CN111521259B - Grinding machine detection method, device and equipment - Google Patents
Grinding machine detection method, device and equipment Download PDFInfo
- Publication number
- CN111521259B CN111521259B CN202010367485.3A CN202010367485A CN111521259B CN 111521259 B CN111521259 B CN 111521259B CN 202010367485 A CN202010367485 A CN 202010367485A CN 111521259 B CN111521259 B CN 111521259B
- Authority
- CN
- China
- Prior art keywords
- mill
- vibration
- real
- data
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H17/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the preceding groups
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B02—CRUSHING, PULVERISING, OR DISINTEGRATING; PREPARATORY TREATMENT OF GRAIN FOR MILLING
- B02C—CRUSHING, PULVERISING, OR DISINTEGRATING IN GENERAL; MILLING GRAIN
- B02C23/00—Auxiliary methods or auxiliary devices or accessories specially adapted for crushing or disintegrating not provided for in preceding groups or not specially adapted to apparatus covered by a single preceding group
Landscapes
- Engineering & Computer Science (AREA)
- Food Science & Technology (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The disclosure provides a mill detection method, a mill detection device and mill detection equipment. The detection method of the mill comprises the following steps: acquiring a plurality of groups of real-time vibration signals and a group of real-time operation data from a plurality of vibration sensors at preset positions of the mill, wherein the real-time operation data comprises a plurality of preset kinds of operation data; extracting multiple groups of vibration data to be detected from the real-time vibration signals according to preset conditions, wherein the vibration data to be detected comprises the frequency, amplitude and phase of the vibration signals to be detected which meet the preset conditions in the real-time vibration signals; inputting vibration data to be detected and real-time operation data into a preset neural network model so as to obtain detection parameters of the mill output by the preset neural network model, wherein the detection parameters comprise the filling rate of grinding media filled in the mill, the comprehensive filling rate, the height of a lifting strip and the abrasion degree of a lining plate. The embodiment of the disclosure can directly output the detection parameters of the mill according to the real-time vibration signals and the real-time operation data of the mill.
Description
Technical Field
The disclosure relates to the technical field of data processing, in particular to a mill detection method, device and equipment.
Background
Mills are an important apparatus for grinding mineral materials and are classified into autogenous mills, semi-autogenous mills and ball mills. For the grinding machine, the operation parameters are related to the stability of ore grinding operation, and the production efficiency and the product quality of the grinding machine are influenced. For example, in the case of a semi-autogenous mill, the filling rate of the grinding media, the overall filling rate, the height of the lifting bars, and the degree of wear of the lining plates are important indicators of the grinding operation. The grinding machine is a multi-parameter, nonlinear and strongly coupled system, and the failure to effectively detect the operation parameters brings great difficulty to load control.
In the related technology, a mill detection method for transmitting mill vibration signals to display equipment through a vibration sensor is available, but the method needs mill responsible personnel to monitor the operation condition of the mill according to numerical values, and only can judge whether the mill needs to be overhauled through experience, so that the efficiency and accuracy rate are required to be improved, the labor cost and time cost are high, and the requirements on the efficiency, accuracy rate and cost of mill detection and control cannot be met.
It is to be noted that the information disclosed in the above background section is only for enhancement of understanding of the background of the present disclosure, and thus may include information that does not constitute prior art known to those of ordinary skill in the art.
Disclosure of Invention
The invention aims to provide a grinding machine detection method, a grinding machine detection device and grinding machine detection equipment, which are used for overcoming the problem that whether a grinding machine needs to be overhauled or not can not be accurately and timely judged due to the limitations and the defects of the related technology at least to a certain extent.
According to a first aspect of the present disclosure, there is provided a mill detection method comprising: acquiring a plurality of groups of real-time vibration signals and a group of real-time operation data from a plurality of vibration sensors at preset positions of the mill, wherein the real-time operation data comprises a plurality of preset kinds of operation data; extracting multiple groups of vibration data to be detected from the real-time vibration signals according to preset conditions, wherein the vibration data to be detected comprises the frequency, amplitude and phase of the vibration signals to be detected which meet the preset conditions in the real-time vibration signals; inputting vibration data to be detected and real-time operation data into a preset neural network model to acquire detection parameters of a mill output by the preset neural network model, wherein the preset type operation data comprise ore property parameters, loaded grinding medium property parameters, mill rotating speed, treatment capacity, power, hard rock quantity and axial pressure, and the detection parameters comprise grinding medium filling rate, comprehensive filling rate, lifting strip height and lining plate abrasion degree loaded by the mill.
In an exemplary embodiment of the present disclosure, extracting a plurality of sets of vibration data to be detected from a real-time vibration signal according to a preset condition includes: preprocessing the real-time vibration signal to generate a first frequency domain signal; extracting frequency domain signals corresponding to a plurality of preset frequency bands from the first frequency domain signal to generate a second frequency domain signal; selecting a plurality of to-be-detected vibration signals which meet preset conditions from the second frequency domain signals; and generating vibration data to be detected according to the frequency, amplitude and phase of the vibration signals to be detected.
In an exemplary embodiment of the present disclosure, preprocessing the real-time vibration signal to generate a first frequency domain signal comprises: determining optimal sampling data according to the real-time vibration signal; capturing a plurality of sampling data from the optimal sampling data through a preset sampling time window; the plurality of sampled data is fourier transformed to generate a first frequency domain signal.
In an exemplary embodiment of the present disclosure, determining the optimal sampling data from the real-time vibration signal includes: and extracting the real-time vibration signals of which the amplitude is within a first preset range and the frequency is within a second preset range from the real-time vibration signals into optimal sampling data through filtering.
In an exemplary embodiment of the present disclosure, the preset condition is a vibration signal corresponding to an amplitude extremum in the vibration signal from the preset vibration sensor in each preset frequency band.
In an exemplary embodiment of the present disclosure, the preset frequency band includes a frequency band which is determined in advance according to the detection parameter type and is related to the detection parameter type.
In an exemplary embodiment of the present disclosure, further comprising: generating observation vibration signals according to a plurality of vibration signals to be detected, wherein the observation vibration signals comprise single-frequency time domain vibration signals and/or frequency mixing time domain vibration signals; and displaying and observing the vibration signal.
In an exemplary embodiment of the present disclosure, the neural network model is preset as a long-term memory neural network model.
In an exemplary embodiment of the present disclosure, the training process of the preset neural network model includes: acquiring m groups of real-time operation data of the mill under m groups of detection parameters and m x n groups of real-time vibration signals from n vibration sensors, wherein m and n are integers greater than 1; acquiring m × p vibration signals to be detected according to the m × n groups of real-time vibration signals, wherein p is a natural number; extracting the frequency, amplitude and phase of the m × p vibration signals to be detected to generate m × p groups of vibration data to be detected; extracting a plurality of sample subsets according to m x p groups of vibration data to be detected, m groups of real-time operation data and m groups of detection parameters, wherein the plurality of sample subsets comprise a training data set and a test data set; and training and generalizing the preset neural network according to the plurality of sample subsets.
