Control device, turbidimeter, determination method, and learning method

Abstract

The present application provides a control device, a turbidimeter, a determination method, and a learning method. The control device (20) includes a control unit (21) that determines turbidity based on a light-receiving waveform of light received via a measured liquid, acquires waveform data representing the light-receiving waveform of the received light when the determined turbidity rises above a threshold value, executes an identification process for identifying a kind of a substance mixed in the measured liquid using the acquired waveform data as input, thereby acquiring an identification result of the kind of the substance, and determines the kind of the substance with reference to the acquired identification result.

Description

[0001] This application claims priority based on Japanese application No. 2022-052047, filed on March 28, 2022, and all disclosures of that earlier application are incorporated herein by reference. Technical Field

[0002] This disclosure relates to control devices, turbidimeters, determination methods, and learning methods. Background Technology

[0003] A turbidimeter is known to perform optical measurement of the turbidity of test liquids such as industrial water. Turbidity is an indicator of the degree of turbidity in water. Measurement methods used in turbidimeters can be classified as transmitted light method, scattered light method, transmitted-scattered light method, surface scattered light method, and integrating sphere method, among others. In the transmitted-scattered light method, light incident from the light source onto the test liquid is divided into transmitted light and scattered light for detection. In the surface scattered light method, light is shone onto the surface of the test liquid, and the scattered light from the surface is measured. In the integrating sphere method, the ratio of the intensity of transmitted light (or total incident light) to the intensity of scattered light is calculated.

[0004] Patent document 1 describes a technique related to a turbidity meter that optically determines turbidity based on the amount of transmitted and scattered light.

[0005] Existing technical documents

[0006] Patent documents

[0007] Patent Document 1: Japanese Patent Application Publication No. 2006-329629 Summary of the Invention

[0008] The technical problem that the invention aims to solve

[0009] In the technology described in Patent Document 1, when turbidity changes, in order to determine the cause, users such as equipment administrators need to visually observe the state of the liquid being measured.

[0010] The purpose of this disclosure is to make it easy to determine the cause of a change in turbidity even without visually observing the state of the liquid being tested.

[0011] Technical solutions for solving technical problems

[0012] Several embodiments of the control device include a control unit that measures turbidity based on the light-receiving waveform of light received through the test liquid, acquires waveform data representing the light-receiving waveform received when the measured turbidity exceeds a threshold and rises, uses the acquired waveform data as input, performs identification processing to identify the types of substances mixed in the test liquid, thereby acquiring identification results of the types of substances, and determines the types of substances by referring to the acquired identification results. In such a control device, the types of substances mixed in the test liquid can be determined based on the light-receiving waveform. Therefore, even without visually observing the state of the test liquid, the cause of the change in turbidity can be easily determined.

[0013] In one embodiment, the control unit of the control device measures turbidity based on the received waveform of light after passing through a low-pass filter. As waveform data, it acquires data representing the received waveform of light before passing through the low-pass filter when turbidity exceeds a threshold and increases. According to this embodiment, turbidity measurement is performed based on the received waveform of light with a limited frequency, thus suppressing the influence of noise and improving the accuracy of the measurement results. On the other hand, the determination of substances mixed in the test liquid is based on the received waveform of light with an unrestricted frequency, thus enabling more accurate identification of the types of substances. This is because the received waveform of light with an unrestricted frequency is considered to more significantly reflect the differences in the types of substances. Therefore, even without visually observing the state of the test liquid, it is easier to determine the cause of changes in turbidity with greater accuracy.

[0014] In one embodiment, the control unit of the control device performs the following process as an identification process: if data representing a light-receiving waveform is input, it refers to a database defining the correspondence between multiple waveform patterns and the types of substances mixed in the liquid through which the light passes, respectively, to determine the type of substance corresponding to the waveform pattern of the light-receiving waveform represented by the input data; and as the identification result, it acquires data representing the type of substance determined by referring to the database. According to this embodiment, the type of substance is determined by referring to a database, thus enabling high-speed identification processing.

[0015] In one embodiment, the control unit of the control device performs a recognition process using a learned model. This learned model takes data representing a light-receiving waveform as input and outputs data representing the types of substances mixed in the liquid through which the light corresponding to the input waveform passes. The data output from the learned model is then acquired as the recognition result. According to this embodiment, the type of substance is determined using a learned model, thus improving recognition accuracy through learning.

