CN113204227B - Cloud collaborative fault diagnosis system and method for layered modular engineering machinery - Google Patents
Cloud collaborative fault diagnosis system and method for layered modular engineering machinery Download PDFInfo
- Publication number
- CN113204227B CN113204227B CN202110454209.5A CN202110454209A CN113204227B CN 113204227 B CN113204227 B CN 113204227B CN 202110454209 A CN202110454209 A CN 202110454209A CN 113204227 B CN113204227 B CN 113204227B
- Authority
- CN
- China
- Prior art keywords
- module
- diagnosis
- fault
- data
- cloud
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0208—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the configuration of the monitoring system
- G05B23/0213—Modular or universal configuration of the monitoring system, e.g. monitoring system having modules that may be combined to build monitoring program; monitoring system that can be applied to legacy systems; adaptable monitoring system; using different communication protocols
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/24—Pc safety
- G05B2219/24065—Real time diagnostics
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Abstract
The invention discloses a layered modular engineering machinery cloud collaborative fault diagnosis system and a layered modular engineering machinery cloud collaborative fault diagnosis method, wherein the system comprises a vehicle-mounted diagnosis unit, a communication contact layer and a cloud diagnosis unit which are sequentially connected; the vehicle-mounted diagnosis unit collects fault data from a machine side and carries out fault diagnosis based on the collected fault data; the vehicle-mounted diagnosis unit can send the acquired fault data to the cloud diagnosis unit; and the cloud diagnosis unit carries out fault diagnosis based on the received fault data. The invention adopts the ideas of layered architecture and modularization to strip functional modules and hardware, improves the reusability, transportability, maintainability and expansibility of the fault diagnosis system, and further improves the overall encapsulation of the fault diagnosis system.
Description
Technical Field
The invention relates to the technical field of engineering machinery fault diagnosis, in particular to a layered modular engineering machinery cloud collaborative fault diagnosis system and method.
Background
At present, with the continuous development of economy and continuous improvement of infrastructure construction in China, the engineering machinery industry is in the key period of vigorous development. The engineering machinery has severe working environment and complex working terrain, particularly in some dangerous working areas, the intelligent equipment is urgently needed to replace manual driving, so that the control logic is gradually complicated, the maintenance cost is higher and higher, and along with the rise of an intelligent diagnosis system, the fuzzy diagnosis and expert diagnosis strategies are continuously perfected, and new requirements are provided for the engineering machinery fault diagnosis system.
Aiming at an engineering vehicle, on one hand, an accurate physical model cannot be established for managing and monitoring the vehicle condition of the engineering vehicle through mechanism analysis; on the other hand, with the improvement of the intelligent requirement, a large amount of equipment for sensing the surrounding environment is additionally arranged, and data reflecting the running state in the running process of the vehicle is stored, so that how to more effectively and more accurately mine a relation model behind mass data is an urgent problem to be solved.
At the present stage, because the controller products are of various types, the diagnostic logic is combined with the bottom port more tightly, once the controller or the message protocol is replaced, the diagnostic logic often needs to be rewritten, which greatly increases the labor intensity of developers and introduces logic errors again. Meanwhile, fault diagnosis depends on sensing of an external environment by a sensor, hardware needs to be replaced again if the sensor fails, parameter threshold values and range mapping often need to be written in a source program again, maintenance is not easy, and errors are easily introduced.
In addition, the existing technical scheme of the fault diagnosis system cannot separate functions from hardware, the whole encapsulation performance of the diagnosis system is insufficient, the coupling performance of functional software and hardware drive is strong, the openness is weak, the secondary development workload is large, and hardware modification usually brings about modification of a series of software code layers. Due to the adoption of the framework mode of the main subprogram, the functional module cannot be effectively stripped for independent development, so that the program is often independently developed and maintained by one person, and huge economic loss is brought to an enterprise once a person leaves the office. Because the algorithm is coupled with the algorithm, the algorithm cannot be stripped and upgraded independently.
Disclosure of Invention
Aiming at the problems, the invention provides a layered modular engineering machinery cloud collaborative fault diagnosis system and a layered modular engineering machinery cloud collaborative fault diagnosis method.
In order to achieve the technical purpose and achieve the technical effects, the invention is realized by the following technical scheme:
in a first aspect, the invention provides a cloud collaborative fault diagnosis system for a layered modular engineering machinery, which comprises a vehicle-mounted diagnosis unit, a communication connection layer and a cloud diagnosis unit which are sequentially connected;
the vehicle-mounted diagnosis unit collects fault data from the machine side and carries out fault diagnosis based on the collected fault data; the vehicle-mounted diagnosis unit also sends the acquired fault data to the cloud diagnosis unit;
and the cloud diagnosis unit carries out fault diagnosis based on the received fault data.
Optionally, the on-board diagnostics unit comprises:
the port driving module is used for reading machine side fault data in real time;
the port data acquisition module is connected with the port driving module;
the bus data acquisition module is connected with the port driving module;
the port diagnosis logic module is connected with the port data acquisition module and is used for carrying out port fault diagnosis based on fault data received from the port data acquisition module;
the bus diagnosis logic module is connected with the bus data acquisition module and is used for carrying out bus fault diagnosis based on fault data received from the bus data acquisition module;
and the local/remote switching module is respectively connected with the port data acquisition module and the bus data acquisition module and controls the port data acquisition module and the bus data acquisition module to output fault data to the port diagnosis logic module and the bus diagnosis logic module or output fault data to the cloud diagnosis unit.
