CN115906938A

CN115906938A - Method for detecting and analyzing oil of mine gearbox and system machine storage medium

Info

Publication number: CN115906938A
Application number: CN202211392344.2A
Authority: CN
Inventors: 刘志明; 陶伟忠; 金刚; 王妙云; 李滢
Original assignee: China Coal Industry Group Information Technology Co ltd
Current assignee: China Coal Industry Group Information Technology Co ltd
Priority date: 2022-11-08
Filing date: 2022-11-08
Publication date: 2023-04-04

Abstract

The invention discloses a method and a system for detecting and analyzing oil in a mine gearbox and a storage medium, and the method comprises the following steps: s1, establishing a prediction model for oil of a target equipment gearbox; s2, establishing a real-time model for oil of a gearbox of target equipment; s3, carrying out oil detection on the target equipment, and bringing various detected data into a prediction model and a real-time model to carry out gear box state simulation and data processing; s4, establishing a user interaction interface, and transmitting the prediction data of the prediction model and the detection data of the real-time model to the user interaction interface; and S5, constructing a mine equipment fault analysis model, extracting characteristic values of the data obtained in the step S3, bringing the characteristic values into the fault diagnosis model, and carrying out fault analysis on the gear box of the target equipment by combining with detection on the running condition of the easily damaged parts in the target gear box. The invention provides the method for detecting the oil on line, thereby avoiding a series of problems caused by manual oil sample detection and greatly improving the detection efficiency.

Description

Method for detecting and analyzing oil of mine gearbox and system machine storage medium

Technical Field

The invention relates to the technical field of maintenance of mine equipment, in particular to a method and a system for detecting and analyzing oil of a mine gearbox and a storage medium.

Background

The underground working environment is quite severe, the machine is not only polluted by huge impact loads from coal, gangue and the like, coal dust, gas and the like during operation, but also equipment such as a coal mining machine, a scraper conveyor, a heading machine and the like are in an environment which is constantly pushed and dynamically developed, the equipment is coal mining core equipment with the worst working condition and the most complex load condition in complete equipment of a fully mechanized mining face, and the failure rate of gear boxes of the equipment is very high under the high-strength severe condition for a long time. Coal mining equipment often has little or no ability to detect vibration or temperature changes at the early stages of failure development. More importantly, these methods fail to identify wear conditions of the equipment and provide a prediction of equipment failure. The oil monitoring technology is characterized in that a lubricating oil sample of coal mining equipment is sampled, physical and chemical indexes, pollutant indexes and wear indexes of the sample are analyzed by various monitoring technical means, and data result analysis is carried out by combining actual working conditions, lubricating states and the like of the equipment.

The common sampling and submission mode of the oil detection at present utilizes manual sampling and submission, and the manual sampling and submission can have the following defects: the installation environment of the coal mine part equipment is severe, the manual sampling mode is greatly limited by regions, and the coal mine part equipment is easily polluted by the environment; manually sampling on site, and sending to a ground laboratory for detection, wherein the sending and detecting period is long; the sampling and detection work has great dependence on the quality of staff, and the consistency and the stability of the work quality cannot be ensured; and the manual off-line mode can not monitor in real time in large area

Therefore, the two technologies of "online oil monitoring" and "fault diagnosis" are important directions for the research of oil monitoring technology, and a method or a system is needed to be provided, so that the oil of the gear box can be detected online in real time, the cause and the part of the fault of target equipment can be scientifically positioned, and meanwhile, the daily prevention and monitoring of the wear condition of industrial equipment in use can be realized.

Disclosure of Invention

The present invention is directed to solving, at least to some extent, one of the technical problems in the related art.

In order to achieve the aim, the invention provides a method for detecting and analyzing oil in a mine gearbox, which comprises the following steps:

s1, establishing a prediction model for oil of a target equipment gear box, and predicting the oil state at the next time point by operating the target equipment gear box in the current state;

s2, establishing a real-time model for oil of the target equipment gearbox, and performing real-time oil state simulation on the target equipment gearbox in the current state;

s3, oil detection is carried out on the target equipment gearbox, and detected data are brought into a prediction model and a real-time model to carry out gearbox state simulation and data processing;

s4, establishing a user interaction interface, transmitting the prediction data of the prediction model and the detection data of the real-time model to the user interaction interface, and performing graphical or textual processing on the data according to the user requirements;

and S5, constructing a mine equipment fault analysis model, extracting characteristic values of the data obtained in the step S3, bringing the characteristic values into the fault diagnosis model, and carrying out fault analysis on the gear box of the target equipment by combining with detection on the running condition of the easily damaged parts in the target gear box.

