Method for diagnosing equipment fault of fully mechanized coal mining face based on model thought

Abstract

The invention discloses a fully mechanized coal mining face equipment fault diagnosis method based on a model thought. A data model and a fault algorithm model are included. The data model comprises an object model and a business model; the fault algorithm model comprises a fault model, a fault identification algorithm model and a prediction algorithm model; the object model comprises basic attributes, inherent attributes and state attributes of mined and controlled equipment of a working face; the business model comprises process parameters of actions executed by the working face mined and controlled equipment; the fault model comprises a single-table fault model and a multi-table fault model; the fault recognition algorithm model comprises a fault source, a diagnosis strategy and a diagnosis algorithm of faults of mined and controlled equipment on the working face; and the prediction algorithm model comprises a fault source, a trend diagnosis strategy and a diagnosis algorithm of faults of the working surface acquired controlled equipment. A data model and a fault algorithm model are constructed, and a technical scheme for analyzing and solving a fault diagnosis problem in combination with actual equipment is provided for a fully mechanized coal mining face system.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment failure in a fully mechanized mining face, and in particular to a method for diagnosing equipment failure in a fully mechanized mining face based on a model concept. Background Art

[0002] The fully mechanized mining face is at the forefront of coal mine production, with a harsh working environment and products prone to failure. Due to the wide variety of sensors, equipment, and devices (the above-mentioned various detection parts or devices are collectively referred to as equipment), it is difficult to achieve fault diagnosis and analysis of the working face system.

[0003] Based on the analysis of working face equipment data and business, create object models and business models; based on the scenarios of fault conditions, create fault models; based on the summary of fault identification and fault prediction methods, create fault identification algorithm models and fault prediction algorithm models; through the precipitation of human-computer interaction and machine learning processes, form a fault knowledge base to achieve the purpose of quickly and accurately locating faulty equipment.

[0004] There are many technical solutions related to fault diagnosis in the field of industrial Internet of Things, such as the Chinese invention patent with application number CN202410338461.3, entitled A method, system, device and storage medium for diagnosing faults in industrial Internet of Things. Figure 1 As shown, the defect of this type of patent is that the technology mostly focuses on introducing the training and machine learning methods of fault models. These fault identification algorithms are certainly important, but lack the analysis and construction ideas related to the IoT devices themselves. The patented technology focuses on introducing the fault diagnosis technology methods of a specific device or the introduction of the fault identification algorithm model. It lacks systematic analysis and construction ideas, lacks the key methods of the actual implementation process, and it is difficult to realize the fault diagnosis and analysis of the working face system. Summary of the invention

[0005] The purpose of the present invention is to solve the above-mentioned problems and to provide a method for diagnosing equipment faults in a comprehensive mining working face based on a model concept. By analyzing the operating conditions of numerous products in the comprehensive mining working face, a physical model, a business model, a fault model, a fault identification algorithm model, and a fault prediction algorithm model are created. Through the continuous precipitation of human-computer interaction and machine learning, a fault knowledge base is formed to realize the identification and maintenance of working face sensor, equipment or equipment faults.

[0006] In order to achieve the above object, the technical solution of the present invention is: A method for diagnosing equipment faults in a fully mechanized working face based on a model concept, comprising a data model and a fault algorithm model; the data model comprises a physical model and a business model; the fault algorithm model comprises a fault model, a fault identification algorithm model, and a prediction algorithm model; The controlled equipment on the working surface constitutes a physical model, which includes basic attributes, inherent attributes and state attributes of the controlled equipment on the working surface; all businesses of the controlled equipment on the working surface are constructed into a business model, which includes process parameters of the actions performed by the controlled equipment on the working surface; The types of faults of the controlled equipment on the working face are constructed as fault models, including single-meter fault models and multi-meter fault models; the rules for the occurrence of faults of the controlled equipment on the working face are constructed as fault identification algorithm models, and the fault identification algorithm model includes the fault sources, diagnostic strategies and diagnostic algorithms of the faults of the controlled equipment on the working face; the rules for the impending failure of the controlled equipment on the working face are constructed as prediction algorithm models, and the prediction algorithm model includes the fault sources, trend diagnosis strategies and diagnostic algorithms of the faults of the controlled equipment on the working face.