Optionally, the communication contact layer includes a CAN interface and a wireless transceiving interface; the CAN interface is connected with the vehicle-mounted diagnosis unit; the wireless receiving and sending interface is in wireless connection with the cloud diagnosis unit.
Optionally, the cloud diagnosis unit includes:
the cloud data exchange module is connected with the communication contact layer and used for receiving fault data sent by the vehicle-mounted diagnosis unit;
the diagnosis modeling module is connected with the cloud data exchange module and used for training a diagnosis model to obtain a trained diagnosis model;
the parameter optimizing module is connected with the diagnosis modeling module and is used for optimizing parameters to obtain optimal parameters;
the storage module is respectively connected with the diagnosis modeling module and the parameter optimizing module and stores the trained diagnosis model and the trained optimal parameters;
and the diagnosis and identification module is respectively connected with the storage module and the cloud data exchange module, and when the diagnosis and identification module carries out fault diagnosis, the diagnosis and identification of the fault are carried out based on the trained diagnosis model and the optimal parameters called from the storage module, and the identified fault information is sent to the cloud data exchange module.
Optionally, the parameter optimizing module includes a particle swarm optimizing module, a genetic algorithm optimizing module, and a gradient descent method optimizing module, which are independently arranged, input ends of the three modules are respectively connected to the diagnosis modeling module, and output ends of the three modules are respectively connected to the storage module and are respectively used for parameter optimizing.
Optionally, the diagnostic modeling module includes a fuzzy neural network sub-module, a support vector machine sub-module and a decision tree sub-module which are independently arranged, input ends of the fuzzy neural network sub-module, the support vector machine sub-module and the decision tree sub-module are all connected with the cloud data exchange module, and data ends of the fuzzy neural network sub-module, the support vector machine sub-module and the decision tree sub-module are all connected with an input end of the parameter optimization module.
Optionally, the cloud diagnosis unit further includes a data preprocessing module, and the data preprocessing module is connected to the cloud data exchange module and is configured to preprocess the received fault data.
Optionally, the data preprocessing module comprises a filtering submodule, an invalid value eliminating submodule, a variable screening submodule and a principal component analysis submodule which are connected in sequence; the filtering submodule carries out filtering processing on the received fault data, the processed data are transmitted to an invalid value removing module to remove invalid values, then the processed data are transmitted to a variable screening submodule to screen related variables, and finally the data are processed by a principal component analysis module to further compress data dimensions.
Optionally, the cloud collaborative fault diagnosis system of the layered modular engineering machinery further comprises a display screen, wherein the display screen is respectively connected with the vehicle-mounted diagnosis unit and the cloud diagnosis unit and is used for displaying faults and controlling and setting parameters of the vehicle-mounted diagnosis unit and the cloud diagnosis unit.
In a second aspect, the invention provides a cloud collaborative fault diagnosis method for a layered modular engineering machinery, which includes the following steps.
Collecting fault data from a machine side by using a vehicle-mounted diagnosis unit, and performing fault diagnosis based on the collected fault data;
and/or the vehicle-mounted diagnosis unit is used for sending the collected fault data to the cloud terminal diagnosis unit, and the cloud terminal diagnosis unit carries out fault diagnosis based on the received fault data.
Compared with the prior art, the invention has the beneficial effects that:
the invention adopts the ideas of layered architecture and modularization to strip functional modules and hardware, improves the reusability, transportability, maintainability and expansibility of the fault diagnosis system, and further improves the overall encapsulation of the fault diagnosis system.
Drawings
In order that the present disclosure may be more readily and clearly understood, reference is now made to the following detailed description of the present disclosure taken in conjunction with the accompanying drawings, in which:
FIG. 1 is a schematic diagram of a fault diagnosis system according to an embodiment of the present invention;
FIG. 2 is a schematic structural diagram of a vehicle-mounted fault diagnosis unit according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a vehicle data collection system according to an embodiment of the present invention;
FIG. 4 is a block diagram of a remote fault diagnosis module according to an embodiment of the present invention;
fig. 5 is a schematic flow chart of a fault diagnosis system according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the scope of the invention.
The following detailed description of the principles of the invention is provided in connection with the accompanying drawings.
Example 1
The embodiment of the invention provides a cloud collaborative fault diagnosis system for a layered modular engineering machinery, which comprises a vehicle-mounted diagnosis unit, a communication layer, a cloud diagnosis unit and a display screen, wherein the vehicle-mounted diagnosis unit, the communication layer, the cloud diagnosis unit and the display screen are sequentially connected;
the vehicle-mounted diagnosis unit collects fault data from a machine side and carries out fault diagnosis based on the collected fault data; the vehicle-mounted diagnosis unit also sends the acquired fault data to the cloud diagnosis unit;
the cloud diagnosis unit carries out fault diagnosis based on the received fault data;
the display screen is respectively connected with the vehicle-mounted diagnosis unit and the cloud diagnosis unit and used for displaying faults and controlling and setting parameters of the vehicle-mounted diagnosis unit and the cloud diagnosis unit.
Therefore, the cloud collaborative fault diagnosis system of the layered modular engineering machinery in the embodiment of the invention can integrate the computing power of the cloud into the fault diagnosis system, fully utilize the computing power of high computation of the cloud, and quickly realize fault diagnosis and identification on the premise of ensuring the precision. The system comprises a cloud diagnosis system, a communication layer and a vehicle-mounted diagnosis system from top to bottom. The vehicle-mounted fault diagnosis system achieves the effect of separating from bottom hardware by separating the vehicle-mounted data acquisition module from the driving module, only the diagnosis algorithm logic at the top layer needs to be concerned at the vehicle-mounted controller end, and the portability and algorithm reusability of the vehicle-mounted end diagnosis logic are greatly improved; the communication connection layer mainly comprises wireless communication equipment, is responsible for establishing communication between a remote upper computer and a local lower computer and consists of a CAN-to-4G module; the cloud fault diagnosis system only exchanges data with the cloud data exchange module and follows a corresponding data exchange protocol, other modules are completely independent of hardware design and can be integrally transplanted, and the system moves the core algorithm to the cloud host machine, so that the system is faster, more efficient and safer compared with local calculation.
As shown in fig. 2, in a specific implementation manner of the embodiment of the present invention, the on-board diagnosis unit includes:
the port driving module is used for reading machine side fault data in real time; in a specific implementation process, the machine side fault data comprises engine parameters, hydraulic system parameters, electrical system parameters, working device parameters and the like.
The port data acquisition module is connected with the port driving module; the bus data acquisition module is connected with the port driving module;
the port diagnosis logic module is connected with the port data acquisition module and is used for carrying out port fault diagnosis based on fault data received from the port data acquisition module;
the bus diagnosis logic module is connected with the bus data acquisition module and is used for carrying out bus fault diagnosis based on fault data received from the bus data acquisition module;
and the local/remote switching module is respectively connected with the port data acquisition module and the bus data acquisition module and controls the port data acquisition module and the bus data acquisition module to output fault data to the port diagnosis logic module and the bus diagnosis logic module or output fault data to the cloud diagnosis unit, namely the local/remote switching module is responsible for switching between the local diagnosis logic and the remote diagnosis logic, and the local/remote switching module needs to press a switching button in the vehicle-mounted display screen when being triggered. In a specific implementation process, the types of faults that the port diagnosis logic module and the bus diagnosis logic module can mainly identify include: the physical value overrun fault, the bus communication fault, the port short-circuit to the ground, the power short-circuit and other faults can only be identified, and the faults caused by nonlinearity and multiple reasons can not be effectively identified by means of the remote cloud diagnosis unit.
In a specific implementation manner of the embodiment of the present invention, as shown in fig. 3, the communication layer includes a CAN interface and a wireless transceiving interface; the CAN interface is connected with the vehicle-mounted diagnosis unit (namely a vehicle-mounted main controller in figure 3); the wireless receiving and sending interface is in wireless connection with the cloud diagnosis unit. In order to improve the wireless transceiving efficiency, the wireless transceiving interface may select a 4G transceiving interface.
In a specific implementation manner of the embodiment of the invention, the cloud diagnosis unit is designed in a modularized manner, so that independent updating of an algorithm module is facilitated, a cloud data exchange module is utilized to realize separation of a cloud algorithm and a vehicle-mounted algorithm, and the transportability of the cloud diagnosis unit is improved; as shown in fig. 4, the cloud diagnosis unit includes:
the cloud data exchange module is connected with the communication connection layer and is used for receiving fault data sent by the vehicle-mounted diagnosis unit, namely the cloud data exchange module is mainly used for carrying out data interaction with a machine side and comprises original data interacted with a machine side controller, a diagnosis result, remote diagnosis parameters and the like (selection of an algorithm and selection of a parameter optimization strategy), and as an original data set comprises data noise and irrelevant variables, the data set needs to be further processed in the data preprocessing module;
the data preprocessing module is respectively connected with the cloud data exchange module and the diagnosis modeling module and is used for preprocessing the received fault data, and the data preprocessing module is in charge of processing the original data uploaded from the cloud data exchange module and sending the processed data set to the diagnosis modeling module for modeling;
the diagnosis modeling module is connected with the data preprocessing module and used for training a diagnosis model to obtain a trained diagnosis model;
the parameter optimizing module is connected with the diagnosis modeling module and is used for optimizing parameters to obtain optimal parameters; when the method is used, only parameter configuration is needed locally, and the corresponding optimization algorithm is selected for optimization, so that parameter optimization is greatly facilitated, the accuracy of an algorithm model is ensured, and meanwhile, the phenomenon of overfitting can be effectively prevented;
the storage module is respectively connected with the diagnosis modeling module and the parameter optimizing module and stores the trained diagnosis model and the optimized parameters;
and the diagnosis and recognition module is respectively connected with the storage module and the cloud data exchange module, when the diagnosis and recognition module carries out fault diagnosis, the diagnosis and recognition of faults are carried out based on the trained diagnosis model and the optimal parameters called from the storage module, model training is not required to be carried out again, the fault diagnosis process is greatly accelerated, and the recognized fault information is sent to the cloud data exchange module.
As shown in fig. 4, in a specific implementation manner of the embodiment of the present invention, the data preprocessing module includes a filtering sub-module, an invalid value eliminating sub-module, a variable screening sub-module, and a principal component analysis sub-module, which are connected in sequence; the filtering submodule carries out filtering processing on the received fault data, the processed data are transmitted to an invalid value removing module to remove invalid values, then the processed data are transmitted to a variable screening submodule to screen related variables, and finally the data are processed by a principal component analysis module to further compress data dimensions. Specifically, the method comprises the following steps: 1) the filtering submodule is responsible for filtering the original data set, 2) the invalid value eliminating submodule is responsible for eliminating the invalid data set in the filtered module, 3) the variable screening submodule is responsible for selecting related variables according to a mutual information correlation criterion, finding out the variables related to the fault and outputting the variables as a related variable set of the fault, 4) the principal component analysis submodule is responsible for reducing the dimension of the selected related variable set, screening out related principal components, eliminating the unrelated variables, reducing the modeling complexity and improving the speed of classification and identification.