According to the method, real-time oil sample monitoring is carried out on underground equipment, a prediction model and a real-time monitoring model are established for the conditions of the oil sample at each time interval by combining a neural network algorithm, so that the comparison condition of the oil liquid state at each moment is generated, the analogy is conveniently carried out by workers, the oil sample states at different moments and different moments of the real-time model and the prediction model are observed, the reason analysis is carried out on the internal fault occurrence condition of the gear box according to the oil liquid detection condition, the oil sample result is brought into a fault diagnosis model, the fault reason and position are analyzed in the corresponding fault model, and a series of problems in the manual inspection process are avoided.

Optionally, in S1, the prediction model includes an original prediction model and a real-time prediction model;

the original prediction model forms a model without fault in a theoretical normal working state on the basis of a nameplate of a corresponding equipment gear box, factory specification data and initial data of an oil sample;

and the real-time prediction model calls the real-time model, predicts the oil model at the next time point according to the oil data detected in the S3, and takes the prediction condition that the gearbox of the target equipment works normally without fault.

Further, when the real-time prediction model detects an initial time point, the original prediction model data is called, and oil sample model prediction of the next time point is carried out in sequence.

Further, when the step S2 is performed, the data call processing rule for synchronously building the real-time model includes:

real-time calling is carried out on oil sample detection data to generate a real-time oil sample model, and the model generated at the moment is temporarily stored for calling of a prediction model;

after the oil sample detection data is called at each moment point, the generated real-time oil sample model updates the temporarily stored oil sample model at the last moment point.

Further, in the step S3, the detection of the oil sample includes detection of the oil abrasive particle concentration, viscosity, water content, density, and vibration signal.

Further, when the step S3 is performed, the processing of the detected signal includes:

s31, preprocessing data, namely denoising, amplifying, normalizing and A/D converting signals generated by oil sample detection;

s32, performing threshold judgment, false alarm elimination and vertical integration processing on the data in the S31;

and S33, transmitting the data in the S32 to a real-time model, performing data replacement on the real-time model of the oil sample, and synchronously transmitting the data to a fault diagnosis model so that the fault diagnosis model can perform fault diagnosis on the target equipment.

Further, the fault diagnosis model is a model which is established on the basis of a deep learning algorithm and used for carrying out data analysis processing on the abrasive particle concentration, the viscosity, the water content and the vibration signal of the oil liquid of the gearbox.

The application also provides a mine equipment fluid detecting system, including following module:

the system comprises a first module, a second module and a third module, wherein the first module is used for establishing a prediction model for oil of a gearbox of target equipment and predicting the oil state of the gearbox of the target equipment at the next time point by operating the gearbox of the target equipment in the current state;

the prediction model comprises an original prediction model and a real-time prediction model;

the real-time prediction model calls the real-time model, oil data detected in real time is used for predicting the oil model at the next time point, and the prediction condition is that no fault occurs when a gearbox of target equipment works normally;

the second unit is used for calling original prediction model data when the real-time prediction model detects an initial time point and sequentially predicting the oil sample model at the next time point;

the second module is used for establishing a real-time model for oil of the gearbox of the target equipment and carrying out real-time oil state simulation on the gearbox of the target equipment in the current state;

and a third unit, which is used for synchronously establishing a data calling processing rule of the real-time model in the second module, and comprises the following steps:

calling the oil sample detection data in real time to generate a real-time oil sample model, and temporarily storing the model generated at the moment for calling the prediction model;

after the oil sample detection data is called at each moment point, the generated real-time oil sample model updates the oil sample model temporarily stored at the last moment point;

the third module is used for carrying out oil detection on the target equipment gearbox, and bringing each detected item of data into the prediction model and the real-time model to carry out gearbox state simulation and data processing;

the fourth unit is used for preprocessing data, and carrying out denoising, amplification, normalization and A/D conversion processing on signals generated by oil sample detection;

the fifth unit is used for carrying out threshold judgment, false alarm elimination and vertical integration processing on the data in the fourth unit;

the sixth unit is used for transmitting the data in the fifth unit to the real-time model, performing data replacement on the real-time model of the oil sample, and synchronously transmitting the data to the fault diagnosis model so that the fault diagnosis model can perform fault diagnosis on the target equipment;

the fourth module is used for establishing a user interaction interface, transmitting the prediction data of the prediction model and the detection data of the real-time model to the user interaction interface, and performing graphical or textual processing on the data according to the user requirements;

and a fifth module for constructing a mine equipment fault analysis model, extracting and processing characteristic values of data in the third module and bringing the characteristic values into the fault diagnosis model, and performing fault analysis on the gearbox of the target equipment by combining detection on the running condition of the easily damaged parts in the target gearbox.