[0007] As an improvement to the above technical solution, the basic attributes include but are not limited to the manufacturer, assembly model, and rated voltage of the controlled equipment on the working surface; the inherent attributes include but are not limited to the bus voltage, operating current, and operating torque of the controlled equipment on the working surface; the status attributes include but are not limited to the operating status of the controlled equipment on the working surface.

[0008] As an improvement to the above technical solution, the business model includes but is not limited to the UUID number of the controlled equipment on the working face, the business type, the scheduling method, and the activity list. The business type includes but is not limited to manual process, mining process, and auxiliary process; the scheduling method includes but is not limited to mutually exclusive method, sequential method, and parallel method.

[0009] As an improvement to the above technical solution, the single-meter fault model includes but is not limited to the non-overlapping parts of the UUID number of the controlled equipment on the working surface, fault code, fault name, fault type, fault level, occurrence time, occurring equipment, occurring components, fault phenomenon, fault cause, processing method, processing time, and operator; the multi-meter fault model includes but is not limited to the UUID number of the controlled equipment on the working surface, which refers to the overlapping part of multiple single-meter fault models.

[0010] As an improvement to the above technical solution, the fault source includes but is not limited to the UUID number, diagnostic object, and diagnostic component of the controlled equipment on the working surface; the diagnostic strategy and trend diagnostic strategy include but are not limited to the diagnostic algorithm code and diagnostic algorithm strategy parameters; the diagnostic algorithm includes but is not limited to the algorithm code, algorithm name, and algorithm strategy.

[0011] As an improvement to the above technical solution, the diagnostic strategy includes but is not limited to the UUID number of the controlled device on the working surface, the strategy number As an improvement to the above technical solution, the diagnostic algorithm and trend diagnostic strategy include but are not limited to the algorithm code and algorithm name of the controlled equipment on the working face.

[0012] As an improvement to the above technical solution, the steps of this method are: inputting the data model, generating multiple faults after calculation by the fault algorithm, and outputting multiple faults, and performing modeling analysis, modeling management, and modeling calculation on the entire process.

[0013] Compared with the prior art, the present invention has the following advantages and positive effects: The present invention models the specific "objects", "businesses" and "fault condition scenarios" of the working face, constructs "object models", "business models" and "fault models", and provides a technical solution that combines actual equipment analysis and solves fault diagnosis problems for the comprehensive mining working face system and even more industrial Internet of Things industries. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0015] Figure 1 A flowchart of an existing patent on a method for diagnosing faults in the industrial Internet of Things; Figure 2 A schematic diagram of the input and output relationship of the fault diagnosis system of the present invention; Figure 3 It is a schematic diagram of the structure of the fault diagnosis system based on modeling of the present invention; Figure 4 A schematic diagram of a model definition for a transfer machine model of the present invention; Figure 5 A schematic diagram of a business model definition model of the present invention; Figure 6 A schematic diagram of a model is defined for a fault identification model of the present invention; Figure 7 A model schematic diagram is defined for the fault prediction model of the present invention; Figure 8 A schematic diagram of a model definition for a single meter fault model of the present invention; Fig. 9 A schematic diagram of a fault knowledge base definition model of the present invention. DETAILED DESCRIPTION

[0016] 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, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work, any modifications, equivalent substitutions, improvements, etc., should be included in the protection scope of the present invention.

[0017] This embodiment discloses a method for diagnosing equipment faults in a fully mechanized mining face based on a model concept, including a data model and an algorithm model; the data model includes a physical model, a business model, and a fault model; the algorithm model includes a fault identification algorithm model and a prediction algorithm model; The controlled equipment on the working surface constitutes a physical model, which includes the basic attributes, inherent attributes and state attributes of the controlled equipment on the working surface; all the businesses of the controlled equipment on the working surface are constructed into a business model, and the business model includes the process parameters of the actions performed by the controlled equipment on the working surface; the indicators and states of the fault handling process of the controlled equipment on the working surface are constructed into a fault model, and the fault model includes The rules for the occurrence of faults in the controlled equipment of the working face are constructed as a fault identification algorithm model, which includes the fault sources, diagnostic strategies and diagnostic algorithms of the faults in the controlled equipment of the working face; the rules for the impending failure of the controlled equipment of the working face are constructed as a prediction algorithm model, which includes the fault sources, trend diagnosis strategies and diagnostic algorithms of the faults in the controlled equipment of the working face.

[0018] like Figure 2 As shown, the method for diagnosing equipment faults in the comprehensive mining working face based on the model concept of the present invention is to model the entire process of multiple inputs of the comprehensive mining working face fault diagnosis system, multiple inputs generating multiple faults, and multiple fault outputs, and perform model analysis, model management, and model calculation.

[0019] The specific idea of ​​modeling is: to build a physical model for the controlled equipment on the working face; to build a business model for all the equipment operations; to build a fault identification algorithm model for the rules of equipment failure; to build a prediction algorithm model for the rules of impending failure; and to build a fault model for the indicators and status of the fault handling process.

[0020] Figure 3The figure shows the structure of the fault diagnosis system based on modeling. The system identifies and predicts faults based on the data and events of standardized input from sensors, devices, and equipment, and outputs fault models based on the results of identification and prediction. We can analyze, confirm, remove, insert, and other operations on the facts of the fault model through human-computer interaction and other methods, and finally precipitate this information to form a fault knowledge base; at the same time, we can also train the fault identification algorithm and fault prediction algorithm in the process of human-computer interaction to optimize the efficiency and accuracy of equipment fault diagnosis.

[0021] Generally speaking, it is divided into the following steps: 1. Definition of Object Model and Business Model Based on the analysis of the data of the working face equipment (taking the transfer machine as an example), the idea of ​​building the object model can refer to Figure 4 As shown, the transfer machine equipment is distinguished and defined from multiple angles, including basic attributes, inherent attributes, and status attributes.

[0022] Based on the analysis of working face equipment business, you can refer to Figure 5 As shown, the business model is constructed through multiple dimensions such as business activity type, scheduling method, scheduling method, activity list, etc.

[0023] 2. Definition of fault identification model and fault prediction model For working face equipment, when conducting statistics and analysis on common fault scenarios, we can build a concrete and lightweight fault identification model through methods such as knowledge distillation; in the statistical analysis of fault development trends and laws, we use machine learning methods such as neural networks to form a fault prediction model.

[0024] Figure 6 The figure shows a fault identification model, which can be composed of multiple aspects including fault source, fault diagnosis strategy (fault rule), and fault diagnosis algorithm.

[0025] Figure 7 The figure shows a fault prediction model, which can be composed of multiple aspects such as fault source, fault trend diagnosis strategy (fault development law), and fault diagnosis algorithm.

[0026] 3. Definition of fault model Just like OBD (on-board diagnostic system), we need to summarize and classify all possible faults on the working surface to form a standardized fault model, such as Figure 8 As shown in the figure, the single-meter fault model of the working face equipment can be constructed from multiple angles such as the time, location, cause, result, person, operation, etc. of the fault. Multiple single-meter fault models can be extended to multi-meter fault models if there are overlapping scenarios.

[0027] 4. Definition of fault knowledge base The process of forming a fault knowledge base is roughly as follows: When the system generates fault alarm information through fault diagnosis and fault identification, the system will first compare it with the existing knowledge base. If the comparison result meets the fault condition, it will trigger the fault elimination action according to the existing fault handling mechanism in the knowledge base. If the comparison result does not meet the fault condition, the operator will be prompted to check the fault alarm information through the human-machine interface; after the operator confirms that the fault has no actual impact, the fault information can be eliminated. The fault information will also be fed back to the fault identification algorithm model and the fault prediction model through model training. The next time the same fault occurs, it will be automatically filtered.

[0028] Through the above-mentioned human-computer interaction and machine learning process, the system gradually accumulates and forms a fault knowledge base, optimizing the efficiency and accuracy of fault identification and fault prediction.

[0029] Fig. 9 The fault knowledge base shown provides a way of thinking. Fault knowledge can be defined from the perspectives of fault coding, fault name, fault type, fault level, etc.

[0030] The key point of the present invention is to provide an idea of ​​building "object model", "business model" and "fault model" around the specific "objects", "business" and "fault condition scenarios" of the working face, and to provide a technical solution from physical objects to digital fault diagnosis systems for the comprehensive mining working face system and even more industrial Internet of Things industries. The patent intends to protect the method of processing the entire fault diagnosis process through modeling ideas and the modeling analysis method of each link of the fault diagnosis system.