As shown in fig. 4, in a specific implementation manner of the embodiment of the present invention, the diagnosis modeling module includes a fuzzy neural network sub-module, a support vector machine sub-module, and a decision tree sub-module, which are independently arranged, input ends of the fuzzy neural network sub-module, the support vector machine sub-module, and the decision tree sub-module are all connected to the cloud data exchange module, and data ends of the fuzzy neural network sub-module, the support vector machine sub-module, and the decision tree sub-module are all connected to an input end of the parameter optimization module. The fuzzy neural network submodule combines the advantages of a fuzzy algorithm and a neural network algorithm, introduces a fuzzy rule on the premise of ensuring the modeling speed, and avoids the defect that the traditional neural network algorithm cannot process and describe fuzzy information, so that the current fault type can be distinguished quickly and accurately.
As shown in fig. 4, the sub-module of the algorithm contains a large number of parameters, and if the parameters are not accurately selected, the accuracy of the model will be greatly affected. Therefore, in a specific implementation manner of the embodiment of the present invention, the parameter optimization module includes a particle swarm optimization module, a genetic algorithm optimization module, and a gradient descent method optimization module that are independently arranged, input ends of the three modules are respectively connected to the diagnosis modeling module, output ends of the three modules are respectively connected to the storage module, and are respectively used for performing parameter optimization, and parameter configuration of the particle swarm optimization module, the genetic algorithm optimization module, and the gradient descent method optimization module can be performed by a remote diagnosis parameter configuration page of a display to adjust precision and speed of optimization, thereby preventing over-fitting or under-fitting.
In summary, as shown in fig. 5, the working process of the fault diagnosis system in the embodiment of the present invention specifically includes:
the method comprises the steps of firstly, reading peripheral data (namely machine side fault data) in real time by using a port driving module, and sending the data to a vehicle-mounted data acquisition module for processing (including range conversion and message analysis). Meanwhile, an operator can control the switching of the local/remote diagnosis system on the display screen through a button, and the local/remote diagnosis system is switched to the vehicle-mounted diagnosis mode by default. In the vehicle-mounted diagnosis mode, the types of fault diagnosis which can be realized are relatively limited, and the method mainly aims at faults which have single occurrence reasons, clear formation mechanisms and strong feature uniqueness during occurrence.
The vehicle-mounted fault unit transmits data output by the vehicle-mounted data acquisition module to a corresponding vehicle-mounted fault diagnosis module according to the occurrence type of the fault for fault diagnosis, and the vehicle-mounted fault unit mainly comprises: a port diagnosis module and a bus diagnosis module.
After switching to the high in the clouds diagnosis mode, the high in the clouds data exchange module can acquire data from the machine side through the communication tie layer on the one hand, can transmit the raw data who acquires for data preprocessing module on the one hand and carry out the preliminary treatment of data, specifically includes: filtering, removing invalid values, screening related variables and analyzing principal components. Firstly, the filtering submodule carries out filtering processing, the processed data are transmitted to the invalid value eliminating submodule for eliminating the invalid value, then the processed data are transmitted to the variable screening submodule for screening related variables, and finally the data are further compressed through the processing of the principal component analysis submodule to prepare for subsequent modeling. The diagnosis modeling module comprises three common fault recognition algorithms, the fuzzy neural network module is selected by default to carry out diagnosis modeling, and if the diagnosis modeling needs to be switched, the diagnosis modeling can be carried out on a remote diagnosis parameter configuration interface of the display screen, so that the system has great flexibility, manual switching can be carried out when one diagnosis method has a poor effect, and the accuracy of fault recognition can be guaranteed to the maximum extent. Meanwhile, in order to ensure the optimization of the model parameters, a parameter optimization strategy can be selected through a remote diagnosis parameter configuration interface of a display screen, and overfitting or underfitting of the model due to the fact that the parameters do not meet the requirements is prevented.
The trained diagnosis model and the optimal parameters are stored in the storage module, when the diagnosis and recognition module carries out fault diagnosis, the diagnosis and recognition module needs to call from the storage module, then fault diagnosis and recognition are carried out based on the trained model, and after the fault recognition, fault information is sent to the cloud-end data exchange module, and then the fault display interface which is transmitted to the display screen is used for displaying the fault. Similarly, the vehicle-mounted diagnosis unit can also transmit the fault to the display screen for displaying.
Example 2
The embodiment of the invention provides a cloud collaborative fault diagnosis method for a layered modular engineering machinery, which comprises the following steps.
Collecting fault data from a machine side by using a vehicle-mounted diagnosis unit, and performing fault diagnosis based on the collected fault data;
and/or the vehicle-mounted diagnosis unit is used for sending the collected fault data to the cloud terminal diagnosis unit, and the cloud terminal diagnosis unit carries out fault diagnosis based on the received fault data.
As shown in fig. 2, in a specific implementation manner of the embodiment of the present invention, the on-board diagnosis unit includes:
the port driving module is used for reading machine side fault data in real time; in a specific implementation process, the machine side fault data comprises engine parameters, hydraulic system parameters, electrical system parameters, working device parameters and the like.
The port data acquisition module is connected with the port driving module; the bus data acquisition module is connected with the port driving module;
the port diagnosis logic module is connected with the port data acquisition module and is used for carrying out port fault diagnosis based on fault data received from the port data acquisition module;
the bus diagnosis logic module is connected with the bus data acquisition module and is used for carrying out bus fault diagnosis based on fault data received from the bus data acquisition module;
and the local/remote switching module is respectively connected with the port data acquisition module and the bus data acquisition module and controls the port data acquisition module and the bus data acquisition module to output fault data to the port diagnosis logic module and the bus diagnosis logic module or output fault data to the cloud diagnosis unit, namely the local/remote switching module is responsible for switching between the local diagnosis logic and the remote diagnosis logic, and the local/remote switching module needs to press a switching button in the vehicle-mounted display screen when being triggered. In a specific implementation process, the types of faults that the port diagnostic logic module and the bus diagnostic logic module can mainly identify include: the physical value overrun fault, the bus communication fault, the port short-circuit to the ground, the power short-circuit and other faults can only be identified, and the faults caused by nonlinearity and multiple reasons can not be effectively identified by means of the remote cloud diagnosis unit.
In a specific implementation manner of the embodiment of the present invention, as shown in fig. 3, the communication layer includes a CAN interface and a wireless transceiving interface; the CAN interface is connected with the vehicle-mounted diagnosis unit; the wireless receiving and sending interface is in wireless connection with the cloud diagnosis unit. In order to improve the wireless transceiving efficiency, the wireless transceiving interface may select a 4G transceiving interface.
In a specific implementation manner of the embodiment of the invention, the cloud diagnosis unit is designed in a modularized manner, so that independent updating of an algorithm module is facilitated, a cloud algorithm is separated from a vehicle-mounted algorithm by using a cloud data exchange module, and the transportability of the cloud diagnosis unit is improved; as shown in fig. 4, the cloud diagnosis unit includes:
the cloud data exchange module is connected with the communication layer and is used for receiving fault data sent by the vehicle-mounted diagnosis unit, namely the cloud data exchange module is mainly used for carrying out data interaction with a machine side and comprises original data interacted with a machine side controller, a diagnosis result, remote diagnosis parameters and the like (selection of an algorithm and selection of a parameter optimization strategy), and as an original data set contains data noise and irrelevant variables, the data set needs to be further processed in the data preprocessing module;
the data preprocessing module is respectively connected with the cloud data exchange module and the diagnosis modeling module and is used for preprocessing the received fault data, and the data preprocessing module is in charge of processing the original data uploaded from the cloud data exchange module and sending the processed data set to the diagnosis modeling module for modeling;
the diagnosis modeling module is connected with the data preprocessing module and is used for training a diagnosis model to obtain a trained diagnosis model;
the parameter optimizing module is connected with the diagnosis modeling module and used for optimizing parameters to obtain optimal parameters; when the method is used, only parameter configuration is needed locally, and the corresponding optimization algorithm is selected for optimization, so that parameter optimization is greatly facilitated, the accuracy of an algorithm model is ensured, and meanwhile, the phenomenon of overfitting can be effectively prevented;
the storage module is respectively connected with the diagnosis modeling module and the parameter optimizing module and stores the trained diagnosis model and the optimized parameters;
and the diagnosis and recognition module is respectively connected with the storage module and the cloud data exchange module, when the diagnosis and recognition module carries out fault diagnosis, the diagnosis and recognition of faults are carried out based on the trained diagnosis model and the optimal parameters called from the storage module, model training is not required to be carried out again, the fault diagnosis process is greatly accelerated, and the recognized fault information is sent to the cloud data exchange module.
As shown in fig. 4, in a specific implementation manner of the embodiment of the present invention, the data preprocessing module includes a filtering sub-module, an invalid value eliminating sub-module, a variable screening sub-module, and a principal component analysis sub-module, which are connected in sequence; the filtering submodule carries out filtering processing on the received fault data, the processed data are transmitted to an invalid value removing module to remove invalid values, then the processed data are transmitted to a variable screening submodule to screen related variables, and finally the data are processed by a principal component analysis module to further compress data dimensions. Specifically, the method comprises the following steps: 1) the filtering submodule is responsible for filtering the original data set, 2) the invalid value eliminating submodule is responsible for eliminating the invalid data set in the filtered module, 3) the variable screening submodule is responsible for selecting related variables according to a mutual information correlation criterion, finding out the variables related to the fault and outputting the variables as a related variable set of the fault, 4) the principal component analysis submodule is responsible for reducing the dimension of the selected related variable set, screening out related principal components, eliminating the unrelated variables, reducing the modeling complexity and improving the speed of classification and identification.
As shown in fig. 4, in a specific implementation manner of the embodiment of the present invention, the diagnosis modeling module includes a fuzzy neural network sub-module, a support vector machine sub-module, and a decision tree sub-module, which are independently arranged, input ends of the fuzzy neural network sub-module, the support vector machine sub-module, and the decision tree sub-module are all connected to the cloud data exchange module, and data ends of the fuzzy neural network sub-module, the support vector machine sub-module, and the decision tree sub-module are all connected to an input end of the parameter optimization module. The fuzzy neural network submodule combines the advantages of a fuzzy algorithm and a neural network algorithm, introduces a fuzzy rule on the premise of ensuring the modeling speed, and avoids the defect that the traditional neural network algorithm cannot process and describe fuzzy information, so that the current fault type can be distinguished quickly and accurately.
As shown in fig. 4, the sub-module of the algorithm contains a large number of parameters, and if the parameters are not accurately selected, the accuracy of the model will be greatly affected. Therefore, in a specific implementation manner of the embodiment of the present invention, the parameter optimization module includes a particle swarm optimization module, a genetic algorithm optimization module, and a gradient descent method optimization module that are independently arranged, input ends of the three modules are respectively connected to the diagnosis modeling module, output ends of the three modules are respectively connected to the storage module, and are respectively used for performing parameter optimization, and parameter configuration of the particle swarm optimization module, the genetic algorithm optimization module, and the gradient descent method optimization module can be performed by a remote diagnosis parameter configuration page of a display to adjust precision and speed of optimization, thereby preventing over-fitting or under-fitting.