The application also provides a storage medium, wherein instructions executable by the mine equipment oil detection system are stored, and the instructions are executed by a processor included in the mine equipment oil detection system and are used for realizing the mine gearbox oil detection and analysis method.

Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.

Drawings

The above and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:

FIG. 1 is a schematic diagram of method steps for a method for detecting and analyzing oil in a mine gearbox according to the present invention;

FIG. 2 is a detailed step diagram of the oil detection and analysis method for the mine gearbox according to the invention in S3.

Detailed Description

Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.

The invention provides a method for detecting and analyzing oil in a mine gearbox, which is explained in detail with reference to fig. 1 and 2.

A method for detecting and analyzing oil in a mine gearbox comprises the following steps:

s3, carrying out oil detection on the target equipment gearbox, and bringing detected data into a prediction model and a real-time model to carry out gearbox state simulation and data processing;

In S1, the prediction model comprises an original prediction model and a real-time prediction model;

forming a model without fault in a theoretical normal working state by using a nameplate of a corresponding equipment gear box, factory specification data and initial data of an oil sample as a basis for the original prediction model;

and calling the real-time model by the real-time prediction model, predicting the oil model at the next time point according to the oil data detected in the step S3, and predicting the condition that the gearbox of the target equipment works normally without failure. And when the real-time prediction model detects the initial time point, calling original prediction model data, and sequentially predicting the oil sample model at the next time point.

When the step S2 is performed, the data call processing rule for synchronously establishing the real-time model includes:

In the step S3, the detection of the oil sample includes detection of the oil abrasive particle concentration, viscosity, water content, density, and vibration signal. When the step S3 is performed, the processing of the detected signal includes:

s31, data preprocessing, namely denoising, amplifying, normalizing and A/D conversion processing are carried out on signals generated by oil sample detection;

And in S5, the fault diagnosis model is a model which is established on the basis of a deep learning algorithm and is used for carrying out data analysis processing on the oil abrasive particle concentration, the viscosity, the water content and the vibration signal of the gearbox.

Furthermore, in order to update and monitor the fault diagnosis model in real time, an internal network chain used for uploading the fault diagnosis model is established, the fault diagnosis model is connected into the internal network chain, a fault model update partition and a user operation communication partition which are connected with the network chain are established at the same time, the fault model diagnosis partition can store each fault model transmitted to the partition, an oil sample real-time model of each mine device and oil sample detection data of each mine device are called in real time, comparison diagnosis is carried out on the fault model in the fault model diagnosis partition and an existing fault model through extraction of characteristic values of the detection data, and the position of a fault cause is analyzed; ground staff can monitor large-area underground equipment in real time through the user operation communication subarea, and can update, increase, delete and the like the fault model updating subarea in real time through the user operation communication subarea.

In the fault model updating subarea, a characteristic value extraction area, a model storage area, a model calling comparison area and a model adjusting area are specifically included, the model storage area stores corresponding models for each monitored target device, when each oil sample data of the oil sample real-time model is transmitted to the fault model updating subarea, the characteristic value extraction area performs data analysis on the oil sample, a characteristic value required by a fault model is extracted, after the characteristic value is extracted, the characteristic value is called by the model calling comparison area to the characteristic extraction area to be analyzed, the fault model addressing of the corresponding device is performed in the model storage area according to the characteristic value, each fault model comprises a value range of the characteristic value, a fault model with the characteristic value in the characteristic value range of the next time is selected as a main fault model during addressing, the fault model corresponding to the characteristic value is obtained, the characteristic value range in the fault model is compared with the characteristic value obtained next time, and fault reasons, positions and solutions contained in the fault model are fed back to a user operation communication subarea for a worker to perform fault removal operation according to the method;

in some embodiments, and the fault model whose feature value range is closest to the next feature value, when addressing the fault model according to the feature value, in addition to selecting the fault model whose feature value is within the feature value range as the main fault model, the fault model whose feature value range is closest to the next feature value needs to be addressed for multiple times in proximity to serve as the candidate fault models, where the number of the candidate fault models is greater than or equal to 2, and when the fault cause, the location, and the solution corresponding to the main fault model are wrong or vulnerable, a worker may observe the candidate fault models in the operation user communication partition, and perform combined correction and complement on information included in the main fault model by referring to the fault cause, the location, and the solution included in the candidate fault models. Meanwhile, after the fault repair is completed, the model adjusting and teaching area can temporarily store the use data of the fault model, a worker can update and adjust the model through the user operation communication subarea, and after the model adjustment and update is completed, the new model is stored in the model storage area so as to be called when the follow-up fault occurs.