[0031] Compared with the prior art, the present invention has the following beneficial effects: The advantage of the present invention is that it visualizes the actual equipment fault diagnosis system of the fully mechanized working face through the modeling concept, and modeling and defining and analyzing multiple key links or objects of the fault diagnosis process by combining specific equipment, specific working conditions, specific fault rules and fault trends, which helps to cope with diversified production environments and solve the problems of inconsistency of various equipment fault scenarios, fault rules, etc. Simplify the process from concept to implementation of the industrial Internet of Things equipment fault diagnosis system.

[0032] The modeling idea mentioned in the technical solution can also be equated with the concept of object. Therefore, whether constructing the working face equipment fault diagnosis system from the perspective of objectification or modeling can be regarded as a thought method or technical solution. The purpose is to digitize the analysis of system fault diagnosis and realize it in a concrete way.

Claims

1. A method for diagnosing equipment failure in fully mechanized mining working face based on model thinking, characterized in that: It includes a data model and a fault algorithm model; the data model includes a physical model and a business model; the fault algorithm model includes a fault model, a fault identification algorithm model, and a prediction algorithm model; The controlled equipment on the working surface constitutes a physical model, and the physical model includes basic attributes, inherent attributes and state attributes of the controlled equipment on the working surface; All businesses of the controlled equipment on the working surface are constructed into a business model, and the business model includes process parameters of the actions performed by the controlled equipment on the working surface; The types of faults of the controlled equipment on the working face are constructed as fault models, including single-meter fault models and multi-meter fault models; the rules for the occurrence of faults of the controlled equipment on the working face are constructed as fault identification algorithm models, and the fault identification algorithm model includes the fault sources, diagnostic strategies and diagnostic algorithms of the faults of the controlled equipment on the working face; the rules for the impending failure of the controlled equipment on the working face are constructed as prediction algorithm models, and the prediction algorithm model includes the fault sources, trend diagnosis strategies and diagnostic algorithms of the faults of the controlled equipment on the working face.

2. The method for diagnosing equipment failure in a fully mechanized mining face based on the model concept as claimed in claim 1 is characterized by: The basic attributes include but are not limited to the manufacturer, assembly model, and rated voltage of the controlled equipment on the working surface; the inherent attributes include but are not limited to the bus voltage, operating current, and operating torque of the controlled equipment on the working surface; the status attributes include but are not limited to the operating status of the controlled equipment on the working surface.

3. The method for diagnosing equipment failure in a fully mechanized mining face based on the model concept as claimed in claim 1 is characterized by: The business model includes but is not limited to the UUID number of the controlled equipment on the working face, the business type, the scheduling method, and the activity list. The business type includes but is not limited to manual process, mining process, and auxiliary process; the scheduling method includes but is not limited to mutually exclusive method, sequential method, and parallel method.

4. The method for diagnosing equipment failure in a fully mechanized mining face based on a model concept as claimed in claim 1 is characterized in that: The single-meter fault model includes but is not limited to the non-overlapping parts of the UUID number, fault code, fault name, fault type, fault level, occurrence time, occurring equipment, occurring components, fault phenomenon, fault cause, processing method, processing time, and operator of the equipment controlled by the working surface; the multi-meter fault model includes but is not limited to the UUID number of the equipment controlled by the working surface, which refers to the overlapping part of multiple single-meter fault models.

5. The method for diagnosing equipment failure in fully mechanized mining face based on model concept as claimed in claim 1 is characterized by: The fault sources include but are not limited to the UUID number, diagnostic object, and diagnostic components of the controlled equipment on the working surface; the diagnostic strategy and trend diagnostic strategy include but are not limited to the diagnostic algorithm code and diagnostic algorithm strategy parameters; the diagnostic algorithm includes but is not limited to the algorithm code, algorithm name, and algorithm strategy.

6. The method for diagnosing equipment failure in a fully mechanized mining face based on a model concept as claimed in claim 1 is characterized in that: The diagnostic strategy includes but is not limited to the UUID number and strategy number of the controlled device on the working surface.

7. The method for diagnosing equipment failure in a fully mechanized mining face based on a model concept as claimed in claim 1 is characterized in that: The diagnostic algorithm and trend diagnostic strategy include but are not limited to the algorithm code and algorithm name of the controlled equipment on the working face.

8. The method for diagnosing equipment failure in fully mechanized mining face based on model concept as claimed in claim 1 is characterized by: The data model is input, and after calculation by the fault algorithm, multiple faults are generated and multiple fault outputs are generated. The entire process is modeled for analysis, modeled management, and modeled calculation.

Images