Anomaly detection system, anomaly detection method, and program

Abstract

In the flotation process, the state of bubbles is determined over a wide area. [Solution] The abnormality detection system 1 according to this disclosure determines the state of bubbles in the flotation process. The abnormality detection system 1 includes an acoustic sensor 10 that detects the popping sound of bubbles generated in the flotation cell and generates an acoustic signal based on the popping sound of the bubbles, and a signal processing device 20. The signal processing device 20 includes a processor 26 that performs filtering to extract a specific frequency band from the acoustic signal and determines whether the state of the bubbles is normal or abnormal based on the filtered acoustic signal.

Description

Technical Field

[0001] The present disclosure relates to an abnormality detection system, an abnormality detection method, and a program.

Background Art

[0002] Conventionally, when separating minerals, a flotation process is widely used. In the flotation process, air is fed into an ore slurry to generate bubbles containing the target mineral, thereby separating useful minerals from unnecessary substances.

[0003] Patent Document 1 discloses a method of acquiring image information of bubbles (froth) generated in flotation and determining the state of the bubbles based on the image information in the flotation process.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In the flotation process, when trying to photograph an image of the generated bubbles, it is difficult to photograph a wide range due to the limitation of the camera's angle of view. Therefore, it is difficult to determine the state of the bubbles over a wide range.

[0006] Therefore, an object of the present disclosure is to provide an abnormality detection system, an abnormality detection method, and a program capable of determining the state of bubbles over a wide range in the flotation process.

Means for Solving the Problems

[0007] [1] An abnormality detection system for determining the state of bubbles in a flotation process, An acoustic sensor that detects the popping sound of bubbles generated in a flotation cell and generates an acoustic signal based on the popping sound of the bubbles, Signal processing device and Equipped with, The signal processing device is A filtering process is performed to extract a specific frequency band from the aforementioned acoustic signal. A processor that determines whether the state of the bubbles is normal or abnormal based on the filtered acoustic signal. An anomaly detection system equipped with the following features. Such anomaly detection systems make it possible to determine the state of bubbles over a wide range in the flotation process.

[0008] [2] In the anomaly detection system described in [1] above, The aforementioned processor, A fast Fourier transform is performed on the acoustic signal acquired from the acoustic sensor to generate frequency spectrum data. Filtering may be performed on the frequency spectrum data to extract the specific frequency band. This allows for the extraction of specific frequency components from the popping sound of bubbles and the removal of external noise.

[0009] [3] In the anomaly detection system described in [1] or [2] above, The processor may perform adaptive filtering when executing the filtering, dynamically adjusting the filtering parameters according to the environment. This allows for more effective removal of external noise.

[0010] [4] In the anomaly detection system described in any one of the above items [1] to [3], The aforementioned environment may also be an anomaly detection system characterized by external noise. This allows filtering parameters to be dynamically adjusted according to the real-time characteristics of external noise.

[0011] [5] In the anomaly detection system described in any one of the above items [1] to [4], The signal processing device further includes a memory that stores a learning model in which the normal and abnormal states of the bubble are learned by machine learning based on past acoustic signals. The processor may input the filtered acoustic signal to the learning model to determine whether the state of the bubble is normal or abnormal. This allows for highly accurate determination of whether the bubble state is normal or abnormal using a machine learning model.

[0012] [6] An abnormality detection method for determining the state of bubbles in a flotation process, The steps include detecting the sound of bubbles bursting in a flotation cell and generating an acoustic signal based on the sound of the bubbles bursting, The steps include: performing filtering to extract a specific frequency band from the aforementioned acoustic signal; A step of determining whether the state of the foam is normal or abnormal based on the filtered acoustic signal, An anomaly detection method, including the above. This anomaly detection method makes it possible to determine the state of bubbles over a wide range in the flotation process.

[0013] [7] A program for determining the state of bubbles in a flotation process, A step of acquiring an acoustic signal based on the popping sound of bubbles generated in a flotation cell, The steps include: performing filtering to extract a specific frequency band from the aforementioned acoustic signal; A step of determining whether the state of the foam is normal or abnormal based on the filtered acoustic signal, A program that causes a computer to perform an action that includes [a specific action]. Such a program makes it possible to determine the state of bubbles over a wide range in the flotation process. [Effects of the Invention]

[0014] According to the present disclosure, it is possible to provide an abnormality detection system, an abnormality detection method, and a program that can determine the state of bubbles in a wide range in a flotation process.

Brief Description of the Drawings

[0015] [Figure 1] It is a diagram showing a schematic configuration of an abnormality detection system according to an embodiment. [Figure 2] It is a diagram showing functional blocks of a processor according to an embodiment. [Figure 3] It is a flowchart showing an example of the operation of an abnormality detection system according to an embodiment.

Modes for Carrying Out the Invention

[0016] Hereinafter, an embodiment of the present disclosure will be described with reference to the drawings.

[0017] FIG. 1 is a diagram showing a schematic configuration of an abnormality detection system 1 according to an embodiment. The abnormality detection system 1 can determine the state of bubbles in a flotation process. The abnormality detection system 1 includes an acoustic sensor 10 and a signal processing device 20.

[0018] The acoustic sensor 10 detects the bursting sound of bubbles generated in a flotation process executed in the flotation cell 100. <s [[ID=3x]]

[0019] The flotation cell 100 is a container in which the flotation process is executed. In the flotation process executed in the flotation cell 100, for example, air is sent into the ore slurry, and bubbles containing the target mineral are generated. Since the behavior of the bubbles affects the efficiency of the flotation process, it is important to grasp the state of the bubbles.

[0020] The acoustic sensor 10 may be any acoustic sensor capable of detecting the sound of bubbles bursting. The acoustic sensor 10 may be, for example, a directional microphone that has directionality toward the flotation cell 100. In this case, the acoustic sensor 10 can detect the sound of bubbles bursting that is occurring in the flotation cell 100 with emphasis.

[0021] The signal processing device 20 may be a dedicated electronic device configured to function as the signal processing device 20 of the anomaly detection system 1, or it may be a general-purpose electronic device such as a tablet terminal or a PC (Personal Computer).

[0022] The signal processing device 20 comprises an AD converter 21, a communication unit 22, a memory 23, an input unit 24, an output unit 25, and a processor 26.

[0023] The AD converter 21 converts the analog acoustic signal acquired from the acoustic sensor 10 into a digital acoustic signal and outputs it to the processor 26. The AD converter 21 may be an AD converter with any configuration.

[0024] The communication unit 22 includes a communication module. For example, the communication unit 22 may include a communication module that corresponds to a LAN (Local Area Network). In one embodiment, the signal processing device 20 is connected to a network via the communication unit 22. The communication unit 22 may be able to communicate with a server or the like via the network.

[0025] Memory 23 is, for example, a semiconductor memory, magnetic memory, or optical memory, but is not limited to these. Memory 23 may function as, for example, a main memory, an auxiliary memory, or a cache memory. Memory 23 stores any information used in the operation of the signal processing device 20. For example, memory 23 may store system programs, application programs, and various types of information. Part of memory 23 may be installed outside the signal processing device 20. In that case, the part of memory 23 installed outside may be connected to the signal processing device 20 via any interface.

[0026] The input unit 24 includes one or more input interfaces that detect user input and acquire input information based on user operations. For example, the input unit 24 includes, but is not limited to, physical keys, capacitive keys, a touchscreen integrated with the display of the output unit 25, or a microphone that accepts voice input.

[0027] The output unit 25 includes one or more output interfaces for outputting information and notifying the user. For example, the output unit 25 includes, but is not limited to, a display for outputting information as an image, a speaker for outputting information as sound, etc.

[0028] The processor 26 is a general-purpose processor such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit), or a dedicated processor specialized for a specific process. The processor 26 may also be an FPGA (Field-Programmable Gate Array), a DSP (Digital Signal Processor), or an ASIC (Application Specific Integrated Circuit). The processor 26 may also be a combination of multiple processors. The processor 26 controls each part of the signal processing device 20 and executes processes related to the operation of the signal processing device 20.

[0029] The functions of the signal processing device 20 can be realized by executing a program stored in the memory 23 using the processor 26. In other words, the functions of the signal processing device 20 can be realized by software.

[0030] Figure 2 shows the functional blocks of the functions that the processor 26 can execute. The processor 26 can execute filtering, anomaly detection, and learning functions by reading programs stored in memory 23 and executing the read programs.

[0031] (Operation of the anomaly detection system) Next, we will explain the operation of the anomaly detection system 1.

[0032] In flotation cell 100, the flotation process is performed. When the flotation process is performed, bubbles containing the target mineral are generated. The generated bubbles burst after a certain amount of time has passed since their generation.

[0033] The acoustic sensor 10 is positioned to detect the popping sound of bubbles generated during the flotation process performed in the flotation cell 100.

[0034] When the acoustic sensor 10 detects the popping sound of bubbles generated in the flotation cell 100, it generates an acoustic signal based on the popping sound of the bubbles. The acoustic sensor 10 transmits the generated acoustic signal to the signal processing device 20.

[0035] The AD converter 21 of the signal processing device 20 receives an analog acoustic signal from the acoustic sensor 10 and converts the acquired analog acoustic signal into a digital acoustic signal. The AD converter 21 outputs the converted digital acoustic signal to the processor 26.

[0036] The processor 26 performs a Fast Fourier Transform on the acoustic signal, which has been converted into a digital signal, to generate frequency spectrum data.

[0037] The processor 26 performs filtering on the generated frequency spectrum data to extract specific frequency bands. The processor 26 can perform filtering to extract specific frequency bands, for example, by functioning as a bandpass filter. By extracting specific frequency bands in this way, the processor 26 can extract specific frequency components of the bubble burst sound and remove external noise.

[0038] When performing filtering, the processor 26 may perform adaptive filtering, which dynamically adjusts the filtering parameters according to the environment. For example, the processor 26 may perform adaptive filtering by dynamically adjusting the filtering parameters according to the characteristics of the external noise. This allows the processor 26 to more effectively remove external noise because it can dynamically adjust the filtering parameters according to the real-time characteristics of the external noise.

[0039] When the processor 26 performs filtering, it determines whether the bubble state is normal or abnormal based on the filtered acoustic signal. For example, the processor 26 may use a machine learning model, which has been trained on past acoustic signals, to determine whether the bubble state is normal or abnormal.

[0040] A machine learning model, trained based on past acoustic signals, is stored in memory 23. The machine learning model may take filtered acoustic signals as input and output whether the bubble state is normal or abnormal. The machine learning model may have been pre-trained using past acoustic signals when the bubble state is normal and past acoustic signals when the bubble state is abnormal as training data.

[0041] The processor 26 can determine whether the bubble state is normal or abnormal by inputting the filtered acoustic signal into a learning model stored in memory 23.

[0042] Alternatively, the processor 26 may determine whether the bubble state is normal or abnormal based on the filtered acoustic signal by performing pattern matching. In this case, the memory 23 may store the acoustic signal pattern for when the bubble state is normal and the acoustic signal pattern for when the bubble state is abnormal.

[0043] If the processor 26 determines that the state of the bubbles is abnormal, it may cause the output unit 25 to emit an alarm. The alarm may be a display alarm or an audible alarm. Alternatively, if the processor 26 determines that the state of the bubbles is abnormal, it may transmit a notification to another device via the communication unit 22 indicating that the state of the bubbles is abnormal.

[0044] The operation of the anomaly detection system 1 will be explained with reference to the flowchart shown in Figure 3.

[0045] Step S101: When the acoustic sensor 10 detects the popping sound of bubbles generated in the flotation cell 100, it generates an acoustic signal based on the popping sound of the bubbles. The acoustic sensor 10 transmits the generated acoustic signal to the signal processing device 20.

[0046] Step S102: The processor 26 of the signal processing device 20 performs a Fast Fourier Transform on the acoustic signal acquired from the acoustic sensor 10 to generate frequency spectrum data. The acoustic signal acquired from the acoustic sensor 10 may have been previously converted into a digital signal by the AD converter 21.

[0047] Step S103: The processor 26 performs filtering to extract specific frequency bands from the generated frequency spectrum data. In this case, the processor 26 may perform adaptive filtering, which dynamically adjusts the filtering parameters according to the environment. For example, the processor 26 may perform adaptive filtering by dynamically adjusting the filtering parameters according to the characteristics of the external noise. This allows the processor 26 to more effectively remove external noise by dynamically adjusting the filtering parameters according to the real-time characteristics of the external noise.

[0048] Step S104: After performing filtering, the processor 26 determines whether the state of the bubbles is normal or abnormal based on the filtered acoustic signal.

[0049] According to the anomaly detection system 1 of this embodiment described above, it is possible to determine the state of bubbles over a wide range in the flotation process. More specifically, the anomaly detection system 1 includes an acoustic sensor 10 that detects the popping sound of bubbles generated in the flotation cell 100 and generates an acoustic signal based on the popping sound of the bubbles, and a signal processing device 20. The signal processing device 20 includes a processor 26 that performs filtering to extract a specific frequency band from the acoustic signal and determines whether the state of the bubbles is normal or abnormal based on the filtered acoustic signal. In this way, the acoustic sensor 10 detects the popping sound of bubbles, but since there are no constraints on the field of view, such as those in the method of capturing images of bubbles, the acoustic sensor 10 can detect the popping sound of bubbles over a wide range. Therefore, according to the anomaly detection system 1 of this embodiment, it is possible to determine the state of bubbles over a wide range in the flotation process.

[0050] Furthermore, while methods that capture images of bubbles are difficult to use in dark places, the abnormality detection system 1 according to this embodiment can detect the sound of bubbles bursting even in dark places.

[0051] Furthermore, while workers checking the state of the foam can be subjective, the abnormality detection system 1 according to this embodiment determines the state of the foam based on the sound of the foam bursting detected by the acoustic sensor 10, thus enabling objective and quantitative determination of the foam's state.

[0052] Furthermore, the anomaly detection system 1 according to this embodiment performs filtering to extract a specific frequency band from the acoustic signal, thereby reducing the influence of external noise and emphasizing the sound of bubbles bursting. Therefore, it can contribute to improving the efficiency of the selection process.

[0053] It will be apparent to those skilled in the art that this disclosure can be implemented in other predetermined forms besides the embodiments described above without deviating from its spirit or essential features. Therefore, the prior description is illustrative and not limiting. The scope of the disclosure is defined not by the prior description but by the added claims. Any modifications within their equivalent scope are incorporated therein.

[0054] For example, the arrangement and number of each component described above are not limited to those shown in the above description and drawings. The arrangement and number of each component may be configured arbitrarily, as long as it can achieve its function.

[0055] For example, it is also possible to configure a general-purpose electronic device such as a smartphone or computer to function as the signal processing device 20 according to the above embodiment. Specifically, a program describing the processing content that realizes each function of the signal processing device 20 according to the embodiment can be stored in the memory of the electronic device, and the processor of the electronic device can read and execute the program. Therefore, the disclosure according to one embodiment can also be realized as a program that can be executed by a processor.

[0056] For example, some of the processing performed by the signal processing device 20 in the embodiment described above may be performed by a server that can communicate with the signal processing device 20. For instance, the processing up to the filtering stage may be performed by the signal processing device 20, and the processing to determine whether the state of the bubbles is normal or abnormal based on the filtered acoustic signal may be performed by the server. In this case, the server may store a machine learning model.

[0057] For example, when the signal processing device 20 determines whether the state of the bubbles is normal or abnormal, it may determine whether the state of the bubbles is normal or abnormal by counting the number of times the bubbles burst. Alternatively, the signal processing device 20 may determine whether the state of the bubbles is normal or abnormal based on the acoustic level of the sound of the bubbles bursting. [Explanation of Symbols]

[0058] 1. Anomaly detection system 10 Acoustic sensors 20 Signal Processing Equipment 21 AD Converters 22 Communications Department 23 memory 24 Input section 25 Output section 26 processors 100 flotation cells

Claims

1. An anomaly detection system for determining the state of bubbles in a flotation process, An acoustic sensor that detects the popping sound of bubbles generated in a flotation cell and generates an acoustic signal based on the popping sound of the bubbles, Signal processing device and Equipped with, The signal processing device is A filtering process is performed to extract a specific frequency band from the aforementioned acoustic signal. A processor that determines whether the state of the bubbles is normal or abnormal based on the filtered acoustic signal. An anomaly detection system equipped with the following features.

2. In the anomaly detection system according to claim 1, The aforementioned processor, A fast Fourier transform is performed on the acoustic signal acquired from the acoustic sensor to generate frequency spectrum data. An anomaly detection system that performs filtering to extract a specific frequency band from the frequency spectrum data.

3. In the anomaly detection system according to claim 1, The processor is an anomaly detection system that performs adaptive filtering, dynamically adjusting the filtering parameters according to the environment when performing the filtering.

4. In the anomaly detection system according to claim 3, The aforementioned environment is an anomaly detection system, which is a characteristic of external noise.

5. In the anomaly detection system according to claim 1, The signal processing device further includes a memory that stores a learning model in which the normal and abnormal states of the bubble are learned by machine learning based on past acoustic signals. The processor is an anomaly detection system that inputs the filtered acoustic signal to the learning model to determine whether the state of the bubble is normal or abnormal.

6. An abnormality detection method for determining the state of bubbles in a flotation process, The steps include detecting the sound of bubbles bursting in a flotation cell and generating an acoustic signal based on the sound of the bubbles bursting, The steps include: performing filtering to extract a specific frequency band from the aforementioned acoustic signal; A step of determining whether the state of the foam is normal or abnormal based on the filtered acoustic signal, An anomaly detection method, including the above.

7. In the abnormality detection method described in claim 6, The method further includes the step of performing a fast Fourier transform on the generated acoustic signal to generate frequency spectral data, The step of performing the filtering is an anomaly detection method that performs filtering to extract the specific frequency band from the frequency spectrum data.

8. In the abnormality detection method described in claim 6, The anomaly detection method includes the step of performing the filtering, which involves performing adaptive filtering that dynamically adjusts the filtering parameters according to the environment when performing the filtering.

9. In the abnormality detection method described in claim 8, The aforementioned environment is an anomaly detection method that is a characteristic of external noise.

10. In the abnormality detection method described in claim 6, An anomaly detection method that determines whether the state of the foam is normal or abnormal by inputting the filtered acoustic signal into a learning model that has been trained to recognize the normal and abnormal states of the foam based on past acoustic signals, thereby determining whether the state of the foam is normal or abnormal.

11. A program for determining the state of bubbles in the flotation process, A step of acquiring an acoustic signal based on the popping sound of bubbles generated in a flotation cell, The steps include: performing filtering to extract a specific frequency band from the aforementioned acoustic signal; A step of determining whether the state of the foam is normal or abnormal based on the filtered acoustic signal, A program that causes a computer to perform an action that includes [a specific action].

12. In the program described in claim 11, The computer is instructed to perform an operation that further includes the step of performing a Fast Fourier Transform on the acquired acoustic signal to generate frequency spectrum data. The step of performing the filtering is a program that performs filtering on the frequency spectrum data to extract the specific frequency band.

13. In the program described in claim 11, The step of performing the filtering is a program that performs adaptive filtering, dynamically adjusting the filtering parameters according to the environment when performing the filtering.

14. In the abnormality detection method described in claim 13, The aforementioned environment is a program with external noise characteristics.

15. In the program described in claim 11, The step of determining whether the state of the foam is normal or abnormal is a program that inputs the filtered acoustic signal into a learning model that has been trained to recognize the normal and abnormal states of the foam based on past acoustic signals, and determines whether the state of the foam is normal or abnormal.

Images