[0016] In one embodiment, the control unit of the control device notifies the user of the determination result of the type of substance. According to this embodiment, the user can understand the cause of the change in turbidity, thus making it easier to explore measures to prevent recurrence and to prevent the abnormality from recurring.

[0017] In one embodiment, the control unit of the control device notifies the user of the determination result by displaying the determination result on a display screen. According to this embodiment, the user can visually understand the reason for the change in turbidity.

[0018] In one embodiment, the control unit of the control device, upon determining that the substance is air bubbles based on the identification results, notifies the user that the substance is air bubbles. Air bubbles are a cause of error in turbidity measurements, and it is desirable to physically remove them during turbidity measurement using a degassing device such as a degassing tank. However, according to this embodiment, the user is notified even if air bubbles are not completely removed or are mixed into the liquid being measured. This allows the user to recognize that errors are present in the turbidity meter's measurement. Furthermore, the user can correct turbidity errors and adjust the measurement results.

[0019] In one embodiment, the control unit of the control device acquires imaging data of the liquid being measured and notifies the user of the acquired imaging data along with the determination result. According to this embodiment, the user can confirm the type of substance determined based on the waveform even by visual inspection, and the accuracy of the analysis can be improved by correlating the imaging data with the determination result.

[0020] In one embodiment, the received light waveform represented by waveform data includes the waveform of transmitted light that passes through the test liquid and the waveform of scattered light that is reflected and scattered by the substance. According to this embodiment, identification processing for identifying the type of substance is performed based on a combination of the transmitted light waveform and the scattered light waveform. Therefore, even without visually observing the state of the test liquid, it is easier and more accurate to determine the cause of changes in turbidity.

[0021] Several embodiments of the turbidimeter include: the aforementioned control device; and a light-receiving device that receives light passing through the test liquid and inputs data representing the received light waveform to the control device. In such a turbidimeter, the type of substance mixed into the test liquid can be determined based on the light waveform. Therefore, even without visually observing the state of the test liquid, the cause of the change in turbidity can be easily determined.

[0022] Several implementation methods involve a measurement method that includes: measuring turbidity based on the light-receiving waveform of light received through the test liquid; acquiring waveform data representing the light-receiving waveform received when the measured turbidity exceeds a threshold and rises; and using the acquired waveform data as input, performing identification processing to identify the type of substance mixed in the test liquid, thereby determining the type of substance. In such a determination method, the type of substance mixed in the test liquid can be determined based on the light-receiving waveform. Therefore, even without visually observing the state of the test liquid, the cause of the change in turbidity can be easily determined.

[0023] Several implementation methods involve a learning method that includes: acquiring waveform data representing the light received via a liquid mixed with various substances, using multiple types of substances; creating teacher data by associating the acquired waveform data with type data representing the type of substance for each type of substance; and generating a learned model by performing machine learning using the created teacher data, wherein the learned model takes the data representing the light received waveform as input and outputs data representing the type of substance mixed in the liquid through which the light corresponding to the light received waveform represented by the input data passes. In this learning method, the light received waveform is associated with the type of substance for learning. By using the model obtained through this learning method, the type of substance can be identified with high accuracy based on the light received waveform.

[0024] The effects of the invention

[0025] According to this disclosure, even without visually observing the state of the liquid being measured, it is easy to determine the cause of the change in turbidity. Attached Figure Description

[0026] Figure 1 This is a diagram showing the appearance of a turbidimeter as one embodiment of this disclosure.

[0027] Figure 2 This is a diagram showing an example of the configuration of a turbidimeter as one embodiment of this disclosure.

[0028] Figure 3 This is a diagram illustrating an example configuration of a current-to-voltage conversion amplifier circuit in a turbidimeter, which is one embodiment of this disclosure.

[0029] Figure 4 This is a diagram illustrating an example of waveform data involved in an embodiment of this disclosure.

[0030] Figure 5 This is a block diagram illustrating the configuration of the control device involved in the embodiments of this disclosure.

[0031] Figure 6This is a flowchart illustrating the operation of the control device according to the embodiments of this disclosure.

[0032] Figure 7 This is a diagram showing the generation order of the learned model according to another embodiment of this disclosure.

[0033] Figure 8 This is a diagram showing a modified example of the configuration of a turbidimeter as a scheme of this disclosure.