In summary, as shown in fig. 5, the working process of the fault diagnosis system in the embodiment of the present invention specifically includes:
the method comprises the steps of firstly, reading peripheral data (namely machine side fault data) in real time by using a port driving module, and sending the data to a vehicle-mounted data acquisition module for processing (including range conversion and message analysis). Meanwhile, an operator can control the switching of the local/remote diagnosis system on the display screen through a button, and the local/remote diagnosis system is switched to the vehicle-mounted diagnosis mode by default. In the vehicle-mounted diagnosis mode, the types of fault diagnosis which can be realized are relatively limited, and the method mainly aims at faults which have single occurrence reasons, clear formation mechanism and strong feature uniqueness during occurrence.
The vehicle-mounted fault unit transmits data output by the vehicle-mounted data acquisition module to a corresponding vehicle-mounted fault diagnosis module according to the occurrence type of the fault for fault diagnosis, and the vehicle-mounted fault unit mainly comprises: a port diagnosis module and a bus diagnosis module.
After switching to the high in the clouds diagnosis mode, the high in the clouds data exchange module can acquire data from the machine side through the communication tie layer on the one hand, can transmit the raw data who acquires for data preprocessing module on the one hand and carry out the preliminary treatment of data, specifically includes: filtering, removing invalid values, screening related variables and analyzing principal components. Firstly, the filtering submodule carries out filtering processing, the processed data are transmitted to the invalid value eliminating submodule for eliminating the invalid value, then the processed data are transmitted to the variable screening submodule for screening related variables, and finally the data are processed by the principal component analysis submodule to further compress the data dimension so as to prepare for subsequent modeling. The diagnosis modeling module comprises three common fault recognition algorithms, the fuzzy neural network module is selected by default to carry out diagnosis modeling, and if the diagnosis modeling module needs to be switched, the diagnosis modeling module can be configured on a remote diagnosis parameter configuration interface of a display screen, so that the system has great flexibility, manual switching can be carried out when one of the diagnosis methods is poor in effect, and the accuracy of fault recognition can be guaranteed to the maximum extent. Meanwhile, in order to ensure the optimization of the model parameters, a parameter optimizing strategy can be selected through a remote diagnosis parameter configuration interface of a display screen, and overfitting or underfitting of the model due to the fact that the parameters do not meet requirements is prevented.
The trained diagnosis model and the optimal parameters are stored in the storage module, when the diagnosis and recognition module carries out fault diagnosis, the diagnosis and recognition module needs to call from the storage module, then fault diagnosis and recognition are carried out based on the trained model, and after the fault recognition, fault information is sent to the cloud-end data exchange module, and then the fault display interface which is transmitted to the display screen is used for displaying the fault. Similarly, the vehicle-mounted diagnosis unit can also transmit the fault display to the display screen.
The foregoing shows and describes the general principles and broad features of the present invention and advantages thereof. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention, which fall within the scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (8)
1. A cloud collaborative fault diagnosis system of a layered modular engineering machinery is characterized by comprising a vehicle-mounted diagnosis unit, a communication layer and a cloud diagnosis unit which are sequentially connected;
the vehicle-mounted diagnosis unit collects fault data from a machine side and carries out fault diagnosis based on the collected fault data; the vehicle-mounted diagnosis unit also sends the acquired fault data to the cloud diagnosis unit;
the cloud diagnosis unit carries out fault diagnosis based on the received fault data;
the on-board diagnostic unit includes:
the port driving module is used for reading machine side fault data in real time;
the port data acquisition module is connected with the port driving module;
the bus data acquisition module is connected with the port driving module;
the port diagnosis logic module is connected with the port data acquisition module and is used for carrying out port fault diagnosis based on fault data received from the port data acquisition module;
the bus diagnosis logic module is connected with the bus data acquisition module and is used for carrying out bus fault diagnosis based on fault data received from the bus data acquisition module;
the local/remote switching module is respectively connected with the port data acquisition module and the bus data acquisition module and controls the port data acquisition module and the bus data acquisition module to output fault data to the port diagnosis logic module and the bus diagnosis logic module or output fault data to the cloud diagnosis unit;
the cloud diagnosis unit comprises:
the cloud data exchange module is connected with the communication contact layer and used for receiving fault data sent by the vehicle-mounted diagnosis unit;
the diagnosis modeling module is connected with the cloud data exchange module and used for training a diagnosis model to obtain a trained diagnosis model;
the parameter optimizing module is connected with the diagnosis modeling module and is used for optimizing parameters to obtain optimal parameters;
the storage module is respectively connected with the diagnosis modeling module and the parameter optimizing module and stores the trained diagnosis model and the optimized parameters;
the diagnosis and recognition module is respectively connected with the storage module and the cloud data exchange module, when the diagnosis and recognition module carries out fault diagnosis, diagnosis and recognition of faults are carried out on the basis of the trained diagnosis model and the optimal parameters called from the storage module, and recognized fault information is sent to the cloud data exchange module;
the vehicle-mounted diagnosis unit separates the port data acquisition module, the bus data acquisition module and the port driving module to achieve the effect of separating from bottom hardware;
and only the cloud data exchange module in the cloud diagnosis unit exchanges data with the machine side, and follows a corresponding data exchange protocol, and other modules are completely independent of hardware design.
2. The cloud collaborative fault diagnosis system of a layered modular construction machine according to claim 1, wherein: the communication contact layer comprises a CAN interface and a wireless transceiving interface; the CAN interface is connected with the vehicle-mounted diagnosis unit; the wireless receiving and sending interface is in wireless connection with the cloud diagnosis unit.
3. The cloud collaborative fault diagnosis system for the layered modular engineering machinery of claim 1, wherein the parameter optimization module comprises a particle swarm optimization module, a genetic algorithm optimization module and a gradient descent method optimization module which are independently arranged, input ends of the three modules are respectively connected with the diagnosis modeling module, and output ends of the three modules are respectively connected with the storage module and are respectively used for parameter optimization.
4. The cloud collaborative fault diagnosis system for layered modular construction machinery according to claim 1, wherein: the diagnosis modeling module comprises a fuzzy neural network submodule, a support vector machine submodule and a decision tree submodule which are independently arranged, the input ends of the fuzzy neural network submodule, the support vector machine submodule and the decision tree submodule are all connected with the cloud data exchange module, and the data ends of the fuzzy neural network submodule, the support vector machine submodule and the decision tree submodule are all connected with the input end of the parameter optimizing module.
5. The cloud collaborative fault diagnosis system of a layered modular construction machine according to claim 1, wherein: the cloud diagnosis unit further comprises a data preprocessing module, the data preprocessing module is connected with the cloud data exchange module, and the data preprocessing module is used for preprocessing the received fault data.
6. The cloud collaborative fault diagnosis system of a layered modular construction machine according to claim 1, wherein: the data preprocessing module comprises a filtering submodule, an invalid value eliminating submodule, a variable screening submodule and a principal component analysis submodule which are sequentially connected; the filtering submodule carries out filtering processing on the received fault data, the processed data are transmitted to an invalid value removing module to remove invalid values, then the processed data are transmitted to a variable screening submodule to screen related variables, and finally the data are processed by a principal component analysis module to further compress data dimensions.
7. The cloud collaborative fault diagnosis system of a layered modular construction machine according to claim 1, wherein: the cloud collaborative fault diagnosis system of the layered modular engineering machinery further comprises a display screen, wherein the display screen is respectively connected with the vehicle-mounted diagnosis unit and the cloud diagnosis unit and is used for displaying faults and controlling and setting parameters of the vehicle-mounted diagnosis unit and the cloud diagnosis unit.
8. A cloud collaborative fault diagnosis method for a layered modular engineering machinery is characterized by comprising the following steps:
collecting fault data from a machine side by using a vehicle-mounted diagnosis unit, and performing fault diagnosis based on the collected fault data;
and/or sending the acquired fault data to a cloud diagnosis unit by using the vehicle-mounted diagnosis unit, and carrying out fault diagnosis by the cloud diagnosis unit based on the received fault data;
the on-board diagnostic unit includes:
the port driving module is used for reading machine side fault data in real time;
the port data acquisition module is connected with the port driving module;
the bus data acquisition module is connected with the port driving module;
the port diagnosis logic module is connected with the port data acquisition module and is used for carrying out port fault diagnosis based on fault data received from the port data acquisition module;
the bus diagnosis logic module is connected with the bus data acquisition module and is used for carrying out bus fault diagnosis based on fault data received from the bus data acquisition module;
the local/remote switching module is respectively connected with the port data acquisition module and the bus data acquisition module and controls the port data acquisition module and the bus data acquisition module to output fault data to the port diagnosis logic module and the bus diagnosis logic module or output fault data to the cloud diagnosis unit;
the cloud diagnosis unit comprises:
the cloud data exchange module is connected with the communication contact layer and used for receiving the fault data sent by the vehicle-mounted diagnosis unit;
the diagnosis modeling module is connected with the cloud data exchange module and used for training a diagnosis model to obtain a trained diagnosis model;
the parameter optimizing module is connected with the diagnosis modeling module and is used for optimizing parameters to obtain optimal parameters;
the storage module is respectively connected with the diagnosis modeling module and the parameter optimizing module and stores the trained diagnosis model and the optimized parameters;
the diagnosis and recognition module is respectively connected with the storage module and the cloud data exchange module, when the diagnosis and recognition module carries out fault diagnosis, the diagnosis and recognition of faults are carried out on the basis of the trained diagnosis model and the optimal parameters called from the storage module, and the recognized fault information is sent to the cloud data exchange module;
the vehicle-mounted diagnosis unit separates the port data acquisition module, the bus data acquisition module and the port drive module to achieve the effect of separating from bottom hardware;
in the cloud diagnosis unit, only the cloud data exchange module exchanges data with the machine side, the corresponding data exchange protocol is followed, and other modules are completely independent of hardware design.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110454209.5A CN113204227B (en) | 2021-04-26 | 2021-04-26 | Cloud collaborative fault diagnosis system and method for layered modular engineering machinery |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110454209.5A CN113204227B (en) | 2021-04-26 | 2021-04-26 | Cloud collaborative fault diagnosis system and method for layered modular engineering machinery |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113204227A CN113204227A (en) | 2021-08-03 |
| CN113204227B true CN113204227B (en) | 2023-04-18 |
Family
ID=77028966
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110454209.5A Active CN113204227B (en) | 2021-04-26 | 2021-04-26 | Cloud collaborative fault diagnosis system and method for layered modular engineering machinery |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113204227B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103455022A (en) * | 2012-06-01 | 2013-12-18 | 北汽福田汽车股份有限公司 | Method and device for diagnosing vehicle-mounted electronic control devices |
| CN110196365A (en) * | 2019-05-31 | 2019-09-03 | 中山职业技术学院 | A kind of vehicle power drive system method for diagnosing faults |
| CN111325257A (en) * | 2020-02-14 | 2020-06-23 | 逸驾智能科技有限公司 | Method and device for analyzing vehicle faults |
Family Cites Families (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7379799B2 (en) * | 2005-06-29 | 2008-05-27 | General Electric Company | Method and system for hierarchical fault classification and diagnosis in large systems |
| CN101707503B (en) * | 2009-07-03 | 2014-10-22 | 中兴通讯股份有限公司 | Embedded method and device for controlling automatic positioning of channel communication failure |
| CN102594929A (en) * | 2012-04-09 | 2012-07-18 | 天津一汽夏利汽车股份有限公司 | Vehicle-mounted information center control unit |
| CN202584724U (en) * | 2012-04-09 | 2012-12-05 | 天津一汽夏利汽车股份有限公司 | Vehicle information center control unit |
| CN104281130B (en) * | 2014-09-22 | 2017-05-10 | 国家电网公司 | Hydroelectric equipment monitoring and fault diagnosis system based on big data technology |
| CN104573355B (en) * | 2014-12-30 | 2017-11-07 | 北华大学 | A kind of Diagnosis Method of Transformer Faults of the use parameter optimization SVMs based on photocaustic spectroscopy |
| CN205970918U (en) * | 2016-09-14 | 2017-02-22 | 武汉电动汽车技术开发有限公司 | Electric automobile monitoring device |
| CN106406296B (en) * | 2016-12-14 | 2019-04-23 | 东北大学 | It is a kind of based on vehicle-mounted and cloud train fault diagnostic system and method |
| US10316812B2 (en) * | 2017-04-20 | 2019-06-11 | Gm Global Technology Operations Llc. | Hierarchical fault diagnosis and prognosis of a system |
| CN107272653B (en) * | 2017-07-20 | 2020-03-17 | 南京航空航天大学 | Fault diagnosis method for flight control system |
| CN109034260B (en) * | 2018-08-07 | 2021-12-03 | 东南大学 | Desulfurization tower oxidation fan fault diagnosis system and method based on statistical principle and intelligent optimization |
| CN110597229A (en) * | 2019-09-24 | 2019-12-20 | 中国第一汽车股份有限公司 | Vehicle diagnosis mutual exclusion method and device, vehicle and storage medium |
| CN110888418B (en) * | 2019-12-23 | 2021-08-10 | 深圳市元征科技股份有限公司 | Vehicle diagnosis method, device, equipment and readable storage medium |
| CN111147342B (en) * | 2020-04-03 | 2020-09-08 | 北京全路通信信号研究设计院集团有限公司 | MVB bus fault diagnosis method and system based on communication chip |
| CN111770006A (en) * | 2020-07-29 | 2020-10-13 | 上海仁童电子科技有限公司 | Communication quality evaluation method and device of multifunctional vehicle bus network |
-
2021
- 2021-04-26 CN CN202110454209.5A patent/CN113204227B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103455022A (en) * | 2012-06-01 | 2013-12-18 | 北汽福田汽车股份有限公司 | Method and device for diagnosing vehicle-mounted electronic control devices |
| CN110196365A (en) * | 2019-05-31 | 2019-09-03 | 中山职业技术学院 | A kind of vehicle power drive system method for diagnosing faults |
| CN111325257A (en) * | 2020-02-14 | 2020-06-23 | 逸驾智能科技有限公司 | Method and device for analyzing vehicle faults |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113204227A (en) | 2021-08-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109635677B (en) | Compound fault diagnosis method and device based on multi-label classification convolutional neural network | |
| CN110769032B (en) | System and method for rapidly detecting and maintaining power grid equipment fault | |
| CN110602176A (en) | Method for controlling intelligent street lamp by utilizing edge calculation | |
| CN204883710U (en) | Power system fault diagnoses intelligent system based on rule | |
| CN112383516A (en) | Graph neural network construction method and abnormal flow detection method based on graph neural network | |
| CN109670584A (en) | A kind of fault diagnosis method and system based on big data | |
| CN110427298A (en) | A kind of Automatic Feature Extraction method of distributed information log | |
| CN113408087A (en) | Substation inspection method based on cloud side system and video intelligent analysis | |
| CN110286649A (en) | For the modeling and its method of sensor acquisition data on numerically-controlled machine tool | |
| CN113204227B (en) | Cloud collaborative fault diagnosis system and method for layered modular engineering machinery | |
| CN113900426B (en) | Remote equipment control and fault diagnosis system based on 5G+ industrial Internet | |
| CN112859885A (en) | Cooperative optimization method for path of feeding robot | |
| CN101382802A (en) | Maintenance machinery electrical system status monitoring with diagnose apparatus and monitoring and diagnose method | |
| CN115604013B (en) | Industrial data interaction platform and interaction method | |
| CN115374829A (en) | Deep learning-based bearing fault diagnosis method and system | |
| CN113344468B (en) | Industrial big data-based process quality control analysis platform | |
| CN115438958A (en) | Intelligent energy efficiency management system based on Internet and management method thereof | |
| CN115033253A (en) | Cloud edge collaborative training and deployment system and method of deep learning model | |
| CN114884695A (en) | Industrial control flow protocol identification method based on deep learning | |
| CN113780371A (en) | Insulator state edge recognition method based on edge calculation and deep learning | |
| CN206039171U (en) | Water treatment expert system based on wireless transmission | |
| CN107360260A (en) | A kind of system of remote control Internet of things access equipment | |
| CN206181262U (en) | Numerical control device video overlap intelligent terminal | |
| CN117473347B (en) | Ore dressing full-flow data processing method and system based on rule engine | |
| CN116776650B (en) | Running state control system of box-type substation |
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 |