a front part and a rear part, directly deleting the data item with the characteristic value which is not digital in the track circuit signal data.
4. The ZPW-2000R orbit circuit fault diagnosis method based on OC-SVM and DNN of claim 1, characterized in that the training process of the OC-SVM prediction model comprises:
step A1: loading a training data set, wherein each track circuit signal data in the training data set is used as a sample, and each sample corresponds to a fault label;
step A2: mapping each sample to a high-dimensional feature space through a kernel function, and solving an optimal hypersphere in the feature space by using a formula (1);
s.t.ω Τ φ(χ i )≥ρ-ξ i
ξ i ≥0,i＝1,2,……,l (1)
wherein, ω and ρ represent normal vector and offset of separating hyperplane in feature space, respectively; xi i A relaxation variable representing the ith sample, v is a parameter controlling the upper limit of the number of outliers and the lower limit of the number of all support vectors, l is the number of samples, and phi (χ) is a function mapping the samples x to a feature space;
step A3: obtaining a classification decision function shown in formula (2) by using a Lagrange multiplier method, wherein the classification decision function is an OC-SVM prediction model, and if f (x) is equal to 1, the classification decision function indicates that the sample x is normal data, and if f (x) is equal to-1, the classification decision function indicates that the sample x is novel data;
wherein, kappa (χ) i χ) represents a kernel function; alpha is alpha i Is a sample x i The corresponding lagrangian coefficient.
5. The ZPW-2000R orbit circuit fault diagnosis method based on OC-SVM and DNN of claim 1, the training process of the DNN prediction model comprising:
step B1: loading a training data set, wherein each track circuit signal data in the training data set is used as a sample, and each sample corresponds to a fault label;
step B2: constructing a deep neural network framework, setting network parameters, and initializing a weight value w and a bias item b at random; the network parameters comprise the number of hidden layers, the number of nodes in each hidden layer and a learning rate;
step B3: setting a gradient optimization method, a hidden layer activation function and an output layer activation function;
step B4: and performing forward calculation, feedback calculation and parameter updating on all samples, and determining the optimal values of the weight value w and the bias item b so as to obtain the DNN prediction model.
6. ZPW-2000R track circuit fault diagnosis device based on OC-SVM and DNN, characterized by comprising:
the data acquisition module is used for acquiring track circuit signal data of the monitored section;
the data preprocessing module is used for identifying the section type of the monitored section and preprocessing the track circuit signal data according to a data preprocessing mode corresponding to the section type;
the novel data prediction module is used for predicting the preprocessed track circuit signal data by adopting a trained OC-SVM prediction model and judging whether the track circuit signal data belongs to normal data or novel data according to a prediction result;
the fault classification module is used for directly sending the track circuit signal data belonging to the normal data to the DNN prediction model for fault classification prediction;
the training data set updating module is used for adding the track circuit signal data at the current moment meeting the preset conditions to a training data set and setting a new fault label for the training data set; the preset conditions include: each track circuit signal data collected in a certain continuous time period from the current moment is novel data;
and the model training module is used for training by using the training data set to obtain an OC-SVM prediction model and a DNN prediction model.
7. An OC-SVM and DNN based ZPW-2000R track circuit fault diagnosis device according to claim 6, wherein the zone types comprise a rear-less zone, a front-less zone, a middle zone and a front-less rear zone.
8. An OC-SVM and DNN based ZPW-2000R rail circuit fault diagnosis apparatus in accordance with claim 7, wherein said data preprocessing module is specifically configured to:
if the monitored section is a section without a rear section, merging the track circuit signal data of the monitored section and the track circuit signal data of the section in front of the monitored section, and deleting the data item of which the characteristic value is not digital;
if the monitored section is a section without a front section, merging the track circuit signal data of the monitored section and the track circuit signal data of the section behind the monitored section, and deleting the data item of which the characteristic value is not digital;
if the monitored section is the middle section, merging the track circuit signal data of the monitored section with the track circuit signal data of the front and rear sections of the monitored section, and deleting the data item of which the characteristic value is not digital;
and if the monitored section is a section without a front part and a rear part, directly deleting the data item of which the characteristic value is not digital in the track circuit signal data.