A neural network algorithm is adopted in a fault model updating partition, a fault diagnosis model is self-perfected and updated through real-time monitoring of detection conditions of all mine equipment, effective early warning and detection are carried out on faults of a gear box in the mine equipment by combining with mechanical equipment characteristics, meanwhile, various fault signal models are analyzed, and a residual life prediction model of key parts of a coal mining machine is constructed by adopting a method of combining data denoising, characteristic extraction, residual life prediction model construction, prediction model performance verification and residual life prediction of the key parts based on a deep learning theory according to the actual operation condition and failure mode of the easily damaged parts of fully mechanized mining face equipment.

The invention also provides a system for detecting the oil of the mine equipment, which comprises the following modules:

the real-time prediction model calls the real-time model, oil data detected in real time is used for predicting the oil model at the next time point, and the prediction condition is that no fault occurs when a gearbox of the target equipment works normally;

the second unit is used for calling original prediction model data when the real-time prediction model detects an initial time point and sequentially carrying out oil sample model prediction of the next time point;

after the oil sample detection data are called at each time point, the generated real-time oil sample model updates the temporarily stored oil sample model at the last time point;

and the fifth module is used for constructing a mine equipment fault analysis model, extracting characteristic values of data in the third module, bringing the characteristic values into the fault diagnosis model, and carrying out fault analysis on the gearbox of the target equipment by combining detection on the running condition of easily damaged parts in the target gearbox.

The invention also provides a storage medium, wherein instructions executable by the mine equipment oil detection system are stored, and the instructions are executed by a processor included in the mine equipment oil detection system and are used for realizing any one of the above methods for detecting and analyzing the oil in the mine gearbox.

In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.

Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "a plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.

While embodiments of the invention have been shown and described, it will be understood by those of ordinary skill in the art that: various changes, modifications, substitutions and alterations can be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for detecting and analyzing oil in a mine gearbox is characterized by comprising the following steps:

s2, establishing a real-time model for oil of the gearbox of the target equipment, and performing real-time oil state simulation on the gearbox of the target equipment in the current state;

2. The method for detecting and analyzing the oil in the mine gearbox according to claim 1, wherein in S1, the prediction model comprises an original prediction model and a real-time prediction model;

and the real-time prediction model calls the real-time model, predicts the oil model at the next time point according to the oil data detected in the step S3, and takes the prediction condition that the gearbox of the target equipment works normally without failure.

3. The method for detecting and analyzing oil in a mine gearbox according to claim 2, wherein the real-time prediction model calls original prediction model data when detecting an initial time point, and sequentially performs oil sample model prediction at a next time point.

4. The method for detecting and analyzing the oil in the mine gearbox according to claim 1, wherein in the step S2, the data call processing rules for synchronously establishing the real-time model comprise:

5. The method for detecting and analyzing the oil in the mine gearbox according to claim 1, wherein in the step S3, the detection of the oil sample comprises detection of abrasive grain concentration, viscosity, water content, density and vibration signals of the oil.

6. The method for detecting and analyzing oil in a mine gearbox according to claim 5, wherein the step S3 is performed by processing the detected signals, and comprises the following steps:

7. The method for detecting and analyzing mine gearbox oil according to claim 1, wherein the fault diagnosis model is a model which is built on the basis of a deep learning algorithm and is used for carrying out data analysis processing on the abrasive grain concentration, viscosity, water content and vibration signals of the gearbox oil.

8. A mine equipment oil detection system is characterized by comprising the following modules:

the original prediction model forms a model without fault under a theoretical normal working state on the basis of a nameplate of a corresponding equipment gearbox, factory specification data and initial data of an oil sample;

a third unit, configured to synchronously establish a data call processing rule of the real-time model in the second module, where the data call processing rule includes:

the fifth unit is used for performing threshold judgment, false alarm elimination and vertical integration processing on the data in the fourth unit;

9. A storage medium having stored thereon instructions executable by the system of claim 8, wherein the instructions when executed by a processor comprised by the system of claim 8 are for implementing a method of mine gearbox oil detection and analysis as claimed in any one of claims 1 to 7.