In an exemplary embodiment of the present disclosure, the mill comprises a semi-autogenous mill, an autogenous mill and a ball mill, and the preset positions comprise a driving motor bearing seat, a pinion bearing seat, a mill feeding port bearing seat, a mill discharging port bearing seat and a mill cylinder wall.
According to a second aspect of the present disclosure, there is provided a mill detecting device comprising: the data acquisition module is used for acquiring a plurality of groups of real-time vibration signals and a group of real-time operation data from a plurality of vibration sensors at preset positions of the mill, and the real-time operation data comprises a plurality of preset types of operation data; the data extraction module is used for extracting a plurality of groups of vibration data to be detected from the real-time vibration signals according to preset conditions, wherein the vibration data to be detected comprises the frequency, the amplitude and the phase of the vibration signals to be detected which meet the preset conditions in the real-time vibration signals; the parameter acquisition module is set to input the vibration data to be detected and the real-time operation data into a preset neural network model so as to acquire detection parameters of the mill output by the preset neural network model, wherein the preset type operation data comprise ore property parameters, loaded grinding medium property parameters, mill rotating speed, processing capacity, power, hard rock quantity and axial pressure, and the detection parameters comprise grinding medium filling rate, comprehensive filling rate, lifting strip height and lining plate abrasion degree loaded by the mill.
According to a third aspect of the present disclosure, there is provided a mill detecting apparatus comprising: the vibration sensors are arranged at a plurality of preset positions of the grinding machine and used for detecting real-time vibration signals of the grinding machine; the parameter acquisition device is used for acquiring real-time operation data, and the real-time operation data comprises a plurality of preset kinds of operation data; the signal transmission device is coupled with the vibration sensor and the parameter acquisition device and is used for transmitting the real-time vibration signal and the real-time operation data to the signal processing device; the signal processing device is coupled with the signal transmission device and used for executing the method to output detection parameters of the mill, wherein the preset type operation data comprises ore property parameters, loaded grinding medium property parameters, mill rotating speed, processing capacity, power, hard rock quantity and axial pressure, and the detection parameters comprise the grinding medium filling rate, comprehensive filling rate, lifting bar height and lining plate abrasion degree of the mill loading; and the display device is coupled with the signal processing device and used for displaying the detection parameters.
According to a fourth aspect of the present disclosure, there is provided an electronic device comprising: a memory; and a processor coupled to the memory, the processor configured to perform the method of any of the above based on instructions stored in the memory.
According to a fifth aspect of the present disclosure, there is provided a computer readable storage medium having a program stored thereon, which when executed by a processor, implements the method as defined in any one of the above.
According to the mill detection method provided by the embodiment of the disclosure, the real-time vibration signals and the real-time operation data of the mill are subjected to data extraction, and the extracted data are input into the trained preset neural network model, so that the detection parameters of the mill can be directly obtained according to the real-time vibration signals and the real-time operation data of the mill, the control accuracy of the mill is assisted to be improved, and the operation monitoring efficiency of the mill is improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and together with the description, serve to explain the principles of the disclosure. It is to be understood that the drawings in the following description are merely exemplary of the disclosure, and that other drawings may be derived from those drawings by one of ordinary skill in the art without the exercise of inventive faculty.
FIG. 1 is a flow chart of a mill detection method in an exemplary embodiment of the present disclosure.
Fig. 2 is a schematic view of a vibration sensor mounting position in an exemplary embodiment of the present disclosure.
FIG. 3 is a sub-flow diagram of a mill detection method in an exemplary embodiment of the present disclosure.
Fig. 4 is a training flow diagram of a predictive neural network model in an exemplary embodiment of the disclosure.
FIG. 5 is a block diagram of a mill detection device in an exemplary embodiment of the present disclosure.
FIG. 6 is a block diagram of a mill testing apparatus in an exemplary embodiment of the present disclosure.
FIG. 7 is a block diagram of an electronic device in an exemplary embodiment of the disclosure.
Detailed Description
Example embodiments will now be described more fully with reference to the accompanying drawings. Example embodiments may, however, be embodied in many different forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the disclosure. One skilled in the relevant art will recognize, however, that the subject matter of the present disclosure can be practiced without one or more of the specific details, or with other methods, components, devices, steps, and the like. In other instances, well-known technical solutions have not been shown or described in detail to avoid obscuring aspects of the present disclosure.
Further, the drawings are merely schematic illustrations of the present disclosure, in which the same reference numerals denote the same or similar parts, and thus, a repetitive description thereof will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different networks and/or processor devices and/or microcontroller devices.
The following detailed description of exemplary embodiments of the disclosure refers to the accompanying drawings.
FIG. 1 is a flow chart of a mill detection method in an exemplary embodiment of the present disclosure. Referring to FIG. 1, mill detection method 100 may include:
step S1, acquiring a plurality of groups of real-time vibration signals and a group of real-time operation data from a plurality of vibration sensors at preset positions of the mill, wherein the real-time operation data comprises a plurality of preset types of operation data;
step S2, extracting multiple groups of vibration data to be detected from the real-time vibration signals according to preset conditions, wherein the vibration data to be detected comprises the frequency, amplitude and phase of the vibration signals to be detected which meet the preset conditions in the real-time vibration signals;
and step S3, inputting vibration data to be detected and real-time operation data into a preset neural network model to obtain detection parameters of the mill output by the preset neural network model, wherein the preset type operation data comprise ore property parameters, loaded grinding medium property parameters, mill rotating speed, treatment capacity, power, hard rock quantity and axial pressure, and the detection parameters comprise grinding medium filling rate, comprehensive filling rate, lifting strip height and lining plate abrasion degree loaded by the mill.
According to the mill detection method provided by the embodiment of the disclosure, the data extraction is carried out on the real-time vibration signal and the real-time operation data of the mill, and the extracted data is input into the trained neural network model, so that the detection parameters of the mill can be directly obtained according to the real-time vibration signal and the real-time operation data of the mill, the control accuracy of the mill is assisted to be improved, and the monitoring efficiency of the operation of the mill is improved.
Next, each step of the mill detection method 100 will be described in detail.
In step S1, a plurality of sets of real-time vibration signals from a plurality of vibration sensors at predetermined locations of the mill and a set of real-time operational data are obtained, the real-time operational data including a plurality of predetermined types of operational data.
Fig. 2 is a schematic view of a mounting position of a vibration sensor in an embodiment of the present disclosure.
Referring to fig. 2, in the disclosed embodiment, the mill 200 may be, for example, a semi-autogenous mill, and vibration sensors may be installed at one or more locations of a drive motor bearing housing 21, a pinion bearing housing 22, a mill inlet bearing housing 23, a mill outlet bearing housing 24, a mill barrel wall 25, and the like, of the semi-autogenous mill. The vibration sensor may be a wireless or wired vibration sensor, a single-axis or multi-axis sensor, an acceleration or velocity sensor, and the present disclosure is not particularly limited thereto.
In the running process of the semi-autogenous mill, the steel balls and the ores are lifted to a certain height by the rotating roller under the action of centripetal force and friction force of the roller body and then fall along a parabola under the action of self gravity. The impacts between the steel balls, ore and the drum liner and lifter bars, between the steel balls, and between the steel balls and the ore can cause vibrations of the semi-autogenous mill that propagate along the drum and bearings. Therefore, the vibration characteristics of the semi-autogenous mill can be measured by installing the measuring devices on the semi-autogenous mill cylinder body and the bearing, and the running state of the semi-autogenous mill can be obtained by analyzing the vibration characteristics.
In the embodiment shown in fig. 2, when the vibration sensors are installed at multiple positions of the dual-drive semi-autogenous mill, the installation directions of the vibration sensors need to be the same, that is, the spatial directions (x \ y \ z axis orientations) of the vibration sensors need to be the same.
When the mill runs, n vibration sensors (n is more than or equal to 1) can output n groups of real-time vibration signals, and the real-time vibration signals can be transmitted to signal processing equipment, such as a computer, through a wired or wireless signal transmission device.
In addition, in the disclosed embodiment, real-time mill operation data is also collected to assist in determining mill detection parameters along with the real-time vibration signals.
In one embodiment, the real-time operation data comprises a plurality of preset species operation data, the preset species operation data may comprise one or more of ore property parameters, charged grinding medium property parameters, mill rotation speed, treatment capacity, power, hard rock amount and shaft pressure, for example, in other embodiments, the preset species operation data may also comprise other species operation data, and the preset species operation data may be set by a person skilled in the art according to actual conditions. The preset type of operation data may be from a human-computer interaction interface, that is, obtained through a data input operation of a user on the human-computer interaction interface, or may be from a plurality of measuring instruments, such as a pressure sensor, an infrared sensor, a mill controller, a voltmeter, an ammeter, and the like, which is not limited in this disclosure.
It can be understood that the type of the real-time operation data obtained in this step is completely consistent with the type of the real-time operation data subsequently used for training the preset data model.
In step S1, the number of vibration sensors and the number of preset kinds of operation parameters may each be, for example, a natural number within 100.
In step S2, multiple sets of vibration data to be detected are extracted from the real-time vibration signals according to the preset conditions, where the vibration data to be detected includes the frequencies, amplitudes, and phases of the multiple vibration signals to be detected that meet the preset conditions in the real-time vibration signals.
Fig. 3 is a sub-flowchart of step S2 in the embodiment of the present disclosure.
Referring to fig. 3, in an exemplary embodiment of the present disclosure, step S2 may include:
step S21, preprocessing the real-time vibration signal to generate a first frequency domain signal;
step S22, extracting frequency domain signals corresponding to a plurality of preset frequency bands from the first frequency domain signal to generate a second frequency domain signal;
step S23, selecting a plurality of vibration signals to be detected which meet preset conditions from the second frequency domain signals;
and step S24, generating vibration data to be detected according to the frequency, amplitude and phase of the vibration signals to be detected.
In one embodiment, in step S21, the optimal sampling data may be first determined according to the real-time vibration signal, a plurality of sampling data may be extracted from the optimal sampling data through a preset sampling time window, and then the plurality of sampling data may be fourier transformed to generate the first frequency domain signal.
In one embodiment, the real-time vibration signal with the amplitude within a first preset range and the frequency within a second preset range in the real-time vibration signal may be extracted as the optimal sampling data through a filtering means. The first predetermined range and the second predetermined range may be, for example, a range of amplitude and a range of frequency determined by analytical calculations and experiments that are most representative of the operating conditions of the mill, and the disclosure is not limited thereto.
When the frequency of the real-time vibration signal is between 1500Hz and 5000Hz, the real-time vibration signal can be preliminarily filtered by adopting a band-pass filter, and the filtering method can be various as long as the effects of retaining effective data, reducing errors, reducing calculated amount and the like can be achieved. Then, a rectangular window (a preset sampling time window) may be added to the filtered time domain signal, and various windowing methods are possible, which is not specifically limited by the present disclosure. Finally, the optimal sampling signal can be converted from a time domain signal to a frequency domain signal through a fast fourier transform mode or other modes to obtain a first frequency domain signal.
The different manifestations of the signal in the time and frequency domains reflect two different sides of the signal. The observation of the analysis signal in the time domain is more perceptual, so that the understanding is easy, and the observation of the analysis signal in the frequency domain is more reasonable and difficult to understand, but deeper and more essential things can be obtained. Therefore, the optimal sampling signal in the time domain can be converted into a frequency domain signal through a fast Fourier transform mode or other modes so as to analyze and extract frequency data.
In step S22, in one embodiment, the plurality of preset frequency bands include a frequency band associated with the detection parameter type determined in advance according to the detection parameter type.
The factors causing vibration of the semi-autogenous mill are various, and the vibration caused by vibration of a transmission system, vibration caused by asymmetric moment of inertia of each part, vibration caused by equipment installation error and the like are removed from vibration generated among steel balls, ore and roller lining plates, steel balls and ore. Therefore, in some embodiments, the frequency band related to the detected parameter can be determined according to analysis, calculation and experiments to reduce the extraction range of the vibration data to be detected and improve the efficiency and accuracy of determining the vibration data to be detected. One or more of the detection parameters may correspond to one or more predetermined frequency bands, for example, if it is determined from analytical calculations or experiments that a signal indicating the detection parameter B is most likely to occur in the frequency bands a1, a2, A3, the vibration signals corresponding to the frequency bands a1, a2, A3 may be added to the second frequency domain signal when it is determined that the detection parameter B is to be obtained through the predetermined neural network model. And after frequency domain signals corresponding to a plurality of frequency bands are selected from the first frequency domain signals, second frequency domain signals corresponding to the detection parameters are obtained.
In step S23, the preset condition may be, for example, a signal corresponding to an amplitude extreme value in the vibration signals from the preset vibration sensors in each preset frequency band.
Since each vibration sensor has different test sensitivity to different frequency bands, it can be set to select the vibration signal from which vibration sensor is used as the vibration signal representing the frequency band. For example, if the vibration sensor No. 1 or 3 is sensitive to the a1 frequency band (i.e., the detection accuracy is high), and the vibration sensor No. 2 is sensitive to the a2 frequency band, in the second frequency domain signal, corresponding to the a1 frequency band, only the vibration signals from the vibration sensor No. 1 and the vibration sensor No. 3 are selected to be detected, and corresponding to the a2 frequency band, only the vibration signals from the vibration sensor No. 2 are selected to be detected. The number of the vibration sensors corresponding to one frequency band can be one or multiple, and the vibration sensors can be set by a person skilled in the art according to actual working condition tests.
In one embodiment, a vibration signal to be detected can be selected for each frequency band, that is, a vibration signal corresponding to an amplitude extremum in a vibration signal from a preset vibration sensor corresponding to the frequency band is used as the vibration signal to be detected.
In another embodiment, two or more vibration signals to be detected may be selected for each frequency band. For example, if the second frequency domain signal includes k frequency bands (k may be a natural number smaller than 100, for example), i vibration signals to be detected may be extracted for the k frequency bands, where i is a natural number larger than k and smaller than n × k. Namely, more than k vibration signals to be detected are extracted from k frequency bands.
In other embodiments, the selection criterion of the vibration signal to be detected may be other criteria as long as the selected vibration signal can accurately reflect the characteristics related to the detection parameters.
In step S24, in addition to adding the amplitude, phase and frequency of each vibration signal to be detected to the vibration data to be detected, each vibration signal to be detected corresponds to a group of vibration data to be detected, and other characteristics of the vibration signal to be detected may also be added to the vibration data to be detected, which is not limited in this disclosure.
In some embodiments, the optimal sampling data may not be determined, and the first frequency domain signal is generated according to all real-time vibration signals, and then the processing in steps S22 to S23 is performed to obtain the vibration data to be detected.
In step S3, the vibration data to be detected and the real-time operation data are input into the preset neural network model to obtain the detection parameters of the mill output by the preset neural network model.
In an exemplary embodiment of the present disclosure, the preset neural network model may be, for example, a Long Short-Term Memory (LSTM) neural network model. The long-time memory neural network model is a time cycle neural network model, is suitable for processing and predicting important events with very long intervals and delays in a time sequence, can memorize numerical values with indefinite time length, and is provided with a gate in a block for determining whether input data is important enough to be memorized and can not be output. According to analysis, calculation and experiments, the applicant finds that the long-term memory neural network model can more accurately output detection parameters of the grinding machine, so that in the embodiment of the disclosure, the long-term memory neural network model is used for data processing.
In one embodiment, the sensed parameters may include, for example, one or more of a mill media fill rate, a composite fill rate, a lifter bar height, and a liner wear level of the mill charge. In other embodiments, a person skilled in the art can determine the type of the detection parameter according to the actual situation and the type of the mill, and determine and select the frequency band corresponding to the vibration signal to be detected according to the type of the detection parameter.
In the embodiment of the disclosure, the mill detection method may further include a training process for the preset neural network model.
FIG. 4 is a flow chart of training a neural network model according to an embodiment of the present disclosure.
Referring to fig. 4, the training process 400 may include:
step S41, acquiring m groups of real-time operation data of the mill under m groups of detection parameters and m x n groups of real-time vibration signals from n vibration sensors, wherein m and n are integers greater than 1;
step S42, acquiring m x p vibration signals to be detected according to the m x n groups of real-time vibration signals, wherein p is a natural number;
step S43, extracting the frequency, amplitude and phase of m × p vibration signals to be detected to generate m × p groups of vibration data to be detected;
step S44, extracting a plurality of sample subsets according to m × p groups of vibration data to be detected, m groups of real-time operation data and m groups of detection parameters, wherein the plurality of sample subsets comprise a training data set and a test data set;
and step S45, training and generalizing the preset neural network according to the plurality of sample subsets.
In steps S42 and S43, the method for extracting the first frequency domain signal and the second frequency domain signal according to the n sets of real-time vibration signals collected at the same time, and further extracting the p sets of vibration data to be detected from the second frequency domain signal has been described in detail in the embodiment shown in fig. 3, and is not described herein again.
In the process of training the neural network model, the training samples are data obtained after processing multiple groups of real-time vibration signals obtained in test experiments of multiple working conditions (different mill-filled grinding medium filling rates, comprehensive filling rates, lifting strip heights and lining plate abrasion degrees) and multiple groups of real-time operation data (such as ore property parameters, filled grinding medium property parameters, mill rotating speeds, processing amounts, power, hard rock volumes and shaft pressure) at corresponding moments. The pre-processing of the real-time vibration signal may include, for example, the operations of filtering, windowing, fourier transform, extracting amplitude, frequency, phase data, and the like in steps S21 to S23, and may also include a processing step of first determining an optimal set of sampling signals in step S21, and performing the processing on the optimal sampling signals as shown in the description in steps S21 to S23. The real-time vibration signals in the training samples can be processed in accordance with the real-time vibration signals during real-time detection
In the embodiment of the disclosure, the long-term and short-term memory neural network may be configured to include an input layer, a hidden layer, and an output layer. The number of input layer neurons can be determined from the number of input data; the number of hidden layer neurons can be calculated by reference to empirical formula
Trial and error determination, where n is the number of hidden layer nodes, niFor the number of input nodes, njI is a constant between 1 and 10 and is the number of output nodes; the output layer may include, for example, 4 neurons to represent the grinding media fill rate, the aggregate fill rate, the lifter bar height, the liner wear level, and other parameters of the mill charge, respectively. The activation function may be, for example, a commonly used Sigmoid function.
Next, the ADAM algorithm may be used to train the long-term and short-term memory neural network, and the algorithm combines the past accumulated momentum and gradient, and corrects the accumulated momentum by considering the momentum to reach the minimum point quickly.
In some embodiments, the preset neural network model may be configured to be continuously automatically modified during use. For example, a feedback entry may be provided for receiving measured values of the detection parameters from multiple sources, comparing the measured values of the detection parameters with the detection parameters output by the model, and triggering parameter modification of the model when the deviation is greater than a threshold value. The parameter correction can be manual correction by reminding maintenance personnel through reminding information, and can also be automatic correction of model parameters according to feedback detection parameter values. The multiple sources of the detected parameter measured values may include manual input actions performed by a user through a human-computer interaction interface after manual measurement, or feedback values automatically output by multiple instruments in the maintenance and repair process.
The related settings of the types of the preset neural network models are only examples, and in practical application, other types of neural network models and corresponding hierarchical settings may be used as long as accurate parameters can be output through training.
In one embodiment of the disclosure, the detection parameters output by the preset neural network model can be displayed on the human-computer interaction interface in various ways, so that a mill maintenance worker can judge whether to carry out field maintenance on the mill or not in time according to the detection parameters. In addition, a detection parameter alarm threshold value can be set, and when one or more of the detection parameters exceed the threshold value, maintenance prompt information is sent, and a mill maintenance person is reminded to carry out field maintenance. The application of the detection parameters may be various, and the disclosure is not limited thereto.
In yet another embodiment, in addition to displaying the detected parameters on the human-computer interface, vibration signals that can be observed and analyzed can be displayed together. For example, one or more observed vibration signals may be generated from a plurality of suspected vibration signals and then displayed. Wherein the observed vibration signal may comprise a single frequency time domain vibration signal and/or a mixed frequency time domain vibration signal. Namely, the vibration signal to be detected can be reduced into one or more time domain signals in a single-frequency or mixing mode and displayed on a human-computer interaction interface so as to assist a mill maintenance person to observe a representative real-time vibration signal. For example, if there are 3 vibration signals to be detected (at this time, the second frequency domain data may include less than or equal to 3 frequency bands), 3 single-frequency time domain signals may be generated and displayed, so that the 3 vibration signals to be detected may be combined to display 1 mixing time domain signal, and any two of the vibration signals to be detected may be mixed and displayed in the form of mixing time domain signals according to user settings. Wherein, if the real-time vibration data is n groups and the second frequency domain signal includes k frequency bands, the number j of the displayed time domain signals can be smaller than (n × k) |! Because the number of mixing signals is smaller than (n x k) | when a plurality of vibration signals to be examined are used to generate the mixing signals. The reduced time domain signals are time domain vibration data after filtering and decoupling, and are easier to understand by mill maintainers than frequency domain signals.
Although a semi-autogenous mill is taken as an example in the embodiment of the present disclosure, the mill detection method provided by the present disclosure can also be applied to an autogenous mill and a ball mill, and a person skilled in the art can adjust the training data and the output parameters of the neural network model according to the type of the mill applied.
The above method is explained below with an application scenario.
With continued reference to fig. 2, in an embodiment of the present disclosure, the following sets of vibration sensors and associated signal transmission device mounting schemes are provided by way of example:
1. two wired three-axis acceleration sensors and active data sending ends are respectively arranged at the two driving motor bearing seats 21;
2. four wired triaxial acceleration sensors and active data transmitting ends are arranged at two groups of four pinion bearing seats 22;
3. a wired three-axis acceleration sensor and an active data transmitting end are arranged on a main bearing seat 23 of a mill feeding port;
4. a wired three-axis acceleration sensor and an active data transmitting end are arranged on a main bearing seat 24 at the discharge port of the grinding machine;
5. a wireless three-axis acceleration sensor and a remote data transmitting end are arranged on the wall 25 of the grinding machine cylinder, and a data receiving end is arranged within two meters away from the edge of the grinding machine.
The above mounting schemes may exist simultaneously or partially, and the disclosure is not limited thereto.
In the above exemplary embodiment, a band-pass filter may be used to filter multiple sets of real-time vibration signals acquired from multiple vibration sensors, then window the signals, convert the filtered signals into frequency domain signals (first frequency domain signals) by applying fast fourier transform, select a vibration signal to be detected corresponding to an amplitude extremum from the frequency domain signals (second frequency domain signals) corresponding to each vibration sensor, input amplitude, frequency, and phase data of the vibration signal to be detected and real-time operation parameters acquired in the same data acquisition process into a trained long-short time memory neural network model according to a sampling sequence to perform calculation, so as to obtain output values of detection parameters such as ore grinding medium filling rate, comprehensive filling rate, lifting bar height, lining plate wear degree and the like of corresponding mill filling.
The long-term memory neural network model can comprise an input layer, a hidden layer and an output layer, wherein the input layer can comprise 1800 neuron nodes, the hidden layer comprises 50 neuron nodes, and the output layer comprises 4 neuron nodes. The training samples can be set as data and real-time operation parameters after multi-position vibration signal processing, which are obtained in a test experiment of a large number of different ore property parameters, comprehensive filling rates, loaded ore grinding medium property parameters, medium filling rates, mill rotating speeds, processing amounts, power, hard rock volumes, shaft pressures, lining plate wear rates, lifting strip wear rates, lining plate wear rates and lifting strip wear rates, the activation function of the neural network model can be a Sigmoid function, and the training method can be an adaptive matrix estimation (ADAM) optimization algorithm.
In an exemplary embodiment of the present disclosure, the time domain signal of the real-time vibration data may be preprocessed, for example, amplitude is limited, or a certain number or time length of amplitude signals are averaged, or smoothed, and a person skilled in the art may set a preprocessing rule of the signal according to an actual situation, or may not perform the preprocessing, which is not limited in the present disclosure.
The preprocessed or unprocessed real-time vibration data may be filtered, windowed, fourier transformed, and the like to generate a first frequency domain signal, and frequency domain signals corresponding to a plurality of preset frequency bands are extracted from the first frequency domain signal to generate a second frequency domain signal, which may include a plurality of frequency segments, for example. Vibration data meeting preset conditions can be extracted from a plurality of frequency segments of the second frequency domain signal, and vibration data to be detected are formed according to amplitude, frequency and phase data of the vibration signals. And inputting the extracted vibration data to be detected and real-time operation data comprising a plurality of preset types of operation data into the neural network model after training generalization to obtain the detection parameters of the grinding machine output by the model.
The real-time operation data may be preprocessed, for example, amplitude limitation, averaging data of a certain amount or time length, smoothing, etc., and a person skilled in the art may set a preprocessing rule of the data according to an actual situation, or may not perform preprocessing, which is not particularly limited by the present disclosure.
To sum up, this disclosed embodiment is through the real-time vibration signal and the real-time operating data of gathering the mill to use the neural network model of predetermineeing through the training to handle real-time vibration signal and the real-time operating data after the preliminary treatment, can directly reachd mill detection parameters such as the grinding medium filling rate that the mill filled, comprehensive filling rate, promotion strip height, welt degree of wear, have convenient to use, degree of automation is high, generalization performance is strong, precision and reliability are high.
Corresponding to the method embodiment, the present disclosure also provides a mill detection apparatus, which may be used to execute the method embodiment.
FIG. 5 is a block diagram of a mill detection device in an exemplary embodiment of the present disclosure.
Referring to fig. 5, the mill detection device 500 may include:
a data acquisition module 51 configured to acquire a plurality of sets of real-time vibration signals from a plurality of vibration sensors at a preset position of the mill and a set of real-time operation data, the real-time operation data including a plurality of preset types of operation data;
the data extraction module 52 is configured to extract a plurality of sets of vibration data to be detected from the real-time vibration signal according to a preset condition, where the vibration data to be detected includes the frequency, amplitude and phase of a plurality of vibration signals to be detected in the real-time vibration signal, which meet the preset condition.
The parameter acquisition module 53 is set to input the vibration data to be detected and the real-time operation data into a preset neural network model so as to acquire the detection parameters of the mill output by the preset neural network model, wherein the preset type operation data comprise ore property parameters, the loaded ore grinding medium property parameters, the mill rotating speed, the processing capacity, the power, the hard rock amount and the axial pressure, and the detection parameters comprise the ore grinding medium filling rate, the comprehensive filling rate, the lifting strip height and the lining plate abrasion degree loaded by the mill.
In an exemplary embodiment of the present disclosure, the data extraction module 52 is configured to: preprocessing the real-time vibration signal to generate a first frequency domain signal; extracting frequency domain signals corresponding to a plurality of preset frequency bands from the first frequency domain signal to generate a second frequency domain signal; selecting a plurality of to-be-detected vibration signals which meet preset conditions from the second frequency domain signals; and generating vibration data to be detected according to the frequency, amplitude and phase of the vibration signals to be detected.
In an exemplary embodiment of the present disclosure, the data extraction module 52 is configured to: determining optimal sampling data according to the real-time vibration signal; capturing a plurality of sampling data from the optimal sampling data through a preset sampling time window; the plurality of sampled data is fourier transformed to generate a first frequency domain signal.
In an exemplary embodiment of the present disclosure, the data extraction module 52 is configured to: and extracting the real-time vibration signals of which the amplitude is within a first preset range and the frequency is within a second preset range from the real-time vibration signals into optimal sampling data through filtering.
In an exemplary embodiment of the present disclosure, the preset condition is a vibration signal corresponding to an amplitude extremum in the vibration signal from the preset vibration sensor in each preset frequency band.
In an exemplary embodiment of the present disclosure, the preset frequency band includes a frequency band which is determined in advance according to the detection parameter type and is related to the detection parameter type.
In an exemplary embodiment of the present disclosure, the apparatus further includes an observation signal exhibiting module 54 configured to generate a plurality of observation vibration signals according to the plurality of vibration signals to be detected, where the observation vibration signals include single-frequency time-domain vibration signals and/or mixed-frequency time-domain vibration signals; a plurality of observed vibration signals are displayed.
In an exemplary embodiment of the present disclosure, the neural network model is preset as a long-term memory neural network model.
In an exemplary embodiment of the present disclosure, the model training module 55 is further configured to obtain m sets of real-time operation data of the mill under m sets of detection parameters and m × n sets of real-time vibration signals from n vibration sensors, where m and n are integers greater than 1; acquiring m × p vibration signals to be detected according to the m × n groups of real-time vibration signals, wherein p is a natural number; extracting the frequency, amplitude and phase of the m × p vibration signals to be detected to generate m × p groups of vibration data to be detected; extracting a plurality of sample subsets according to m x p groups of vibration data to be detected, m groups of real-time operation data and m groups of detection parameters, wherein the plurality of sample subsets comprise a training data set and a test data set; and training and generalizing the preset neural network according to the plurality of sample subsets.
In an exemplary embodiment of the present disclosure, the mill comprises a semi-autogenous mill, an autogenous mill, or a ball mill, and the preset positions comprise a driving motor bearing seat, a pinion bearing seat, a mill feed inlet bearing seat, a mill discharge outlet bearing seat, and a mill barrel wall.
Since the functions of the apparatus 500 have been described in detail in the corresponding method embodiments, the disclosure is not repeated herein.
FIG. 6 is a block diagram of a mill testing apparatus in an exemplary embodiment of the present disclosure.
Referring to fig. 6, the mill detecting apparatus 600 may include:
the vibration sensors 61 are arranged at a plurality of preset positions of the grinding machine 1 and are used for detecting real-time vibration signals of the grinding machine;
the parameter acquisition device 62 is used for acquiring real-time operation data, and the real-time operation data comprises a plurality of preset types of operation data;
the signal transmission device 63 is coupled to the vibration sensor 61 and the parameter acquisition device 62 and is used for transmitting the real-time vibration signal and the real-time operation data to the signal processing device;
a signal processing device 64, coupled to the signal transmission device 63, for executing the mill detection method according to the embodiment shown in fig. 1 to 4 to output the detection parameters of the mill, wherein the preset type operation data includes ore property parameters, charged grinding medium property parameters, mill rotation speed, processing capacity, power, hard rock amount, and axial pressure, and the detection parameters include a charging rate of the grinding medium charged by the mill, a comprehensive charging rate, a lifting bar height, and a wear degree of the lining plate;
the display device 65 is coupled to the signal processing device 64 for displaying the detection parameters.
In some embodiments, the parameter collecting device 62 may include a human-computer interface (e.g., information obtained through an input device such as a mouse, a keyboard, a touch screen, etc.), various sensors, and various electronic devices, and the specific implementation of the parameter collecting device 62 is determined according to the kind of real-time operation data, and may be set by a person skilled in the art according to actual situations.
In some embodiments, the signal processing device 64 may restore the frequency, amplitude and phase of the plurality of vibration signals to be detected into one or more time domain signals (the one or more time domain signals may be, for example, a single frequency signal, a mixed frequency signal, a single frequency signal and a mixed frequency signal), and then output the one or more time domain signals to the display device 65. The display device 64 can display the one or more time domain signals and the detection parameters of the mill at the same time, so that the operation condition of the mill can be judged visually and timely by a mill maintenance personnel.
It should be noted that although in the above detailed description several modules or units of the device for action execution are mentioned, such a division is not mandatory. Indeed, the features and functionality of two or more modules or units described above may be embodied in one module or unit, according to embodiments of the present disclosure. Conversely, the features and functions of one module or unit described above may be further divided into embodiments by a plurality of modules or units.
In an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided.
As will be appreciated by one skilled in the art, aspects of the present invention may be embodied as a system, method or program product. Thus, various aspects of the invention may be embodied in the form of: an entirely hardware embodiment, an entirely software embodiment (including firmware, microcode, etc.) or an embodiment combining hardware and software aspects that may all generally be referred to herein as a "circuit," module "or" system.
An electronic device 700 according to this embodiment of the invention is described below with reference to fig. 7. The electronic device 700 shown in fig. 7 is only an example and should not bring any limitation to the functions and the scope of use of the embodiments of the present invention.
As shown in fig. 7, electronic device 700 is embodied in the form of a general purpose computing device. The components of the electronic device 700 may include, but are not limited to: the at least one processing unit 710, the at least one memory unit 720, and a bus 730 that couples various system components including the memory unit 720 and the processing unit 710.
Where the memory unit stores program code, the program code may be executed by the processing unit 710 such that the processing unit 710 performs the steps according to various exemplary embodiments of the present invention as described in the above-mentioned "exemplary methods" section of this specification. For example, processing unit 710 may perform a method as shown in fig. 1.
The storage unit 720 may include readable media in the form of volatile memory units, such as a random access memory unit (RAM)7201 and/or a cache memory unit 7202, and may further include a read only memory unit (ROM) 7203.
The storage unit 720 may also include a program/utility 7204 having a set (at least one) of program modules 7205, such program modules 7205 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may comprise an implementation of a network environment.
The electronic device 700 may also communicate with one or more external devices 800 (e.g., keyboard, pointing device, bluetooth device, etc.), with one or more devices that enable a user to interact with the electronic device 700, and/or with any devices (e.g., router, modem, etc.) that enable the electronic device 700 to communicate with one or more other computing devices. Such communication may occur via an input/output (I/O) interface 750. Also, the electronic device 700 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network such as the internet) via the network adapter 760. As shown, the network adapter 760 communicates with the other modules of the electronic device 700 via the bus 730. It should be appreciated that although not shown in the figures, other hardware and/or software modules may be used in conjunction with the electronic device 700, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.
Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein may be implemented by software, or by software in combination with necessary hardware. Therefore, the technical solution according to the embodiments of the present disclosure may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (which may be a CD-ROM, a usb disk, a removable hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which may be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.
In an exemplary embodiment of the present disclosure, there is also provided a computer-readable storage medium having stored thereon a program product capable of implementing the above-described method of the present specification. In some possible embodiments, aspects of the invention may also be implemented in the form of a program product comprising program code means for causing a terminal device to carry out the steps according to various exemplary embodiments of the invention described in the above-mentioned "exemplary methods" section of the present description, when the program product is run on the terminal device.
The program product for implementing the above method according to an embodiment of the present invention may employ a portable compact disc read only memory (CD-ROM) and include program codes, and may be run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited in this regard and, in the present document, a readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
A computer readable signal medium may include a propagated data signal with readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A readable signal medium may also be any readable medium that is not a readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C + +, Python, etc., as well as conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., through the internet using an internet service provider). The remote computing device may include, for example, an edge computing device and/or a cloud computing device.
Furthermore, the above-described figures are merely schematic illustrations of processes involved in methods according to exemplary embodiments of the invention, and are not intended to be limiting. It will be readily understood that the processes shown in the above figures are not intended to indicate or limit the chronological order of the processes. In addition, it is also readily understood that these processes may be performed synchronously or asynchronously, e.g., in multiple modules.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
Claims (13)
1. A mill testing method, comprising:
acquiring a plurality of groups of real-time vibration signals and a group of real-time operation data from a plurality of vibration sensors at preset positions of a mill, wherein the real-time operation data comprises a plurality of preset kinds of operation data;
extracting multiple groups of vibration data to be detected from the real-time vibration signals according to preset conditions, wherein the vibration data to be detected comprises the frequency, amplitude and phase of a plurality of vibration signals to be detected which meet the preset conditions in the real-time vibration signals;
inputting the vibration data to be detected and the real-time operation data into a preset neural network model so as to obtain detection parameters of the mill output by the preset neural network model, wherein the preset type operation data comprises ore property parameters, loaded grinding medium property parameters, mill rotating speed, processing capacity, power, hard rock quantity and axial pressure, and the detection parameters comprise grinding medium filling rate, comprehensive filling rate, lifting strip height and lining plate abrasion degree loaded by the mill;
wherein, it is right to wait to examine vibration data to the real-time vibration signal extraction multiunit according to the condition of predetermineeing includes:
preprocessing the real-time vibration signal to generate a first frequency domain signal;
extracting frequency domain signals corresponding to a plurality of preset frequency bands from the first frequency domain signal to generate a second frequency domain signal;
selecting a plurality of to-be-detected vibration signals which meet the preset conditions from the second frequency domain signals;
and generating the vibration data to be detected according to the frequency, the amplitude and the phase of the vibration signals to be detected.
2. The mill detection method of claim 1 wherein the pre-processing the real-time vibration signal to generate a first frequency domain signal comprises:
determining optimal sampling data according to the real-time vibration signal;
capturing a plurality of sampling data from the optimal sampling data through a preset sampling time window;
fourier transforming the plurality of sampled data to generate the first frequency domain signal.
3. The mill testing method of claim 2 wherein said determining optimal sampling data from said real-time vibration signal comprises:
and extracting the real-time vibration signals of which the amplitude is within a first preset range and the frequency is within a second preset range from the real-time vibration signals into the optimal sampling data through filtering.
4. A mill testing method according to claim 1, wherein said predetermined condition is a vibration signal corresponding to an extreme amplitude value in vibration signals from predetermined vibration sensors in each of said predetermined frequency bands.
5. The mill testing method as claimed in claim 1, wherein said preset frequency band comprises a frequency band associated with said testing parameter type determined in advance according to said testing parameter type.
6. The mill testing method of claim 1, further comprising:
generating observation vibration signals according to the plurality of vibration signals to be detected, wherein the observation vibration signals comprise single-frequency time domain vibration signals and/or frequency mixing time domain vibration signals;
displaying the observed vibration signal.
7. The mill testing method of claim 1, wherein the pre-defined neural network model is an episodic memory neural network model.
8. The mill detection method according to claim 1 or 7, wherein the training process of the preset neural network model comprises:
acquiring m groups of real-time operation data of the mill under m groups of detection parameters and m x n groups of real-time vibration signals from n vibration sensors, wherein m and n are integers greater than 1;
acquiring m × p vibration signals to be detected according to the m × n groups of real-time vibration signals, wherein p is a natural number;
extracting the frequency, amplitude and phase of the m × p vibration signals to be detected to generate m × p groups of vibration data to be detected;
extracting a plurality of sample subsets according to the m groups of vibration data to be detected, the m groups of real-time operation data and the m groups of detection parameters, wherein the plurality of sample subsets comprise a training data set and a testing data set;
and training and generalizing the preset neural network according to the plurality of sample subsets.
9. The mill testing method as claimed in claim 1, wherein said mill comprises a semi-autogenous mill, an autogenous mill and a ball mill, and said predetermined positions comprise a driving motor bearing seat, a pinion bearing seat, a mill inlet bearing seat, a mill outlet bearing seat, and a mill barrel wall.
10. A mill testing apparatus, comprising:
the data acquisition module is used for acquiring a plurality of groups of real-time vibration signals and a group of real-time operation data from a plurality of vibration sensors at preset positions of the mill, and the real-time operation data comprises a plurality of preset types of operation data;
the data extraction module is used for extracting a plurality of groups of to-be-detected vibration data from the real-time vibration signals according to preset conditions, wherein the to-be-detected vibration data comprise the frequency, amplitude and phase of a plurality of to-be-detected vibration signals which meet the preset conditions in the real-time vibration signals;
the parameter acquisition module is used for inputting the vibration data to be detected and the real-time operation data into a preset neural network model so as to acquire detection parameters of the mill output by the preset neural network model, wherein the preset type operation data comprises ore property parameters, loaded grinding medium property parameters, mill rotating speed, processing capacity, power, stubborn stone quantity and axial pressure, and the detection parameters comprise the grinding medium filling rate, the comprehensive filling rate, the lifting strip height and the lining plate abrasion degree loaded by the mill;
wherein the data extraction module is configured to: preprocessing the real-time vibration signal to generate a first frequency domain signal; extracting frequency domain signals corresponding to a plurality of preset frequency bands from the first frequency domain signal to generate a second frequency domain signal; selecting a plurality of to-be-detected vibration signals which meet the preset conditions from the second frequency domain signals; and generating the vibration data to be detected according to the frequency, the amplitude and the phase of the vibration signals to be detected.
11. A mill testing apparatus, comprising:
the vibration sensors are arranged at a plurality of preset positions of the grinding machine and used for detecting real-time vibration signals of the grinding machine;
the system comprises a parameter acquisition device, a parameter storage device and a parameter processing device, wherein the parameter acquisition device is used for acquiring real-time operation data which comprises a plurality of preset types of operation data;
the signal transmission device is coupled with the vibration sensor and the parameter acquisition device and is used for transmitting the real-time vibration signal and the real-time operation data to the signal processing device;
signal processing means, coupled to the signal transmission means, for executing the mill detection method according to any one of claims 1 to 9, so as to output detection parameters of a mill, wherein the preset category of operation data includes ore property parameters, charged grinding medium property parameters, mill rotation speed, processing capacity, power, hard rock amount, and axial pressure, and the detection parameters include charging rate of grinding medium, comprehensive charging rate, lifting bar height, and wear degree of lining plate of the mill;
and the display device is coupled with the signal processing device and used for displaying the detection parameters.
12. An electronic device, comprising:
a memory; and
a processor coupled to the memory, the processor configured to execute the mill detection method of any of claims 1-9 based on instructions stored in the memory.
13. A computer-readable storage medium, on which a program is stored, which when executed by a processor implements the mill detection method according to any one of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010367485.3A CN111521259B (en) | 2020-04-30 | 2020-04-30 | Grinding machine detection method, device and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010367485.3A CN111521259B (en) | 2020-04-30 | 2020-04-30 | Grinding machine detection method, device and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111521259A CN111521259A (en) | 2020-08-11 |
| CN111521259B true CN111521259B (en) | 2022-02-18 |
Family
ID=71906846
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010367485.3A Active CN111521259B (en) | 2020-04-30 | 2020-04-30 | Grinding machine detection method, device and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111521259B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112665710A (en) * | 2020-12-21 | 2021-04-16 | 陕西宝光集团有限公司 | Method and device for detecting running state of equipment, electronic equipment and storage medium |
| CN113916366B (en) * | 2021-10-21 | 2024-04-19 | 山东鑫海矿业技术装备股份有限公司 | Method and equipment for monitoring operation of impeller of vortex breaker based on vibration signal |
| CN114308358B (en) * | 2022-03-17 | 2022-05-27 | 山东金有粮脱皮制粉设备有限公司 | Safe operation monitoring system of corncob grinding device |
| CN115090378A (en) * | 2022-08-26 | 2022-09-23 | 启东市春晨机械有限公司 | Abnormal vibration monitoring and early warning method for hammer head rotor of metal crusher |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0712638A (en) * | 1993-06-28 | 1995-01-17 | Kawasaki Steel Corp | Method and apparatus for detecting abnormal vibration in winding device of cold |
| CN101358862A (en) * | 2008-03-17 | 2009-02-04 | 西安艾贝尔科技发展有限公司 | Measurement method and device for material status in barrel type grinding mill |
| CN102564494A (en) * | 2011-12-20 | 2012-07-11 | 西安艾贝尔科技发展有限公司 | Method and device for measuring state of substances in drum type grinding machine |
| KR101620507B1 (en) * | 2015-10-07 | 2016-05-13 | 한국지질자원연구원 | A vibrate monitoring apparatus of ball mill system and a method thereof |
| CN105806643A (en) * | 2016-05-24 | 2016-07-27 | 中国矿业大学 | Identification method and device for gas-liquid mixing state of dust remover |
| CN106568503A (en) * | 2016-11-07 | 2017-04-19 | 西安交通大学 | Mill load detection method based on cylinder surface multiple vibration signals |
| CN107132451A (en) * | 2017-05-31 | 2017-09-05 | 广州供电局有限公司 | The winding state detection method and system of transformer |
| CN107462319A (en) * | 2017-09-15 | 2017-12-12 | 安徽理工大学 | The acoustics identifying processing method and experimental provision of a kind of micro-machine noise |
| CN110907207A (en) * | 2019-11-25 | 2020-03-24 | 湃方科技(天津)有限责任公司 | Running state detection method and device for non-marking mechanical equipment |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101623667B (en) * | 2009-07-31 | 2012-05-09 | 浙江中控技术股份有限公司 | Automatic control method and system for mill load |
| US9772219B2 (en) * | 2012-09-11 | 2017-09-26 | S.P.M. Instrument Ab | Apparatus for monitoring the condition of a machine |
| US9955274B2 (en) * | 2015-04-08 | 2018-04-24 | The Boeing Company | Vibration monitoring systems |
| CN105115592B (en) * | 2015-09-02 | 2018-02-16 | 东莞市中光通信科技有限公司 | Cranial cavity method for detecting vibration and device |
| WO2017079452A1 (en) * | 2015-11-03 | 2017-05-11 | Massachusetts Institute Of Technology | System and method for discriminating between origins of vibrations in an object and determination of contact between blunt bodies traveling in a medium |
| CN109871831B (en) * | 2019-03-18 | 2021-01-05 | 太原理工大学 | Emotion recognition method and system |
| FR3102554B1 (en) * | 2019-10-23 | 2021-11-19 | Alstom Transp Tech | Method and system for estimating the wear of a rotating machine comprising a bearing |
-
2020
- 2020-04-30 CN CN202010367485.3A patent/CN111521259B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0712638A (en) * | 1993-06-28 | 1995-01-17 | Kawasaki Steel Corp | Method and apparatus for detecting abnormal vibration in winding device of cold |
| CN101358862A (en) * | 2008-03-17 | 2009-02-04 | 西安艾贝尔科技发展有限公司 | Measurement method and device for material status in barrel type grinding mill |
| CN102564494A (en) * | 2011-12-20 | 2012-07-11 | 西安艾贝尔科技发展有限公司 | Method and device for measuring state of substances in drum type grinding machine |
| KR101620507B1 (en) * | 2015-10-07 | 2016-05-13 | 한국지질자원연구원 | A vibrate monitoring apparatus of ball mill system and a method thereof |
| CN105806643A (en) * | 2016-05-24 | 2016-07-27 | 中国矿业大学 | Identification method and device for gas-liquid mixing state of dust remover |
| CN106568503A (en) * | 2016-11-07 | 2017-04-19 | 西安交通大学 | Mill load detection method based on cylinder surface multiple vibration signals |
| CN107132451A (en) * | 2017-05-31 | 2017-09-05 | 广州供电局有限公司 | The winding state detection method and system of transformer |
| CN107462319A (en) * | 2017-09-15 | 2017-12-12 | 安徽理工大学 | The acoustics identifying processing method and experimental provision of a kind of micro-machine noise |
| CN110907207A (en) * | 2019-11-25 | 2020-03-24 | 湃方科技(天津)有限责任公司 | Running state detection method and device for non-marking mechanical equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111521259A (en) | 2020-08-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111521259B (en) | Grinding machine detection method, device and equipment | |
| Fugate et al. | Vibration-based damage detection using statistical process control | |
| CN108168811A (en) | The Portable acquiring analytical equipment and method of a kind of vibration signal | |
| EP2836881B1 (en) | Embedded prognostics on plc platforms for equipment condition monitoring, diagnosis and time-to-failure/service prediction | |
| US6694286B2 (en) | Method and system for monitoring the condition of an individual machine | |
| CN101726383B (en) | Multi-rope winder steel wire rope tension test method | |
| US8907243B2 (en) | Maintenance system for wire transport system of wire discharge processing machine | |
| CN105067239B (en) | The beam crack fault detection means and method vibrated based on swept frequency excitation | |
| CN109141886A (en) | A kind of vibration and the state of wear combined monitoring experiment porch of shaft and bearing | |
| CN103775832A (en) | Transient flow problem method-based oil pipeline dropping detection device | |
| CN110455517B (en) | Tower barrel health monitoring method of wind generating set | |
| KR20110009615A (en) | Data collection device, and diagnosis device of facility management with data collection device thereof | |
| CN115943516A (en) | Vehicle battery unbalance detection method and device, electronic equipment and storage medium | |
| JP6511573B1 (en) | Method and apparatus for diagnosing abnormality of rolling bearing, abnormality diagnosis program | |
| CN112729793B (en) | Weak fault feature extraction method based on nonlinear spectrum analysis | |
| CN114201831A (en) | Rolling bearing working condition quantitative analysis method based on vibration signal real-time acquisition | |
| CN114216640A (en) | Method, apparatus and medium for detecting fault status of industrial equipment | |
| CN115982896B (en) | Bearing retainer service life detection method and device | |
| CN117171657A (en) | Wind power generation equipment fault diagnosis method and device, electronic equipment and storage medium | |
| JP7082585B2 (en) | Bearing information analysis device and bearing information analysis method | |
| CN115655717A (en) | Bearing fault diagnosis method based on depth domain generalization network | |
| CN105675321B (en) | A kind of equipment performance degeneration radar map determines method | |
| CN107977679A (en) | Method based on frequency response function and operation response characteristic diagnosis of complex device initial failure | |
| CN110726542A (en) | Analysis method for fatigue life of spring | |
| Koshekov et al. | An intelligent system for vibrodiagnostics of oil and gas equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |