Fault processing method and device of coal mining machine, storage medium and electronic equipment
By integrating multi-channel sensor data acquisition, abnormal feature extraction and knowledge graph retrieval technology of convolutional neural network model in coal miner fault processing, the problems of low degree of automation of coal miner fault processing and underutilized knowledge graph potential are solved, efficient and accurate fault detection and type judgment are achieved, and the practicality and applicability of the system are improved.
Patent Information
- Application Number
- CN202411974394.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-30
- Publication Date
- 2025-04-25
AI Technical Summary
The existing technology has low degree of automation in the fault handling of coal mining machines in the field of coal mining, and it is impossible to achieve unsupervised independent fault learning, the timeliness of fault identification are not high, and the potential of the knowledge graph model is not fully utilized, and the output content is thin, so it cannot provide comprehensive fault diagnosis and processing references.
By collecting the operation data of the multi-channel sensor of the coal mining machine, using the convolutional neural network model to extract abnormal features and classify and identify them to generate fault data. Then, using the knowledge graph search to index the fault data, a matching scheme for handling coal miner failures is generated. The method also supports regular updates to the model and knowledge base to enhance the system's self-learning and applicability.
It realizes efficient and accurate fault detection and type judgment, has real-time prediction and early warning functions, and can accurately determine the most likely fault situations. The use of knowledge graph models and RAG technology has improved the practicality of the system and helped on-site engineers to easily obtain first-hand information for fault repairs.
Smart Images
Figure CN119885020A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of coal mining equipment, and in particular to a fault handling method, device, storage medium and electronic equipment for a coal mining machine. Background Art
[0002] In recent years, with the continuous advancement of science and technology, the technology of using convolutional neural network models to conduct deep learning and accurately identify equipment failures has made rapid progress and has been widely used in many fields such as industry and manufacturing, providing strong support for equipment maintenance and production safety; at the same time, with the help of professional knowledge integrated by knowledge graphs and intelligent reasoning of RAG technology, the level of automatic generation of solutions for technical failures is also constantly improving, and the prospects are promising.
[0003] Currently, in the field of coal mining, there are cases in the existing technology that use the above-mentioned artificial intelligence technology to build equipment diagnosis platforms and provide intelligent services for workers. For example, after personnel report a fault or the software platform alarms, data that can be connected to the big model is obtained, and then the RAG technology and knowledge graph are combined with the big model to obtain an equipment abnormality analysis plan.
[0004] The existing technology has significant defects: first, it relies too much on manual fault identification, making it difficult to achieve an unsupervised autonomous fault learning model, resulting in low automation and inability to effectively use intelligent means to make accurate judgments when underground coal machinery encounters complex fault situations, and the timeliness of fault identification is not high; second, it fails to fully tap the potential of the knowledge graph big model. The accuracy of the analysis results using the knowledge graph big model is low, and the output content is not rich enough, which cannot provide comprehensive reference information for fault diagnosis and processing. Summary of the invention
[0005] The main purpose of the present invention is to solve the problems in the related art of over-reliance on manual fault identification, low degree of automation, weak ability to analyze complex situations, low timeliness of fault identification, inability to give full play to the potential of the knowledge graph large model, thin output content, and inability to provide comprehensive reference information for fault diagnosis and processing.
[0006] In order to achieve the above object, according to one aspect of the present invention, a method for troubleshooting a coal mining machine is provided, comprising the following steps:
[0007] Collect the operating data of the coal mining machine's multi-channel sensors to form a data source;
[0008] Extracting abnormal features from the data source through the first model, and classifying and identifying the abnormal features to obtain fault data;
[0009] Using the fault data as an index, performing knowledge graph retrieval to obtain retrieval results;
[0010] A matching solution for handling coal mining machine failure is generated according to the search results.
[0011] In some embodiments, the coal mining machine fault processing method further includes the following steps: collecting the retrieved uncovered data, performing intervention processing on the retrieved uncovered data to form a sample set;
[0012] The first model for identifying and classifying the abnormal features is trained periodically based on the sample set.
[0013] In some embodiments, classifying and identifying the abnormal features includes:
[0014] Classify the abnormal features according to known fault categories and unknown fault categories, and output classification results;
[0015] Identify the abnormal state of the abnormal feature and output the identification result;
[0016] The classification results and the identification results are summarized as fault information and organized into known fault data and unknown fault data.