[0034] Explanation of reference numerals in the attached figures

[0035] 10. Turbidity meter;

[0036] 20 control devices;

[0037] 21. Control Department;

[0038] 22. Storage Unit;

[0039] 23 Ministry of Communications;

[0040] 24 Input Section;

[0041] 25 output units;

[0042] 26 Low-pass filter;

[0043] 27AD converter;

[0044] 30. Light receiving device;

[0045] 31. Light-receiving element;

[0046] 32 optical receivers;

[0047] 33 Processing circuit;

[0048] 40 body;

[0049] 41. The solution to be tested;

[0050] 401 is the outlet of the liquid being measured;

[0051] 402 is the inlet of the solution to be measured;

[0052] 42 network cameras;

[0053] 50 light source devices;

[0054] 51 light sources;

[0055] 52 lenses;

[0056] 60 coaxial cable;

[0057] Waveform data for 70, 71, 72, 73, and 74;

[0058] TL transmitted light;

[0059] SL scattered light. Detailed Implementation

[0060] The present disclosure will now be described with reference to the accompanying drawings.

[0061] In the accompanying drawings, the same or equivalent parts are labeled with the same reference numerals. In the description of this disclosure, descriptions of the same or equivalent parts are appropriately omitted or simplified.

[0062] Reference Figure 1 and Figure 2 The configuration of the turbidimeter 10, which is one embodiment of this disclosure, will be described.

[0063] The turbidimeter 10, as one embodiment of this disclosure, is a device for measuring turbidity based on the light waveform received by the liquid being measured 41. Hereinafter, an example of a turbidimeter 10 using the transmitted light scattering method will be described.

[0064] The turbidimeter 10 includes a body 40, a control device 20, a light receiving device 30, and a light source device 50. The body 40 has a liquid inlet 402 for the liquid to be measured 41 to flow into the body 40 and a liquid outlet 401 for the liquid to flow out.

[0065] The light source device 50 includes a light source 51 and a lens 52. The light source 51 is, for example, a light bulb or an LED. "LED" is an abbreviation for light-emitting diode.

[0066] The light-receiving device 30 includes one or more light-receiving elements 31. Figure 2 The diagram shows three light-receiving elements 31, but the number of light-receiving elements 31 can be arbitrary and not limited to three. Each light-receiving element 31 converts the received light into an electrical signal. For example, a photodiode is a light-receiving element 31.

[0067] The control device 20 is a computer. The control device 20 may be a dedicated device, a mobile device, a general-purpose device such as a PC, or a server device belonging to a cloud computing system or other computing system. "PC" is an abbreviation for personal computer. Mobile devices include mobile phones, smartphones, or tablets.

[0068] The control device 20 is housed inside the turbidimeter 10. Alternatively, the control device 20 can be located outside the turbidimeter 10 and communicate with it wirelessly or via a wired connection. Communication is conducted via a network such as a LAN or the Internet. "LAN" is an abbreviation for local area network. For example, communication can use communication interfaces corresponding to mobile communication standards such as LTE, 4G, or 5G, wireless LAN communication standards such as IEEE 802.11, or wired LAN communication standards such as Ethernet. "LTE" is an abbreviation for Long Term Evolution. "4G" is an abbreviation for 4th generation. "5G" is an abbreviation for 5th generation. "IEEE" is an abbreviation for the Institute of Electrical and Electronics Engineers.

[0069] The light receiving device 30 can communicate with the control device 20 directly or via a network such as a LAN or the Internet. For example, the light receiving device 30 can communicate with the control device 20 using a communication interface that corresponds to mobile communication standards such as LTE, 4G, and 5G, wireless LAN communication standards such as IEEE 802.11, or wired LAN communication standards such as Ethernet.

[0070] Reference Figure 2 The operating principle of turbidimeter 10 will be explained.

[0071] In the turbidimeter 10, the liquid to be measured 41 passes through the interior of the body 40 in the direction of arrow X. Light irradiated onto the liquid to be measured 41 by the light source 51 of the light source device 50 is received by the light receiving element 31 of the light receiving device 30 via the lens 52 and the liquid to be measured 41. Here, the light incident on the liquid to be measured 41 is received by the light receiving element 31 as transmitted light TL passing through the liquid to be measured 41 in a straight line and scattered light SL scattered by substances mixed in the liquid to be measured 41. The light received by the light receiving element 31 is output to the control device 20 after passing through the current-to-voltage conversion and amplification circuit described later. The control device 20 measures the turbidity of the liquid to be measured 41 based on the waveform of the output voltage value, i.e., the received light waveform.

[0072] Reference Figure 3 An example of the configuration of a current-to-voltage conversion amplifier circuit will be explained.

[0073] Figure 3The current-to-voltage conversion amplifier circuit shown is configured as a circuit from the front-end amplifier of the light-receiving device 30 to the CPU of the control device 20. "CPU" is an abbreviation for Central Processing Unit.

[0074] A current-to-voltage conversion amplifier circuit is mounted on analog substrate BD1 and CPT substrate BD2. "CPT" is an abbreviation for Capacitive Power Transfer. Analog substrate BD1 and CPT substrate BD2 are connected via a coaxial cable 60. Figure 3 In the diagram, the upper part represents the signal path of the transmitted light TL, and the lower part represents the signal path of the scattered light SL. One of the signal paths for the scattered light SL is omitted from the diagram. Both the transmitted light TL and the scattered light SL are processed in the same way. On the signal paths of the transmitted light TL and the scattered light SL on the analog substrate BD1, a light-receiving element 31, an optical receiver 32, and a processing circuit 33 are respectively provided. The optical receiver 32 is, for example, a TIA (Transimpedance Amplifier). The processing circuit 33 is, for example, a linear drive amplifier. Data is transmitted to the optical receiver 32 via a serial bus. The serial bus is, for example, a two-wire serial bus such as an I2C bus (Inter-Integrated Circuit). On the signal paths of the transmitted light TL and the scattered light SL on the CPT substrate BD2, a low-pass filter 26 and an AD converter 27 are respectively provided. AD is an abbreviation for Analog to Digital. Figure 3 In this configuration, analog substrate BD1 is disposed on light-receiving device 30, and CPT substrate BD2 is disposed on control device 20. However, this disclosure is not limited to this configuration. Any configuration can be used as long as it can convert the current signal from light-receiving element 31 into a voltage signal, and then into a digital signal and output it to the CPU, as described later. Furthermore, three light-receiving elements 31, three optical receivers 32, three processing circuits 33, three low-pass filters 26, and three AD converters 27 are each provided. Figure 3 Each element represents two, but the number of each element is not limited to this and can be arbitrarily selected according to the number and type of light signals received.

[0075] exist Figure 3In the process, the light passing through the measured liquid 41 is received by the light-receiving element 31 and converted into a current signal with a current value corresponding to the light intensity. This current signal is then converted into a voltage signal with a voltage value corresponding to the light intensity by the subsequent light receiver 32. After being amplified by the processing circuit 33, this voltage signal is converted into an AD signal by the subsequent AD converter 27 via the coaxial cable 60 after the high-frequency components have been reduced in the low-pass filter 26, and then output to the CPU.

[0076] When measuring turbidity, the cutoff frequency of the low-pass filter 26 is typically lowered to prevent signals other than DC components from being input to the AD converter 27. This is to prevent signal aliasing caused by components, especially AC components, contained in the signal before AD conversion, and to improve the measurement accuracy by suppressing errors in the turbidity measurement results.

[0077] Hereinafter, several implementation methods will be described as specific examples of this disclosure.

[0078] (First Implementation)

[0079] Reference Figure 4 The outline of this embodiment will be described below.

[0080] The control device 20 measures turbidity based on the received light waveform of the light received via the test liquid 41. The control device 20 acquires waveform data 70, which represents the received light waveform when the measured turbidity exceeds a threshold and rises. Using the acquired waveform data 70 as input, the control device 20 performs identification processing to identify the types of substances mixed in the test liquid 41, thereby determining the type of substance. The control device 20 then notifies the user of the determination result regarding the type of substance.

[0081] It is understood that when different types of substances are measured optically, the resulting light waveforms will differ depending on parameters such as the size, transparency, concentration, or mobility of the measured object. The inventors, focusing on this, discovered that the types of substances mixed in the measured liquid 41 can be identified based on the light waveform obtained when turbidity increases.

[0082] In this embodiment, waveform data 70 represents raw data showing the waveforms of transmitted light and scattered light as observed using a measuring instrument such as an oscilloscope. When observing the waveforms, the resolution, response, or storage medium usage can be adjusted by changing the sampling rate as needed. Alternatively, sound waves can be generated based on the waveforms. Specifically, the type or volume of the generated sound can be changed according to the type or size of the waveform. As an example, timbre can be distinguished using transmitted and scattered light.