9. The ZPW-2000R orbit circuit fault diagnosis device based on OC-SVM and DNN of claim 6, wherein the model training module is used for training the process of obtaining the OC-SVM prediction model, and comprises the following steps:
step A1: loading a training data set, wherein each track circuit signal data in the training data set is used as a sample, and each sample corresponds to a fault label;
step A2: mapping each sample to a high-dimensional feature space through a kernel function, and solving an optimal hypersphere in the feature space by using a formula (1);
s.t.ω Τ φ(χ i )≥ρ-ξ i
ξ i ≥0,i＝1,2,……,l (1)
wherein, ω and ρ represent normal vector and offset of separating hyperplane in feature space, respectively; xi i A relaxation variable representing the ith sample, v is a parameter controlling the upper limit of the number of outliers and the lower limit of the number of total support vectors, l is the number of samples, and phi (χ) is a function mapping samples x to feature space；
Step A3: obtaining a classification decision function shown in formula (2) by using a Lagrange multiplier method, wherein the classification decision function is an OC-SVM prediction model, and if f (x) is equal to 1, the classification decision function indicates that the sample x is normal data, and if f (x) is equal to-1, the classification decision function indicates that the sample x is novel data;
wherein, kappa (χ) i χ) represents a kernel function; alpha (alpha) ("alpha") i Is a sample x i The corresponding lagrangian coefficient.
10. The ZPW-2000R orbit circuit fault diagnosis apparatus based on OC-SVM and DNN of claim 6, wherein the model training module is used for training the DNN prediction model by the steps of:
step B1: loading a training data set, wherein each track circuit signal data in the training data set is used as a sample, and each sample corresponds to a fault label;
step B2: constructing a deep neural network framework, setting network parameters, and initializing a weight value w and a bias item b at random; the network parameters comprise the number of hidden layers, the number of nodes in each hidden layer and a learning rate;
step B3: setting a gradient optimization method, a hidden layer activation function and an output layer activation function;
step B4: and performing forward calculation, feedback calculation and parameter updating on all samples, and determining the optimal values of the weight value w and the bias term b to obtain the DNN prediction model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210621985.4A CN115047313B (en) | 2022-06-02 | 2022-06-02 | ZPW-2000R orbit circuit fault diagnosis method and device based on OC-SVM and DNN |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210621985.4A CN115047313B (en) | 2022-06-02 | 2022-06-02 | ZPW-2000R orbit circuit fault diagnosis method and device based on OC-SVM and DNN |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115047313A true CN115047313A (en) | 2022-09-13 |
| CN115047313B CN115047313B (en) | 2024-04-26 |
Family
ID=83158856
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210621985.4A Active CN115047313B (en) | 2022-06-02 | 2022-06-02 | ZPW-2000R orbit circuit fault diagnosis method and device based on OC-SVM and DNN |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115047313B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116520817A (en) * | 2023-07-05 | 2023-08-01 | 贵州宏信达高新科技有限责任公司 | ETC system running state real-time monitoring system and method based on expressway |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5483170A (en) * | 1993-08-24 | 1996-01-09 | New Mexico State University Technology Transfer Corp. | Integrated circuit fault testing implementing voltage supply rail pulsing and corresponding instantaneous current response analysis |
| CN103714348A (en) * | 2014-01-09 | 2014-04-09 | 北京泰乐德信息技术有限公司 | Rail transit fault diagnosis method and system based on decision-making tree |
| CN103914735A (en) * | 2014-04-17 | 2014-07-09 | 北京泰乐德信息技术有限公司 | Failure recognition method and system based on neural network self-learning |
| CN110286668A (en) * | 2019-07-15 | 2019-09-27 | 广东毓秀科技有限公司 | A kind of rail friendship signal system VIM board faults prediction technique based on big data |
| CN110852365A (en) * | 2019-10-31 | 2020-02-28 | 北京交通大学 | ZPW-2000A type non-insulated rail circuit fault diagnosis method |
| CN111626416A (en) * | 2020-04-24 | 2020-09-04 | 黑龙江瑞兴科技股份有限公司 | Automatic rail circuit fault diagnosis method based on deep convolutional neural network |
| CN112949715A (en) * | 2013-12-31 | 2021-06-11 | 北京泰乐德信息技术有限公司 | SVM (support vector machine) -based rail transit fault diagnosis method |
| CN113139335A (en) * | 2021-04-09 | 2021-07-20 | 郑州宥新算法智能科技有限公司 | BP neural network-based track circuit fault intelligent diagnosis method |
| CN114462475A (en) * | 2021-12-21 | 2022-05-10 | 南京邮电大学 | Unsupervised machine abnormal sound detection method and unsupervised machine abnormal sound detection device based on single classification algorithm |
-
2022
- 2022-06-02 CN CN202210621985.4A patent/CN115047313B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5483170A (en) * | 1993-08-24 | 1996-01-09 | New Mexico State University Technology Transfer Corp. | Integrated circuit fault testing implementing voltage supply rail pulsing and corresponding instantaneous current response analysis |
| CN112949715A (en) * | 2013-12-31 | 2021-06-11 | 北京泰乐德信息技术有限公司 | SVM (support vector machine) -based rail transit fault diagnosis method |
| CN103714348A (en) * | 2014-01-09 | 2014-04-09 | 北京泰乐德信息技术有限公司 | Rail transit fault diagnosis method and system based on decision-making tree |
| CN103914735A (en) * | 2014-04-17 | 2014-07-09 | 北京泰乐德信息技术有限公司 | Failure recognition method and system based on neural network self-learning |
| CN110286668A (en) * | 2019-07-15 | 2019-09-27 | 广东毓秀科技有限公司 | A kind of rail friendship signal system VIM board faults prediction technique based on big data |
| CN110852365A (en) * | 2019-10-31 | 2020-02-28 | 北京交通大学 | ZPW-2000A type non-insulated rail circuit fault diagnosis method |
| CN111626416A (en) * | 2020-04-24 | 2020-09-04 | 黑龙江瑞兴科技股份有限公司 | Automatic rail circuit fault diagnosis method based on deep convolutional neural network |
| CN113139335A (en) * | 2021-04-09 | 2021-07-20 | 郑州宥新算法智能科技有限公司 | BP neural network-based track circuit fault intelligent diagnosis method |
| CN114462475A (en) * | 2021-12-21 | 2022-05-10 | 南京邮电大学 | Unsupervised machine abnormal sound detection method and unsupervised machine abnormal sound detection device based on single classification algorithm |
Non-Patent Citations (3)
| Title |
|---|
| 卢皎，等: "基于卷积神经网络的ZPW-2000R轨道电路智能故障诊断方法", 《工业工程》, vol. 24, no. 4, 31 August 2021 (2021-08-31), pages 128 - 134 * |
| 文武臣: "基于人工智能的ZPW-2000A轨道电路 故障诊断方法研究", 《铁路通信信号工程技术》, vol. 18, no. 10, 25 October 2021 (2021-10-25), pages 34 - 38 * |
| 谢旭旭，等: "基于深度学习的无绝缘轨道电路故障诊断研究", 《铁道学报》, vol. 42, no. 6, 30 June 2020 (2020-06-30), pages 79 - 85 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116520817A (en) * | 2023-07-05 | 2023-08-01 | 贵州宏信达高新科技有限责任公司 | ETC system running state real-time monitoring system and method based on expressway |
| CN116520817B (en) * | 2023-07-05 | 2023-08-29 | 贵州宏信达高新科技有限责任公司 | ETC system running state real-time monitoring system and method based on expressway |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115047313B (en) | 2024-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112784965B (en) | Large-scale multi-element time series data anomaly detection method oriented to cloud environment | |
| Liu et al. | Accurate prediction scheme of water quality in smart mariculture with deep Bi-S-SRU learning network | |
| CN104992223B (en) | Intensive Population size estimation method based on deep learning | |
| CN111178553A (en) | Industrial equipment health trend analysis method and system based on ARIMA and LSTM algorithms | |
| CN114266301A (en) | Intelligent power equipment fault prediction method based on graph convolution neural network | |
| CN112232316B (en) | Crowd gathering detection method and device, electronic equipment and storage medium | |
| Cao et al. | Multimodal process monitoring based on variational Bayesian PCA and Kullback-Leibler divergence between mixture models | |
| CN110046663A (en) | A kind of complex electromechanical systems fault critical state discrimination method | |
| CN115047313A (en) | ZPW-2000R track circuit fault diagnosis method and device based on OC-SVM and DNN | |
| CN113884807B (en) | Power distribution network fault prediction method based on random forest and multi-layer architecture clustering | |
| CN117077870B (en) | Water resource digital management method based on artificial intelligence | |
| CN117035013A (en) | Method for predicting dynamic network link by adopting impulse neural network | |
| Kolev et al. | ARFA: automated real-time flight data analysis using evolving clustering, classifiers and recursive density estimation | |
| CN115761647A (en) | Computer vision-based method and system for detecting motion trail of object in transformer substation | |
| CN110728768A (en) | Switch operation and maintenance method and system | |
| CN111988192A (en) | Train communication network fault diagnosis method based on machine learning | |
| CN114004258B (en) | Semi-supervised electrocardiographic abnormality detection method | |
| CN115796361A (en) | Wind speed interval prediction method and device for ground stage of overhead line engineering | |
| CN112185543A (en) | Construction method of medical induction data flow classification model | |
| CN112101244A (en) | Method for predicting optical fiber vibration signal danger level based on deep learning | |
| CN110837932A (en) | Thermal power prediction method of solar heat collection system based on DBN-GA model | |
| CN116415196A (en) | Intelligent industry-based data anomaly detection method, storage medium and terminal | |
| CN115496131B (en) | Equipment health state classification method based on multiple pre-training neural networks | |
| CN116563953B (en) | Bottom-up weak supervision time sequence action detection method, system, equipment and medium | |
| CN116481668A (en) | Deep learning-based optical fiber temperature measurement system danger level prediction method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |