Mill load monitoring system and method based on cylinder vibration signal

Abstract

The invention relates to the technical field of mineral separation automation, and discloses a mill load monitoring system and method based on a barrel vibration signal, a signal acquisition part of the system is installed on the outer side of the surface of a mill barrel, and equipment power supply and data transmission are achieved in a wireless charging and wireless communication mode; upper computer software is operated on the industrial personal computer to receive vibration signals, relevant information is stored and displayed in real time, and meanwhile, the signals are analyzed based on a monitoring method, and the load of the mill is calculated. The system can continuously and reliably operate in an actual production environment, effectively overcomes the influence of field noise and environmental interference, and monitors the load condition of the mill in real time.

Description

Technical Field

[0001] The present invention relates to the technical field of ore dressing automation, and particularly to a mill load monitoring system and method based on cylinder vibration signals. Background Art

[0002] A mill is the main equipment for realizing the grinding process. When the mill is operating, the motor drives the mill cylinder to rotate continuously. By the continuous impact and extrusion between the steel balls, ores filled inside the cylinder and the mill liner, the mineral size is reduced, and the grinding of the ores is realized. The mill load refers to the total loading amount of ores, water and steel balls inside the mill, and is an important parameter during the operation of the mill. An excessive mill load (overload) will lead to insufficient grinding of the ores, easily cause internal blockage, and even trigger a production shutdown accident; a too small mill load (underload) will result in excessive useless power consumption and steel consumption of the ball mill, and accelerate equipment damage. Therefore, keeping the mill operating in a good load state is crucial for improving the energy utilization efficiency and production benefits of the ore dressing plant.

[0003] However, since the materials inside the mill rotate continuously in a closed space, it is impossible to directly detect the internal state of the mill and calculate the mill load. The existing mill load monitoring means usually adopt an indirect measurement and calculation method, including monitoring the ore feeding amount, power, sound (grinding sound), etc. when the mill is operating stably. Although these variables can provide certain information, the mapping relationship between them and the mill load is not direct and is relatively complex, and it is difficult to ensure the long-term stability and accuracy of the monitoring system. In addition, due to the influence of various factors such as ore properties, steel balls, and liners during the grinding process, and at the same time there are interactions with subsequent processing links, it is very difficult to establish a long-term effective mechanism model to accurately determine the relationship between each variable and the mill load. To sum up, in actual production, it is of great significance to use equipment that can collect signals reflecting the internal state of the mill, establish a data-driven model based on the collected signals, judge the mill load situation, and then optimize the production operation and process.

[0004] Existing mill load monitoring systems usually collect the sound emitted by the mill during operation, i.e., the grinding sound, and establish a data-driven or mechanism-based load identification model based on the characteristics of the grinding sound to achieve the judgment of the internal condition of the mill. Although the load monitoring system based on the grinding sound has been practically applied in many ore dressing plants and brought certain economic benefits, this method has the following limitations: 1. Environmental interference problem: The method of collecting sound based on a microphone is extremely vulnerable to the influence of industrial site noise. The actual industrial site environment is noisy, including the sounds emitted by other large equipment during operation, the sounds generated by the activities of workers, and the noises brought by production operations such as adding steel balls, etc., which may all be captured by the microphone. The existence of these background noises makes the grinding sound signal always contain a large amount of interference. Especially when multiple mills are operating simultaneously, their respective grinding sounds interfere with each other. These grinding sound signals have similar frequencies and are difficult to effectively distinguish during the signal processing process. The existence of noise greatly affects the accuracy and reliability of the data-driven model. 2. Signal interpretation problem: The grinding sound signal can only be used as an indirect evaluation basis for the mill load. Although there is an empirical rule that "when the mill load is high, the grinding sound is small and dull, and when the load is low, the grinding sound is large and loud" for the change in the sound amplitude and frequency when the internal filling amount of the mill changes, this rule cannot directly and clearly characterize the load size inside the mill, and this type of interpretation based on sound characteristics does not always apply to all production environments. Therefore, in different ore dressing plants, the load identification system based on the grinding sound may not achieve consistent high efficiency and accuracy. It can be seen that the existing mill load monitoring system based on the grinding sound has problems of being unable to effectively monitor the load change inside the mill and having a low applicability in the actual grinding process. Summary of the Invention

[0005] The present invention provides a mill load monitoring system and method based on the vibration signal of the cylinder body to solve the problems in the prior art that the load change inside the mill cannot be effectively monitored and the applicability is not high.

[0006] To achieve the above object, the present invention is realized through the following technical solutions:

[0007] In the first aspect, the present invention provides a mill load monitoring system based on the vibration signal of the cylinder body, including: a signal transmitting end, a signal receiving end, a signal processing end, a wireless charging power module, and an active space positioning module. The signal transmitting end is arranged inside a housing, and the housing is arranged on the surface of the mill cylinder body and rotates synchronously with the cylinder body during the operation of the mill;

[0008] The signal transmitting end is used to collect the vibration signal on the surface of the mill cylinder body;

[0009] The signal receiving end is used to receive the vibration signal and send the vibration signal to the signal processing end;

[0010] The wireless charging power module is used to supply power to the signal transmitting end, the signal receiving end, and the active spatial positioning module;

[0011] The active spatial positioning module is used to calculate the phase of the mill cylinder in space in real time and send the phase to the signal processing end;

[0012] The signal processing end is used to calculate the mill load based on the vibration signal and the phase.

[0013] Optionally, the signal transmitting end includes a vibration sensor, a signal conditioning circuit, an ADC converter, an MCU, a wireless communication module, and a first antenna;

[0014] The vibration sensor is arranged on the surface of the mill cylinder and rotates synchronously with the cylinder during the operation of the mill, and is used to convert the vibration on the surface of the mill cylinder into an analog electrical signal; the signal conditioning circuit is used to preprocess the analog electrical signal and send it to a channel of the ADC converter; the ADC converter is used to sample the analog electrical signal input to this channel at a set sampling frequency to obtain a digital vibration signal; the MCU is used to drive the ADC converter to perform analog-to-digital conversion and communicate with the ADC converter bidirectionally through a high-speed bus to read the sampled digital vibration signal; the wireless communication module is used to communicate with the MCU bidirectionally through a high-speed bus and transmit the read information to the signal receiving end through the first antenna at a set frequency band.

[0015] Optionally, the signal transmitting end further includes a DCDC buck module, and the DCDC buck module is respectively connected to the signal conditioning circuit, the ADC converter, the MCU, and the wireless communication module for power supply.

[0016] Optionally, the active spatial positioning module includes a photoelectric sensor and a reflector. The photoelectric sensor is arranged inside the housing, and the reflector is installed at a position within a set range from the mill cylinder; the photoelectric sensor is used to emit infrared light outward. When the infrared light is not reflected by any object, the photoelectric sensor outputs a low level; when the infrared light is reflected by the reflector and received by the photoelectric sensor, the photoelectric sensor outputs a high level; wherein, the time between two high levels is regarded as a complete circle; the phase is calculated by the ratio of the time difference between the current moment and the previous trigger of the photoelectric sensor to the time interval between two adjacent triggers;

[0017] The calculation process of the phase satisfies the following relational expression:

[0018] Phase=(Time Current -Trigger 1 ) / (Trigger 2-Trigger 1 )·360;

[0019] wherein, Trigger2 - Trigger1 represents the time interval between two adjacent high levels output by the photoelectric sensor, and Time Current represents the current moment.

[0020] Optionally, the wireless charging power module includes a switching power supply, a wireless charging transmitting host, a wireless charging transmitting coil, a wireless charging receiving host, and a wireless charging receiving coil; the switching power supply, the wireless charging transmitting host, and the wireless charging transmitting coil are arranged at a position within a set range from the mill cylinder body, and the switching power supply is connected to the wireless charging transmitting host for power supply; the wireless charging transmitting host is used to convert the input direct current into high-frequency alternating current and transmit the high-frequency alternating current to the wireless charging transmitting coil; the wireless charging receiving coil identifies the wireless charging transmitting coil through magnetic field coupling and receives electric energy, and converts the high-frequency alternating current back into direct current through the wireless charging receiving host; and the output end of the wireless charging receiving host is connected in parallel to the input end of the DCDC buck module.

[0021] Optionally, the signal receiving end includes a second antenna and a wireless base station, the second antenna is communicatively connected to the first antenna, and the second antenna receives the vibration signal from the first antenna and transmits it to the wireless base station.

[0022] Optionally, the signal processing end includes an industrial computer, the industrial computer is connected to the wireless base station, the industrial computer is used to read the vibration signal transmitted by the wireless base station, display the waveform of the vibration signal on the upper computer interface in real time, and display relevant parameters. The industrial computer is also used to identify the mill load by using a load identification model. The input of the load identification model is the vibration signal and the phase position, and the output of the identification model is the mill load.

[0023] In a second aspect, the present application further provides a method for monitoring the mill load based on the vibration signal of the cylinder body, which is applied to the mill load monitoring system based on the vibration signal of the cylinder body described in the first aspect. The method includes:

[0024] Collect the vibration signal on the surface of the mill cylinder body by using the signal transmitting end;

[0025] The signal receiving end is used to receive the vibration signal and send the vibration signal to the signal processing end;

[0026] Use the wireless charging power module to supply power to the signal transmitting end, the signal receiving end, and the active space positioning module;

[0027] An active spatial positioning module is used to calculate the phase of the mill cylinder in space in real time, and the phase is sent to the signal processing end;

[0028] The signal processing end is used to calculate the mill load based on the vibration signal and the phase;

[0029] Optionally, the signal processing end calculating the mill load based on the vibration signal and the phase includes:

[0030] Step 1: Segment the vibration signal according to the phase, and truncate the data in a single signal file in a full-circle form in time according to the preset phase start point and end point;

[0031] Step 2: Preprocess the vibration signal segmented by circle to obtain a signal that meets the basic requirements for the input of the load recognition model;

[0032] The preprocessing includes: first subtracting the amplitude mean value of the vibration signal in the time domain to remove the trend term, then performing maximum absolute value normalization on the waveform amplitude so as to represent the vibration signal in relative intensity, and finally performing pre-emphasis processing on the high-frequency band of the vibration signal;

[0033] Step 3: Convert the preprocessed vibration signal from the time-domain waveform to a time-frequency spectrogram and input it into the load recognition model of the mill to obtain the mill load output by the load recognition model;

[0034] The input model time-frequency spectrogram is a time-frequency spectrogram group formed by superimposing three-layer time-frequency spectrograms. The frequency range of the first-layer time-frequency spectrogram only covers the key frequency band divided based on prior knowledge, the frequency range of the second layer is the remaining frequency band except the key frequency band, and the third layer is the time-frequency representation converted to the logarithmic scale; by adjusting the frame length and frame shift of the time-frequency spectrogram, ensure that each layer of time-frequency spectrogram has the same dimension in time and frequency; the three-layer time-frequency spectrograms have the same size and are independent of each other, regarded as three different channels.

[0035] Beneficial effects:

[0036] For the mill load monitoring system based on the cylinder vibration signal provided by the present invention, the signal acquisition part is installed on the outer surface of the mill cylinder, and wireless charging and wireless communication are used to realize device power supply and data transmission; the upper computer software runs on the industrial control computer to receive the vibration signal, save and display relevant information in real time, and at the same time identify the mill load based on the monitoring method to monitor the mill load situation in real time. This system can operate continuously and reliably in the actual production environment, effectively overcoming the influence of on-site noise and environmental interference.

[0037] In a further technical solution, the recognition model based on the deep learning neural network can accurately extract the inherent features of the vibration signals of the mill cylinder under different load conditions, and realize the real-time monitoring of the mill load through the analysis of the inherent features of the vibration signals. Description of the Drawings

[0038] Figure 1 It is a block diagram of a mill load monitoring system based on the vibration signals of the cylinder according to a preferred embodiment of the present invention;

[0039] Figure 2 It is a waveform display diagram of the upper computer according to a preferred embodiment of the present invention;

[0040] Figure 3 It is a block diagram of the recognition model according to a preferred embodiment of the present invention;

[0041] Figure 4 It is a block diagram of a noise reduction feature fusion residual sub-module based on spatial attention according to a preferred embodiment of the present invention;

[0042] Figure 5 It is a structural diagram of a noise reduction feature fusion unit (SAD Unit) based on spatial attention according to a preferred embodiment of the present invention;

[0043] Figure 6 It is a confusion matrix for classifying the mill load status according to a preferred embodiment of the present invention;

[0044] Figure 7 It is a field physical diagram in an example according to a preferred embodiment of the present invention;

[0045] Figure 8 It is a flowchart of a mill load monitoring method based on the vibration signals of the cylinder according to a preferred embodiment of the present invention. Detailed Embodiments

[0046] The technical solutions of the present invention will be described clearly and completely below. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts fall within the protection scope of the present invention.

[0047] Unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the ordinary meanings understood by those of ordinary skill in the art to which the present invention pertains. The "first", "second" and similar terms used in the present invention do not denote any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "a" or "an" do not denote a quantity limitation, but mean that there is at least one. The terms "connected" or "coupled" and the like are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The terms "upper", "lower", "left", "right" and the like are only used to indicate relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship also changes accordingly.

[0048] Please refer to Figure 1 , a mill load monitoring system based on the vibration signal of the cylinder body provided by the present application, includes: a signal transmitting end, a signal receiving end, a signal processing end, a wireless charging power supply module and an active spatial positioning module. The signal transmitting end is arranged in a housing, and the housing is arranged on the surface of the mill cylinder body and rotates synchronously with the cylinder body during the operation of the mill;

[0049] The signal transmitting end is used to collect the vibration signal on the surface of the mill cylinder body;

[0050] The signal receiving end is used to receive the vibration signal and send the vibration signal to the signal processing end;

[0051] The wireless charging power supply module is used to supply power to the signal transmitting end, the signal receiving end and the active spatial positioning module;

[0052] The active spatial positioning module is used to calculate the phase of the mill cylinder body in space in real time and send the phase to the signal processing end;

[0053] The signal processing end is used to calculate the mill load based on the vibration signal and the phase.

[0054] It should be noted that the vibration signal of the mill cylinder body has obvious advantages compared with the grinding sound signal. First of all, the cylinder body vibration signal will not be interfered by the operation of other equipment on site. The cylinder body vibration signal of each mill is only related to the operation state of that mill, thus ensuring the correlation and accuracy of the signal. Secondly, the vibration of the mill cylinder body can directly reflect the change of its internal load. Any change in load will cause corresponding changes in the vibration characteristics. This consistent correlation relationship shows stability among different mills and concentrator plants, improving the universality and reliability of the signal. Therefore, in the present application, by arranging the signal transmitting end in the housing, and the housing is arranged on the surface of the mill cylinder body and rotates synchronously with the cylinder body during the operation of the mill; it can directly reflect the change of its internal load, improving the universality and reliability of the signal.

[0055] Further, it is worth noting that the mill may run continuously for several months after startup. Its surface vibrates violently and it rotates continuously for a long time. Coupled with the complex and harsh industrial site environment, there is currently a lack of equipment that can stably collect the vibration signals of the mill cylinder in the industrial site for a long time. At the same time, the vibration signals required for analysis have certain requirements for the sampling frequency, and the commonly used LoRa transmission in the industrial site cannot meet the signal transmission rate at the required sampling frequency. Therefore, in this application, by setting up a wireless charging power module for power supply and an antenna for signal transmission, signal transmission can be quickly realized under self-powered conditions, and continuous monitoring can be achieved.

[0056] In summary, the above-mentioned mill load monitoring system based on the cylinder vibration signal has its acquisition part installed on the outer surface of the mill cylinder, and uses wireless charging and wireless communication methods to realize the power supply of the equipment and the transmission of data; the host computer software runs on the industrial control computer and loads the monitoring method to receive the vibration signal, save and display relevant information in real time, and identify the mill load based on the monitoring method, so as to monitor the mill load situation in real time. This system can operate reliably in a continuous production environment and effectively overcome the influence of on-site noise and environmental interference.

[0057] Optionally, the signal transmitter includes a vibration sensor, a signal conditioning circuit, an ADC converter, an MCU, a wireless communication module, and a first antenna; the vibration sensor is arranged on the surface of the mill cylinder and rotates synchronously with the cylinder during the operation of the mill, and is used to convert the vibration on the surface of the mill cylinder into an analog electrical signal; the signal conditioning circuit is used to preprocess the analog electrical signal and send it to a channel of the ADC converter; the ADC converter is used to sample the analog electrical signal input to this channel at a set sampling frequency to obtain a digital vibration signal; the MCU is used to drive the ADC converter to perform analog-to-digital conversion, and communicate with the ADC converter bidirectionally through a high-speed bus to read the sampled digital vibration signal; the wireless communication module is used to communicate with the MCU bidirectionally through a high-speed bus, and transmit the read information to the signal receiver through the first antenna at a set frequency band.

[0058] Furthermore, the signal transmitter further includes a DCDC buck module, and the DCDC buck module is respectively connected to the signal conditioning circuit, the ADC converter, the MCU, and the wireless communication module for power supply. The DCDC buck module further steps down the voltage through its own LDO to meet the requirements of different chip pins for the input voltage.

[0059] Optionally, the active spatial positioning module includes a photoelectric sensor and a mirror. The photoelectric sensor is disposed inside the housing, and the mirror is installed at a position within a set range from the mill cylinder. The photoelectric sensor is used to emit infrared light outward. When the infrared light is not reflected by any object, the photoelectric sensor outputs a low level. When the infrared light is reflected by the mirror and received by the photoelectric sensor, the photoelectric sensor outputs a high level. Among them, the time between two high levels is regarded as a complete circle. The phase is calculated by the ratio of the time difference between the current moment and the previous trigger of the photoelectric sensor to the time interval between two adjacent triggers.

[0060] Among them, the calculation process of the phase satisfies the following relational expression:

[0061] Phase=(Time Current -Trigger 1 ) / (Trigger 2 -Trigger 1 )·360;

[0062] In the formula, Trigger2 - Trigger1 represents the time interval between two adjacent high levels output by the photoelectric sensor, and Time Current represents the current moment.

[0063] Optionally, the wireless charging power module includes a switching power supply, a wireless charging transmitting host, a wireless charging transmitting coil, a wireless charging receiving host, and a wireless charging receiving coil. The switching power supply, the wireless charging transmitting host, and the wireless charging transmitting coil are disposed at a position within a set range from the mill cylinder. The switching power supply is connected to the wireless charging transmitting host for power supply. The wireless charging transmitting host is used to convert the input direct current into high-frequency alternating current and transmit the high-frequency alternating current to the wireless charging transmitting coil. The wireless charging receiving coil identifies the wireless charging transmitting coil through magnetic field coupling and receives electric energy, and converts the high-frequency alternating current back into direct current through the wireless charging receiving host. And the output end of the wireless charging receiving host is connected in parallel to the input end of the DCDC buck module.

[0064] In a feasible embodiment, the housing is installed at the middle position of the ball mill cylinder. The wireless charging receiving coil is wound into a circle and installed under the top cover of the housing. The wireless charging transmitting coil is installed at a position at a set distance from the top of the housing.

[0065] Optionally, the signal receiving end includes a second antenna and a wireless base station. The second antenna is communicatively connected to the first antenna, and the second antenna receives the vibration signal from the first antenna and transmits it to the wireless base station.

[0066] Optionally, the signal processing end includes an industrial control computer, which is connected to the wireless base station. The industrial control computer is used to read the vibration signals transmitted by the wireless base station, display the waveforms of the vibration signals on the upper computer interface in real time, and display relevant parameters. The industrial control computer is also used to identify the mill load using a load identification model. The input of the load identification model is the vibration signal and the phase position, and the output of the identification model is the mill load.

[0067] Specifically, in an installation example, all components of the signal transmitting end, the wireless charging receiving host of the wireless charging power module, the wireless charging receiving coil, and the photoelectric sensor of the active space positioning module are installed in the acquisition system housing. The housing is installed on the surface of the mill cylinder and fixed by bolts on the cylinder surface. It rotates synchronously with the cylinder during the operation of the mill. The wireless base station of the signal receiving end, the switching power supply of the wireless charging power module, the wireless charging transmitting host, the wireless charging transmitting coil, and the reflector of the active space positioning module are installed at a position near the mill. Among them, the wireless charging transmitting coil and the reflector are installed at an appropriate position away from the mill cylinder through an installation bracket, and the rest of the equipment is installed at a suitable position on site. Specifically, the wireless charging transmitting coil is installed and fixed at a position 5 - 15 cm away from the mill cylinder wall on the side of the mill cylinder, and is aligned horizontally with the housing installed on the mill. During the operation of the mill, the housing rotates synchronously with the mill cylinder. Whenever it rotates into the effective range of the wireless charging transmitting coil, a magnetic field coupling is generated between the transmitting coil and the receiving coil, and the receiving coil receives the electric energy transmitted by the transmitting coil.

[0068] Specifically, the working process of the mill load monitoring system for the vibration signal of the cylinder provided by this application is as follows:

[0069] The vibration sensor converts the vibration on the surface of the mill cylinder into an electrical signal (analog quantity). The generated analog signal is transmitted to the signal conditioning circuit through a cable. This circuit processes the electrical signal, such as amplification and filtering. The processed analog vibration signal is transmitted to a channel of the ADC converter through a cable. The ADC converter samples the analog signal input to this channel at a certain frequency to obtain a vibration signal in digital quantity form. The sampling frequency can be adjusted according to different actual situations at different sites, with a minimum not lower than the lower limit frequency required by the subsequent vibration signal analysis algorithm and a maximum not higher than the upper limit frequency that can be achieved for stable transmission by the wireless module under on-site conditions. In this way, while ensuring the stable transmission of the system, the data quality required for algorithm analysis can be satisfied.

[0070] Specifically, the normal communication rate (unit: bytes per second) is: the length of a single data L 1 (unit: bytes) and the additional overhead per byte of data brought by the communication protocol L 2(Unit: byte) sum, multiplied by the sampling frequency (unit: Hertz).

[0071] Speed=(L 1 +L 2 )·SampleRate

[0072] The MCU drives the ADC converter to perform analog-to-digital conversion, and communicates bidirectionally with the ADC converter through a high-speed bus to read the digital vibration signal obtained by sampling; the wireless module communicates bidirectionally with the MCU through a high-speed bus and wirelessly transmits it to the wireless base station through the 2.4GHz frequency band; the upper computer software deployed on the industrial computer reads the vibration signal transmitted by the wireless base station to the industrial computer, as Figure 2 shown, and displays the signal waveform on the upper computer interface in real time, and displays relevant parameters such as communication quality, helping the operator to more intuitively understand the current working state of the system. The communication quality is defined as the ratio of the actual number of bytes Num RealReceved received per unit time to the number of bytes Num Total that should be received per unit time:

[0073] Quality=Num RealReceved / Num Total

[0074] At the same time, the upper computer software will regularly save data to the industrial computer for subsequent analysis algorithms to provide data.

[0075] Specifically, the working process of the mill load monitoring method of the above-mentioned mill load monitoring system for the vibration signal of the cylinder body is as follows:

[0076] Step 1: Segment the vibration signal according to the phase, and truncate the data in a single signal file in the form of a complete circle in time according to the preset phase start point and end point;

[0077] Step 2: Preprocess the vibration signal segmented by circle to obtain a signal that meets the basic requirements for input to the load recognition model;

[0078] The preprocessing includes: first subtracting the amplitude mean value of the vibration signal in the time domain to remove the trend item, then performing maximum absolute value normalization on the waveform amplitude so as to represent the vibration signal in relative intensity, and finally performing pre-emphasis processing on the high-frequency band of the vibration signal;

[0079] Step 3: Convert the preprocessed vibration signal from the time domain waveform to a time-frequency spectrogram and input it into the load recognition model of the mill to obtain the mill load output by the load recognition model;

[0080] The time-frequency spectrogram of the input model is a group of time-frequency spectrograms formed by superimposing three-layer time-frequency spectrograms. The frequency range of the first-layer time-frequency spectrogram only covers the key frequency bands divided based on prior knowledge. The frequency range of the second layer is the remaining frequency bands except the key frequency bands. The third layer is the time-frequency representation converted to the logarithmic scale. By adjusting the frame length and frame shift of the time-frequency spectrogram, it is ensured that each layer of the time-frequency spectrogram has the same dimensions in time and frequency. The three-layer time-frequency spectrograms have the same size and are independent of each other, regarded as three different sound channels. Since the amount of information input to the model by the three sound channels is the same, and the key frequency bands only account for a small part of all the frequency bands, this multi-channel design significantly improves the relative importance of the key frequency bands for model decision-making. At the same time, the logarithmic scale enhances the low-frequency resolution, making the input of the model contain more information. The load identification model based on the neural network can effectively extract and fuse the features in the above input information, so as to obtain a more accurate load identification result. As Figure 6 shown, the model finally classifies the load of the mill into five different situations according to the same standard as the industrial site: normal load, full grinding, empty grinding, partial full grinding, and partial empty grinding.

[0081] Specifically, the load identification model of the mill in step 3 includes an input convolution module, 3 noise reduction feature fusion residual modules based on spatial attention, and an output fully connected module connected in sequence as Figure 3 shown. Specifically, the input convolution module includes a convolution layer, a batch normalization (Batch Normal, BN) layer, a ReLU activation function, and a pooling layer. The convolution kernel size of the convolution layer is set to 5, so that the network has a large receptive field at the beginning. The pooling layer performs a max pooling operation.

[0082] Specifically, the noise reduction feature fusion residual module based on spatial attention (SAD-Res Block) includes two residual sub-modules and a max pooling layer. The residual sub-module includes a first convolution layer, a BN layer, a ReLU activation function, a second convolution layer, a BN layer, and a spatial attention-based denoising feature fusion (SAD) unit connected in sequence. After the output of the SAD unit is added to the input value directly sent from the input of the sub-module through the residual connection structure, it passes through a ReLU activation function to obtain the final output of the noise reduction feature fusion residual module based on spatial attention. Among them, the residual sub-module is Figure 4 , and the SAD unit is as Figure 5 shown.

[0083] Specifically, the SAD unit first performs global max pooling and global average pooling on the input features according to the channel dimension of the figure respectively, to obtain a group of 2×H×W pooling features [F MaxPool , F AvgPoolThis set of pooled features passes through two convolutional layers and a Sigmoid function to obtain a set of spatial attention α. The α is separated in the channel dimension into an attention A with a size of 1×H×W MaxPool and A AvgPool , and is multiplied element-wise with the corresponding pooled features. The average of these two sets of products is taken to obtain a set of hybrid attention A Mix . The channel dimension of A Mix is equally expanded so that it has the same size as the original feature map, and this set of expanded parameters is used as the noise reduction soft threshold of the original feature map. The thresholding function is shown as follows:

[0084]

[0085] where x is the input feature, y is the output feature, τ is the threshold, and is a positive parameter. The input feature is thresholded through the thresholding function, and the output feature map is the feature after noise reduction is achieved.

[0086] Specifically, the output fully connected module includes a fully connected layer and a softmax classifier, which are used to convert the extracted features into the final classification result of the mill load.

[0087] Next, a complete example is used to introduce the above-mentioned mill load monitoring system for the vibration signal of the cylinder body:

[0088] Refer to Figure 1 , in this embodiment, the system is composed of four parts: a signal transmitting end, a signal receiving end, a wireless charging power module, and an active space positioning module. The signal transmitting end includes a vibration sensor, a signal conditioning circuit, an ADC converter, an MCU, a wireless communication module, an antenna, and a 5V DCDC buck module; the signal receiving end includes an antenna, a wireless base station, an industrial computer, and a host computer software deployed on the industrial computer; the wireless charging power module includes a wireless charging transmitting part composed of a 48V switching power supply, a wireless charging transmitting host, and a wireless charging transmitting coil, and a wireless charging receiving part composed of a wireless charging receiving host and a wireless charging receiving coil; the active space positioning module includes a photoelectric sensor and a reflector. Among them, the signal transmitting end, the wireless charging receiving part, and the photoelectric sensor are installed inside the system housing, and the housing is fixed on the wall of the mill cylinder body and rotates synchronously with the operation of the mill.

[0089] Specifically, for an actual implementation case of a semi-autogenous mill in a certain concentrator, the diameter of the semi-autogenous mill is 7m, the length is 3.5m, and the rotational speed during operation is about 12 rpm. The on-site physical picture is as shown in Figure 7As shown in the figure. When the system is working, the housing rotates synchronously with the mill cylinder, measures the vibration on the cylinder wall along the circumferential direction of the mill cylinder, and at the same time determines the time when the housing passes the position of the mirror each time through the active space positioning module, so as to divide the phase corresponding to the vibration signal. When the system starts, the wireless module establishes a connection with the wireless base station, and then starts to transmit the signals collected by the vibration sensor. At this time, open the host computer software on the industrial control computer, and you can receive the signals transmitted by the system, see the real-time waveform, and at the same time the data will be saved to the disk of the industrial control computer. After being actually put into use on site, the system can stably collect the cylinder vibration signals of the semi-autogenous mill in the industrial field for a long time and the load monitoring is relatively accurate.

[0090] As a transformable implementation manner, in another example, specifically, for an actual implementation case of a ball mill in a certain ore dressing plant, the diameter of the ball mill is 5m, the length is 8m, and the rotational speed during operation is about 14rpm. When the system is working, measure the vibration of its cylinder along the circumferential direction of the ball mill, and determine the phase through the active space positioning module. The remaining settings can refer to Embodiment 1.

[0091] As Figure 8 As shown in the figure, the present application also provides a method for monitoring the load of a mill based on the cylinder vibration signal, which is applied to the above-mentioned mill load monitoring system based on the cylinder vibration signal. The method includes:

[0092] Use the signal transmitter to collect the vibration signal on the surface of the mill cylinder;

[0093] The signal receiver is used to receive the vibration signal and send the vibration signal to the signal processing end;

[0094] Use the wireless charging power module to supply power to the signal transmitter, signal receiver and active space positioning module;

[0095] Use the active space positioning module to calculate the phase of the mill cylinder in space in real time and send the phase to the signal processing end;

[0096] Use the signal processing end to calculate the mill load based on the vibration signal and the phase.

[0097] Optionally, the step of using the signal processing end to calculate the mill load based on the vibration signal and the phase includes:

[0098] Step 1: Segment the vibration signal according to the phase, and truncate the data in a single signal file in the form of a full circle in time according to the preset phase start point and end point;

[0099] Step 2: Preprocess the vibration signal segmented by circles to obtain a signal that meets the basic requirements for the input of the load recognition model;

[0100] The preprocessing includes: first, subtracting the amplitude mean of the vibration signal in the time domain to remove the trend term, then performing maximum absolute value normalization on the waveform amplitude so as to represent the vibration signal in relative intensity, and finally performing pre-emphasis processing on the high-frequency band of the vibration signal;

[0101] Step 3: Convert the preprocessed vibration signal from the time-domain waveform into a time-frequency spectrogram and input it into the load recognition model of the mill to obtain the mill load output by the load recognition model;

[0102] The input model's time-frequency spectrogram is a time-frequency spectrogram group formed by superimposing three-layer time-frequency spectrograms. Among them, the frequency range of the first-layer time-frequency spectrogram only covers the key frequency bands divided based on prior knowledge, the frequency range of the second layer is the remaining frequency bands except the key frequency bands, and the third layer is the time-frequency representation converted to the logarithmic scale; by adjusting the frame length and frame shift of the time-frequency spectrogram, ensure that each layer of the time-frequency spectrogram has the same dimensions in time and frequency; the three-layer time-frequency spectrograms have the same size, are independent of each other, and are regarded as three different channels.

[0103] The mill load monitoring based on the vibration signal of the cylinder can implement each embodiment of the above system and can achieve the same beneficial effects, which will not be elaborated here.

[0104] The preferred specific embodiments of the present invention have been described in detail above. It should be understood that those of ordinary skill in the art can make many modifications and variations according to the concept of the present invention without creative labor. Therefore, all technical solutions that can be obtained by those skilled in the art in the technical field of the present invention based on the concept of the present invention through logical analysis, reasoning, or limited experiments on the basis of the prior art should be within the protection scope determined by the claims.

Claims

1. A mill load monitoring system based on cylinder vibration signal, characterized in that: include: A signal transmitting end, a signal receiving end, a signal processing end, a wireless charging power module and an active space positioning module, wherein the signal transmitting end is arranged in a shell, and the shell is fixed to the surface of the mill cylinder and rotates synchronously with the cylinder during the operation of the mill; The signal transmitting end is used to collect the vibration signal of the surface of the mill cylinder; The signal receiving end is used to receive the vibration signal and send the vibration signal to the signal processing end; The wireless charging power supply module is used to supply power to the signal transmitting end, the signal receiving end and the active space positioning module; The active spatial positioning module is used to calculate the phase of the mill cylinder in space in real time and send the phase to the signal processing end; The signal processing end is used to calculate the mill load based on the vibration signal and the phase.

2. The mill load monitoring system based on cylinder vibration signal according to claim 1 is characterized in that: The signal transmitting end includes a vibration sensor, a signal conditioning circuit, an ADC converter, an MCU, a wireless communication module and a first antenna; The vibration sensor is arranged on the surface of the mill cylinder and rotates synchronously with the cylinder during the operation of the mill, and is used to convert the vibration of the mill cylinder surface into an analog electrical signal; the signal conditioning circuit is used to pre-process the analog electrical signal and then send it to a channel of the ADC converter; the ADC converter is used to sample the analog electrical signal input to the channel at a set sampling frequency to obtain a vibration signal; the MCU is used to drive the ADC converter to perform analog-to-digital conversion, and to communicate bidirectionally with the ADC converter through a high-speed bus to read the sampled vibration signal; the wireless communication module is used to perform bidirectional communication with the MCU through a high-speed bus, and to transmit the read information to the signal receiving end through the first antenna by setting the frequency band.

3. The mill load monitoring system based on cylinder vibration signal according to claim 2 is characterized in that: The signal transmitting end also includes a DCDC step-down module, and the DCDC step-down module is respectively connected to the signal conditioning circuit, the ADC converter, the MCU, and the wireless communication module to provide power.

4. The mill load monitoring system based on cylinder vibration signal according to claim 1 is characterized in that: The active spatial positioning module comprises a photoelectric sensor and a reflector, wherein the photoelectric sensor is arranged in the housing, and the reflector is installed at a position within a set range from the mill cylinder; the photoelectric sensor is used to emit infrared light outward, and when the infrared light is not reflected by any object, the photoelectric sensor outputs a low level; when the infrared light is reflected by the reflector and received by the photoelectric sensor, the photoelectric sensor outputs a high level; wherein the time between two high levels is regarded as a complete circle; the phase is calculated by the time difference between the current moment and the last triggering of the photoelectric sensor, and the ratio of the time interval between two adjacent triggerings; The phase calculation process satisfies the following relationship: Phase=(Time Current -Trigger1) / (Trigger2-Trigger1)·360; In the formula, Trigger2-Trigger1 represents the time interval between two adjacent high levels output by the photoelectric sensor, Time Current Indicates the current moment.

5. The mill load monitoring system based on cylinder vibration signal according to claim 3 is characterized in that: The wireless charging power supply module includes a switching power supply, a wireless charging transmitting host, a wireless charging transmitting coil, a wireless charging receiving host, and a wireless charging receiving coil; the switching power supply, the wireless charging transmitting host, and the wireless charging transmitting coil are arranged at a position within a set range from the grinding mill cylinder, and the switching power supply is connected to the wireless charging transmitting host for power supply; the wireless charging transmitting host is used to convert the input direct current into high-frequency alternating current, and transmit the high-frequency alternating current to the wireless charging transmitting coil; the wireless charging receiving coil identifies the wireless charging transmitting coil through magnetic field coupling and receives electric energy, and converts the high-frequency alternating current back into direct current through the wireless charging receiving host; and the output end of the wireless charging receiving host is connected in parallel to the input end of the DCDC step-down module.

6. The mill load monitoring system based on cylinder vibration signal according to claim 1 is characterized in that: The signal receiving end includes a second antenna and a wireless base station. The second antenna is communicatively connected with the first antenna. The second antenna receives the vibration signal from the first antenna and transmits it to the wireless base station.

7. The mill load monitoring system based on cylinder vibration signal according to claim 6 is characterized in that: The signal processing end includes an industrial computer, which is connected to the wireless base station. The industrial computer is used to read the vibration signal transmitted by the wireless base station, display the waveform of the vibration signal on the upper computer interface in real time, and display related parameters. The industrial computer is also used to identify the mill load using a load identification model. The input of the load identification model is the vibration signal and the phase position, and the output of the identification model is the mill load.

8. A method for monitoring a mill load based on a cylinder vibration signal, applied to a mill load monitoring system based on a cylinder vibration signal as claimed in any one of claims 1 to 7, characterized in that: The method comprises: The signal transmitter is used to collect the vibration signal on the surface of the mill cylinder; Using a signal receiving end to receive the vibration signal, and sending the vibration signal to a signal processing end; An active spatial positioning module is used to calculate the phase of the mill barrel in space in real time, and the phase is sent to the signal processing end; A wireless charging power module is used to power the signal transmitting end, the signal receiving end and the active space positioning module; A signal processing end is used to calculate the mill load based on the vibration signal and the phase.

9. The method according to claim 8, characterized in that The signal processing end is used to calculate the mill load based on the vibration signal and the phase, including: Step 1: Divide the vibration signal according to the phase, and truncate the data in a single signal file in the form of a full circle in time according to the preset phase starting point and end point; Step 2: Preprocess the vibration signal after being divided by circle to obtain a signal that meets the basic requirements of the load identification model input; The preprocessing includes: firstly, subtracting the amplitude mean of the vibration signal in the time domain to remove the trend term, then performing a maximum absolute value normalization process on the waveform amplitude so as to express the vibration signal in relative intensity, and finally performing a pre-emphasis process on the high frequency band of the vibration signal; Step 3: convert the preprocessed vibration signal from a time domain waveform into a time-frequency spectrum and input the spectrum into a load identification model of the mill to obtain the mill load output by the load identification model; The time-frequency spectrum of the input model is a time-frequency spectrum group formed by superimposing three layers of time-frequency spectrum, wherein the frequency range of the first layer of time-frequency spectrum only covers the key frequency bands divided based on prior knowledge, the frequency range of the second layer is the remaining frequency bands except the key frequency bands, and the third layer is a time-frequency representation converted to a logarithmic scale; by adjusting the frame length and frame shift of the time-frequency spectrum, it is ensured that each layer of the time-frequency spectrum has the same dimension in time and frequency; the three layers of time-frequency spectrum are consistent in size and independent of each other, and are regarded as three different channels.

Images