[0083] Figure 4In the example of waveform data 70, we show the first waveform 71, the second waveform 72, the third waveform 73, and the fourth waveform 74. In each waveform, the solid line represents the waveform of transmitted light, the dashed line represents the waveform of scattered light, the horizontal axis represents time, and the vertical axis represents the voltage value. The first waveform 71 is an example of a waveform observed when bonito flakes are added to the test liquid 41, assuming that fallen leaves are mixed into the test path. The second waveform 72 is an example of a waveform observed when kaolinite is added to the test liquid 41, assuming that sludge is mixed into the test path. "Kaolinite" is the common name for clay minerals whose main component is kaolinite (Al2O3·2SiO2·2H2O). The third waveform 73 is an example of a waveform observed when soil collected from the ground is added to the test liquid 41, assuming that sand is mixed into the test path. The fourth waveform 74 is an example of a waveform observed when carbonated water is added to the test liquid 41, assuming that air bubbles are mixed into the test path. Referring to the first waveform 71, the second waveform 72, the third waveform 73, and the fourth waveform 74, it can be seen that the obtained light waveform varies depending on the type of substance mixed in the measured liquid 41.

[0084] According to this embodiment, the type of substance is determined based on waveform data 70. This successfully identifies air bubbles or foreign matter mixed in the measured liquid that would have been undetectable without visual observation until now. Previously, even when turbidity increased, it was impossible to determine whether the cause was air bubbles or foreign matter such as sludge. Therefore, users such as equipment administrators needed to investigate the cause of the turbidity increase whenever an anomaly occurred. However, in cases where the turbidity meter is remotely set up, there are situations where the anomaly disappears by the time the user arrives on-site. In such cases, the cause cannot be determined. Because the cause cannot be determined, it is difficult to explore measures to prevent recurrence, and as a result, the anomaly may recur. According to this embodiment, even without visually observing the state of the measured liquid 41, the cause of the turbidity change can be easily determined. This reduces the burden on the user. Furthermore, it is easier for the user to explore measures to prevent the anomaly from recurring.

[0085] Reference Figure 5 The configuration of the control device 20 involved in this embodiment will be described.

[0086] The control device 20 includes a control unit 21, a storage unit 22, a communication unit 23, an input unit 24, and an output unit 25.

[0087] The control unit 21 includes at least one processor, at least one programmable circuit, at least one dedicated circuit, or any combination thereof. The processor is a general-purpose processor such as a CPU or GPU, or a dedicated processor for specific processing. "GPU" is an abbreviation for Graphics Processing Unit. The programmable circuit is, for example, an FPGA. "FPGA" is an abbreviation for Field-Programmable Gate Array. The dedicated circuit is, for example, an ASIC. "ASIC" is an abbreviation for Application-Specific Integrated Circuit. The control unit 21 controls the various parts of the control device 20 while performing processing related to the operation of the control device 20.

[0088] Storage unit 22 includes at least one semiconductor memory, at least one magnetic memory, at least one optical memory, or any combination thereof. Semiconductor memory is, for example, RAM, ROM, or SSD. RAM is an abbreviation for random access memory. ROM is an abbreviation for read-only memory. SSD is an abbreviation for solid-state drive. RAM is, for example, SRAM or DRAM. SRAM is an abbreviation for static random access memory. DRAM is an abbreviation for dynamic random access memory. ROM is, for example, EEPROM. EEPROM is an abbreviation for electrically erasable programmable read-only memory. Magnetic memory is, for example, HDD. HDD is an abbreviation for hard disk drive. Storage unit 22 functions as, for example, a primary storage device, an auxiliary storage device, or flash memory. Storage unit 22 stores data used in the operation of control device 20 and data obtained through the operation of control device 20. In this embodiment, waveform data 70 is stored in the storage unit 22. As a variation of this embodiment, the waveform data 70 may also be stored in an external system.

[0089] The communication unit 23 includes at least one communication interface. This communication interface may be, for example, a LAN interface, an interface corresponding to mobile communication standards such as LTE, 4G, or 5G, or an interface corresponding to wireless LAN communication standards such as IEEE 802.11 or wired LAN communication standards such as Ethernet (for external networks). The communication unit 23 receives data used in the operation of the control device 20 and transmits data obtained through the operation of the control device 20.

[0090] The input unit 24 includes at least one input interface. The input interface may be, for example, a physical button, a capacitive button, a click device, a touchscreen integrated with the display, a camera, or a microphone. The input unit 24 receives operations that input data used in the operation of the control device 20. The input unit 24 may also be separate from the control device 20 and connected to it as an external input device. As a connection interface, for example, an interface compatible with standards such as USB, HDMI, or Bluetooth can be used. "USB" is an abbreviation for Universal Serial Bus. "HDMI" is an abbreviation for High-Definition Multimedia Interface.