[0017] In some embodiments, the known fault data includes phenomenon information and type information of the fault;
[0018] The unknown fault data includes phenomenon information of the fault.
[0019] In some embodiments, the using the fault data as an index to perform knowledge graph retrieval includes:
[0020] Performing knowledge graph retrieval based on phenomenon information and type information in the known fault data;
[0021] Knowledge graph retrieval and similarity comparison are performed based on the phenomenon information in the unknown fault data.
[0022] In some embodiments, generating a matching solution for handling coal mining machine failure according to the search result includes:
[0023] Template matching is performed on the retrieval results to generate a matching solution based on the coal mining machine fault.
[0024] In some embodiments, performing template matching on the search results includes:
[0025] Performing template matching on the search results to generate prompt information;
[0026] According to the prompt information, the search results are inferred and information fused through a second model.
[0027] According to another aspect of the present invention, there is provided a fault handling device for a coal mining machine, comprising: a data acquisition module, an abnormal classification and identification module, a knowledge graph retrieval module and an output module;
[0028] The data acquisition module is used to collect the operating data of the multi-channel sensors of the coal mining machine to form a data source;
[0029] The abnormal classification and identification module is used to extract abnormal features in the data source, classify and identify the abnormal features, and obtain fault data;
[0030] The knowledge graph retrieval module uses the fault data as an index to perform a knowledge graph retrieval to obtain a retrieval result;
[0031] The output module performs template matching on the search results and outputs a matching solution according to the coal mining machine fault;
[0032] The fault handling device can implement the steps of any of the above-mentioned coal mining machine fault handling methods.
[0033] According to another aspect of the present invention, there is provided a storage medium storing a computer program, wherein when the computer program is executed by a processor, the steps of any one of the above-mentioned coal mining machine fault handling methods are implemented.
[0034] According to another aspect of the present invention, there is provided an electronic device comprising at least a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above-mentioned coal mining machine fault handling methods when executing the computer program on the memory.
[0035] This disclosure uses artificial intelligence models to achieve efficient and accurate detection of abnormal fault conditions and fault types of coal mining machines. It has real-time prediction and early warning functions and can accurately identify the most likely fault situations. The use of knowledge graph big models and RAG technology can effectively improve the practicality of the system and help on-site engineers easily obtain first-hand information on fault repairs. This method has a strong self-learning ability, and the system supports manual intervention and optimization. It can continuously optimize the model's prediction effect and fault type identification accuracy based on accumulated data, and simultaneously update the knowledge base. The applicability gradually increases with the increase in operating time. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] The drawings constituting a part of the present application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:
[0037] Figure 1 A schematic diagram showing the steps of a method for troubleshooting a coal mining machine according to an embodiment of the present disclosure is shown;
[0038] Figure 2 A schematic structural diagram of a fault handling device for a coal mining machine according to an embodiment of the present disclosure is shown;
[0039] Figure 3 A schematic diagram showing the structure of an abnormal classification and identification module according to an embodiment of the present disclosure is shown;
[0040] Figure 4 A specific neural network design schematic diagram based on a CNN-LSTM autoencoder according to an embodiment of the present disclosure is shown;
[0041] Figure 5 A schematic diagram showing the structure of a knowledge graph retrieval module and an output module in an embodiment of the present disclosure is shown; DETAILED DESCRIPTION
[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present invention and its application or use. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0043] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.
[0044] Unless otherwise specifically stated, the relative arrangement of the parts and steps described in these embodiments, numerical expressions and numerical values do not limit the scope of the present invention. At the same time, it should be understood that, for ease of description, the sizes of the various parts shown in the accompanying drawings are not drawn according to the actual proportional relationship. The technology, method and equipment known to ordinary technicians in the relevant field may not be discussed in detail, but in appropriate cases, the technology, method and equipment should be regarded as a part of the authorization specification. In all examples shown and discussed here, any specific value should be interpreted as being merely exemplary, rather than as a limitation. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once a certain item is defined in an accompanying drawing, it does not need to be further discussed in subsequent drawings.
[0045] This disclosure lists artificial intelligence technical means that can realize the technical solutions provided by this disclosure, such as convolutional neural network model CNN, long short-term memory network model LSTM, large language model LLM, etc., and uses this as an example for specific description. It is worth noting that relevant technical personnel should have the ability to use other artificial intelligence means to achieve the above technical solutions for the purpose of achieving the same or similar technical effects, and should be considered to fall within the scope of this disclosure.
[0046] Embodiment 1
[0047] The present disclosure provides a method for troubleshooting a coal mining machine. Figure 1 , 2 From the above, the troubleshooting method includes the following steps:
[0048] S1: Collect the operating data of the coal mining machine's multi-channel sensors to form a data source.
[0049] Specifically, the data source formed by collecting the operating data of the multi-channel sensors of the coal mining machine includes: collecting data from the multi-channel sensors distributed in various key parts of the coal mining machine, such as the operating data of the temperature sensor installed on the motor, the vibration sensor of the coal mining machine cutting drum, and the pressure sensor used to monitor the hydraulic pressure.
[0050] Furthermore, it also includes data cleaning, standardization, and normalization of the operating data to eliminate interference factors. The data collection tool connects the time series data of multiple coal machine sensors to the system, completes data governance to form the format required by the model, and then inputs it into the abnormal classification and identification module 2.
[0051] S2: extracting abnormal features from the data source, classifying and identifying the abnormal features, and obtaining fault data.
[0052] Specifically, this step is completed using the abnormal classification and recognition module 2, and the abnormal classification and recognition module 2 can use the CNN-LSTM autoencoder deep learning model, in which the convolutional neural network CNN is good at extracting local features of the data and can accurately capture the key patterns in the input data; the long short-term memory network LSTM focuses on processing time series data and has excellent memory and analysis capabilities for information with time series; the autoencoder (Autoencoder) can effectively achieve data dimensionality reduction, feature extraction and reconstruction through the encoding and decoding process of the data. The model has shown strong performance and application potential in many fields such as fault diagnosis and time series prediction.
[0053] For further information, see Figure 3 The fault data obtained in this step is divided into known fault data and unknown fault data.
[0054] The known fault data includes the phenomenon information and type information of the fault; the unknown fault data includes the phenomenon information of the fault. It can be considered that the known fault data not only represents the abnormal phenomenon of the fault, but also the existing type of the fault; the unknown fault data only represents the abnormal phenomenon of the fault, and it is currently impossible to know what type the fault should belong to.
[0055] Specifically, Figure 3 As shown, the feature extraction submodule 21 (convolutional neural network model CNN) is first used to extract abnormal features from the data source, and the abnormal classification submodule 22 (such as Softmax function classification) is used to classify the abnormal features according to known fault categories and unknown fault categories, and the classification results are output. At the same time, the abnormal recognition submodule 23 (long short-term memory network model LSTM) is used to identify the abnormal state of the abnormal features and output the recognition results.
[0056] Among them, the classification using the abnormal classification submodule output 22 is based on the known fault category label. Therefore, the output abnormal classification data contains known fault categories and unknown fault categories, and it is impossible to distinguish whether there is a fault; the identification using the abnormal identification submodule 23 is based on the timing to identify whether there is an abnormal state. Therefore, the output abnormal identification data contains the abnormal state of the data, but the fault type cannot be distinguished.
[0057] By organizing the above-mentioned abnormal classification data and abnormal identification data using the fault information summary submodule 24, known fault data and unknown fault data can be obtained. In this way, the autonomous learning advantages of the CNN-LSTM autoencoder can be fully utilized to achieve automatic unsupervised learning of the data source, providing a basis for knowledge graph retrieval.
[0058] Figure 4 A specific neural network design based on CNN-LSTM autoencoder is shown.
[0059] like Figure 4 As shown in the figure, the model input is composed of multi-channel sensor time series data. The number of data sources is the number of input channels S of Conv1. For example, if the data source dimension is 10, the dimension of the input dimension is also 10. Through the convolution operation, the output layer will be converted into more feature maps, such as 16, to extract higher dimensional features.
[0060] The input dimension of the convolution layer Conv2 is the dimension of the feature map generated by Conv1, while the dimension of the output layer is richer. For example, if the input layer is 16-dimensional, the output can be 32-dimensional channels. It is important to coordinate the relationship between the convolution kernel kernel_size and the LSTM time series length seq_len, and select an appropriate value to ensure that kernel_size is less than or equal to seq_len. The input data is extracted through CNN to extract high-dimensional features, and the extracted features are used as the input of the time step and passed to the LSTM encoder.
[0061] In addition, the output of the CNN convolutional network also includes a fully connected layer for classification training, so that the CNN network can be trained and modeled through labeled data, for example, using the Softmax function for classification training.
[0062] The LSTM encoder generates a hidden state layer. In the decoder stage, the decoder generates low-dimensional features of the time step from the hidden state of the LSTM. The low-dimensional features are restored to data close to the input shape through the deconvolution layer (Conv Transpose1d). This method allows the model to optimize both the CNN and LSTM parts during training to achieve efficient reconstruction of the input data. In practice, the LSTM model can be trained and modeled using multi-source data. The LSTM encoder is divided into two groups of models, one for modeling time series data with each independent data source, and the other for modeling all multi-source data sources together and observing data offsets from a single independent data source when anomalies are observed.
[0063] In the specific implementation, after being processed by the CNN-LSTM autoencoder, an abnormal prompt can be given according to the abnormal characteristics; if the fault type is known, the known fault data is generated and the fault type suggestion is generated; if the fault type is unknown, the unknown fault data is generated and the abnormal information prompted indicates the data source of the data and outputs the deviation from the normal value. Then, proceed to the next step.
[0064] S3: Use the fault data as an index to perform knowledge graph search to obtain search results.
[0065] Further, such as Figure 5 As shown, the known fault retrieval submodule 31 performs knowledge graph retrieval based on the phenomenon information and type information of the known fault data, and the unknown fault retrieval module 32 performs knowledge graph retrieval and similarity comparison based on the phenomenon information of the fault included in the unknown fault data.
[0066] Specifically, when searching based on known fault data, according to the known fault data phenomena (such as abnormal machine vibration, sudden temperature rise) and types (such as mechanical failure, circuit failure), it is possible to quickly locate the corresponding past cases, cause analysis, maintenance methods and other related knowledge, thereby more quickly assisting the staff to quickly judge the current fault condition and accurately prescribe the right medicine.
[0067] At the same time, when searching for unknown fault data, since the only and most descriptive clue is the fault phenomenon, it is possible to dig out similar fault scenarios by searching in the knowledge graph with the help of this phenomenon information. Even if a direct solution cannot be obtained, related clues can be found to help technicians open up their thinking, infer the cause of the fault from similar cases, and try to find ways to deal with unknown faults.
[0068] Before this step is performed, it is necessary to construct a knowledge graph about coal mining machine failures, that is, to collect information related to coal mining machine failures from multiple knowledge graph data sources, including technical manuals, maintenance records, sensor data, operator experience descriptions, and expert knowledge of coal mining machines; analyze the knowledge graph data source to identify key entities, including fault types, fault phenomena, and corresponding component names, etc., determine the relationship between entities, and then perform knowledge fusion and verification, and update the knowledge graph regularly based on new data. Since knowledge graph construction is already a relatively mature existing technology, this disclosure does not elaborate on it.
[0069] Furthermore, when performing knowledge graph retrieval, if according to the above steps, abnormal features are extracted through CNN-LSTM autoencoder processing to form unknown fault data, after the knowledge graph retrieval, no valid information can be retrieved, and these data need to be collected as uncovered data.
[0070] Afterwards, if Figure 5 As shown, these uncovered data are input into the intervention module 5 for intervention processing, and the model is fine-tuned using historical data and field test results, and relevant data is added to the sample set, and the model used to identify and classify the abnormal features is retrained regularly based on the sample set. In this way, the system can give a fault result when the same type of fault occurs again.
[0071] After performing a knowledge graph search and obtaining search results, proceed to the next step.
[0072] S4: Generate a matching solution for handling coal mining machine failure according to the search results.
[0073] Furthermore, a matching solution for handling coal mining machine failures is generated based on the retrieval results, including: using the retrieval enhancement submodule 41 to perform template matching on the retrieval results to generate prompt information; using the fusion reasoning submodule 42 to perform reasoning and information fusion based on the prompt information through the large language model LLM to obtain a matching solution.
[0074] That is, this step combines the knowledge graph with RAG technology and provides the retrieved content to the big model for reasoning. In this way, a rich output of relevant maintenance knowledge for fault prediction is obtained through a professional industry professional knowledge base, providing real-time professional operation and maintenance service capabilities for on-site staff and experts.
[0075] Specifically, RAG technology (retrieval enhanced generation technology) is a method that combines information retrieval and language generation, which can perform template matching using the retrieval results retrieved from the knowledge graph.
[0076] For example, in the coal mining machine fault handling scenario, a search is performed based on a fault phenomenon, motor overheating, to obtain search results such as component information related to motor overheating (such as windings, bearings, etc.), possible causes (such as overload, poor heat dissipation, etc.), and historical maintenance cases.
[0077] Then, these search results are template matched. A template can be a predefined text format used to organize the retrieved information. For example, a simple template may be "Fault phenomenon: [specific fault phenomenon]; Possible cause: [list possible causes]; Reference repair case: [list repair cases]". By filling the search results according to this template, a prompt message can be generated: "Fault phenomenon: motor overheating; Possible cause: overload, poor heat dissipation; Reference repair case: At [specific time], the motor overheating problem caused by dust blockage was solved by cleaning the radiator."
[0078] After that, the prompt is input into the large language model LLM to guide the large language model LLM to perform reasoning. The large language model LLM will further analyze the fault situation based on the information in the prompt and its own trained language knowledge and logic. For example, based on the "possible cause" part in the above prompt, the large language model LLM can infer how to further check the overload condition (such as checking whether the load current exceeds the rated value) or poor heat dissipation (such as checking whether the radiator is blocked, whether the fan is operating normally, etc.).
[0079] In addition, the large language model LLM also performs information fusion. It can combine the knowledge graph retrieval results in the prompt with a large amount of text knowledge trained by itself. For example, the large language model LLM may know other consequences of motor overload or the manifestations of poor heat dissipation in different machines, and integrate this knowledge with the specific situation of the coal mining machine motor overheating in the prompt to generate a more comprehensive and accurate matching solution. For example, a matching solution such as "First check whether the load current of the motor exceeds the rated value. If so, consider reducing the load; at the same time, check whether the radiator is blocked and whether the fan is operating normally. If blocked, clean the radiator; if the fan fails, replace the fan."
[0080] In summary, the embodiments of the present disclosure can achieve the following beneficial effects:
[0081] With the help of artificial intelligence models, the abnormal fault conditions and fault types of coal mining machines can be detected efficiently and accurately. It has real-time prediction and early warning functions and can accurately identify the most likely fault situations. The use of knowledge graph big models and RAG technology can effectively improve the practicality of the system and help on-site engineers easily obtain first-hand information on fault repairs. This method has a strong self-learning ability. At the same time, the system supports manual intervention and optimization. It can continuously optimize the model's prediction effect and fault type identification accuracy based on accumulated data, and simultaneously update the knowledge base. The applicability gradually increases with the increase in operating time.
[0082] Embodiment 2
[0083] like Figure 2 As shown, this embodiment provides a fault handling device for a coal mining machine, including: a data acquisition module 1, an abnormal classification and identification module 2, a knowledge graph retrieval module 3 and an output module 4.
[0084] The data acquisition module 1 is used to collect the operating data of the multi-channel sensors of the coal mining machine to form a data source;
[0085] The abnormal classification and identification module 2 is used to extract abnormal features in the data source, classify and identify the abnormal features, and obtain fault data;
[0086] The knowledge graph retrieval module 3 uses the fault data as an index to perform a knowledge graph retrieval to obtain a retrieval result;
[0087] The output module 4 performs template matching on the search results and outputs a matching solution based on the coal mining machine fault.
[0088] The fault handling device provided in this embodiment is used to implement the steps of the fault handling method for the coal mining machine described in the first embodiment.
[0089] Embodiment 3
[0090] A storage medium stores a computer program, characterized in that when the computer program is executed by a processor, the steps of the coal mining machine fault handling method described in Example 1 are implemented.
[0091] Embodiment 4
[0092] An electronic device comprises at least a memory and a processor, wherein the memory stores a computer program, and is characterized in that the processor implements the steps of the coal mining machine fault handling method described in Example 1 when executing the computer program on the memory.
[0093] The storage medium may be included in the electronic device, or may exist independently without being installed in the electronic device.
[0094] The storage medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: obtains at least two Internet Protocol addresses; sends a node evaluation request including at least two Internet Protocol addresses to a node evaluation device, wherein the node evaluation device selects an Internet Protocol address from the at least two Internet Protocol addresses and returns it; receives the Internet Protocol address returned by the node evaluation device; wherein the obtained Internet Protocol address indicates an edge node in a content distribution network.
[0095] Alternatively, the storage medium carries one or more programs, and when the one or more programs are executed by the electronic device, the electronic device: receives a node evaluation request including at least two Internet Protocol addresses; selects an Internet Protocol address from the at least two Internet Protocol addresses; and returns the selected Internet Protocol address; wherein the received Internet Protocol address indicates an edge node in a content distribution network.
[0096] Computer program code for performing operations of the present disclosure may be written in one or more programming languages, or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the passenger computer, partially on the passenger computer, as a separate software package, partially on the passenger computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the passenger computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).
[0097] It should be noted that the storage medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium may also be any storage medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.
[0098] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.
[0099] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware, wherein the name of a unit does not, in some cases, limit the unit itself.
[0100] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.
[0101] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
[0102] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.
[0103] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.
[0104] Although the subject matter has been described in language specific to structural features and / or methodological logical actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims.
[0105] Multiple embodiments of the present disclosure are described in detail above, but the present disclosure is not limited to these specific embodiments. Those skilled in the art can make various variations and modifications to the embodiments based on the concepts of the present disclosure, and these variations and modifications should all fall within the scope of protection required by the present disclosure.
Claims
1. A method for troubleshooting a coal mining machine, characterized in that: The following steps are involved: Collect the operating data of the coal mining machine's multi-channel sensors to form a data source; Extracting abnormal features from the data source through the first model, and classifying and identifying the abnormal features to obtain fault data; Using the fault data as an index, performing knowledge graph retrieval to obtain retrieval results; A matching solution for handling coal mining machine failure is generated according to the search results.
2. The method for troubleshooting a coal mining machine according to claim 1, characterized in that: The following steps are also included: Collecting the uncovered data, performing intervention processing on the uncovered data, and forming a sample set; The first model for identifying and classifying the abnormal features is trained periodically based on the sample set.
3. The method for troubleshooting a coal mining machine according to claim 1, characterized in that: Classifying and identifying the abnormal features, including: Classify the abnormal features according to known fault categories and unknown fault categories, and output classification results; Identify the abnormal state of the abnormal feature and output the identification result; The classification results and the identification results are summarized as fault information and organized into known fault data and unknown fault data.
4. The method for troubleshooting a coal mining machine according to claim 3, characterized in that: in, The known fault data includes phenomenon information and type information of the fault; The unknown fault data includes phenomenon information of the fault.
5. The method for troubleshooting a coal mining machine according to claim 4, characterized in that: The method of using the fault data as an index to perform knowledge graph retrieval includes: Performing knowledge graph retrieval based on phenomenon information and type information in the known fault data; Knowledge graph retrieval and similarity comparison are performed based on the phenomenon information in the unknown fault data.
6. The method for troubleshooting a coal mining machine according to claim 1, characterized in that: A matching solution for handling coal mining machine failure is generated according to the search results, including: Template matching is performed on the retrieval results to generate a matching solution based on the coal mining machine fault.
7. The method for troubleshooting a coal mining machine according to claim 6, characterized in that: Performing template matching on the search results includes: Performing template matching on the search results to generate prompt information; According to the prompt information, the search results are inferred and information fused through a second model.
8. A fault handling device for a coal mining machine, characterized in that: include: Data collection module, abnormal classification and identification module, knowledge graph retrieval module and output module; The data acquisition module is used to collect the operating data of the multi-channel sensors of the coal mining machine to form a data source; The abnormal classification and identification module is used to extract abnormal features in the data source, classify and identify the abnormal features, and obtain fault data; The knowledge graph retrieval module uses the fault data as an index to perform a knowledge graph retrieval to obtain a retrieval result; The output module performs template matching on the search results and outputs a matching solution according to the coal mining machine fault; The fault handling device can implement the steps of the fault handling method for a coal mining machine in any one of claims 1 to 7.
9. A storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the fault handling method for a coal mining machine according to any one of claims 1 to 7 are implemented.
10. An electronic device, comprising at least a memory and a processor, wherein a computer program is stored in the memory, wherein: The processor implements the steps of the fault handling method for a coal mining machine according to any one of claims 1 to 7 when executing the computer program on the memory.