[0091] Output unit 25 includes at least one output interface. The output interface may be, for example, a display or a speaker. The display may be, for example, an LCD or an organic EL display. "LCD" is an abbreviation for liquid crystal display. "EL" is an abbreviation for electroluminescence. Output unit 25 outputs data obtained through the operation of control device 20. Output unit 25 may also be independent of control device 20 and connected to control device 20 as an external output device. For example, an interface compatible with standards such as USB, HDMI, or Bluetooth can be used for connection. An interface compatible with industrial wireless standards such as ISA100 or LoRaWAN can also be used. "ISA" is an abbreviation for International Society of Automation. "LoRaWAN" is an abbreviation for Long Range Wide Area Network.

[0092] The function of the control device 20 is implemented by executing the program according to this embodiment in the processor, which serves as the control unit 21. That is, the function of the control device 20 is implemented by software. The program causes the computer to perform the actions of the control device 20, thereby enabling the computer to function as the control device 20. In other words, the computer functions as the control device 20 by executing the actions of the control device 20 according to the program.

[0093] Programs can be pre-stored on non-transitory computer-readable media. Examples of non-transitory computer-readable media include flash memory, magnetic recording devices, optical discs, optical-magnetic recording media, or ROM. Program distribution can occur, for example, through the sale, transfer, or rental of portable media such as SD cards, DVDs, or CD-ROMs containing the program. "SD" is an abbreviation for Secure Digital. "DVD" is an abbreviation for Digital Versatile Disc. "CD-ROM" is an abbreviation for Compact Disc Read Only Memory. Programs can also be distributed by storing them in server storage and transferring them from the server to other computers. Programs can also be provided as program products.

[0094] Computers may temporarily store programs stored on portable media or transferred from servers in main storage. The computer then reads the program from main storage in its processor and executes the processing according to the read program. Alternatively, the computer can directly read programs from portable media and execute the processing according to the program. The computer can also execute the processing according to the received program each time it is transferred from a server. Processing can also be performed using so-called ASP-type services, which do not involve transferring programs from the server to the computer but only perform the execution of instructions and the retrieval of results. "ASP" is an abbreviation for Application Service Provider. A program includes information for the computer to process, i.e., data that follows the program. For example, data that, while not direct instructions to the computer, has the nature of specifying the computer's processing corresponds to "data that follows the program."

[0095] Some or all of the functions of the control device 20 can also be implemented by a programmable circuit or a dedicated circuit that serves as the control unit 21. That is, some or all of the functions of the control device 20 can be implemented by hardware.

[0096] Reference Figure 6 The operation of the control device 20 according to this embodiment will be explained. This operation is equivalent to the determination method according to this embodiment.

[0097] When the turbidimeter 10 is started or when the measurement begins after startup, the light source 51 of the light source device 50 is driven, and the light source 51 illuminates the liquid to be measured 41. The light illuminated by the light source 51 is incident on the liquid to be measured 41. A portion of the light incident on the liquid to be measured 41 is reflected by substances mixed in the liquid to be measured 41. The light receiving element 31 of the light receiving device 30 receives the light that has passed through the liquid to be measured 41. The light receiving element 31 converts the received light into a current signal. The current signal from the light receiving element 31 is converted into a voltage signal by a current-to-voltage conversion amplifier circuit and output to the control device 20.

[0098] In step S101, the control unit 21 of the control device 20 measures the turbidity based on the received light waveform of the light received via the liquid to be measured 41. Specifically, the control unit 21 measures the turbidity based on the received light waveform after it has passed through the low-pass filter 26. This is because measuring the turbidity based on the received light waveform, which is limited by frequency, can suppress the influence of noise and other factors, thereby suppressing errors in the measurement results.

[0099] In step S102, the control unit 21 of the control device 20 determines whether the measured turbidity exceeds a threshold value and thus increases. Specifically, the control unit 21 compares the measured turbidity with a threshold value. The threshold value can be arbitrarily set, for example, a water quality standard value determined by laws such as the Japanese Waterway Law, water quality standard values ​​of various countries, or a value based on WHO guidelines. "WHO" is an abbreviation for World Health Organization. If the measured turbidity is greater than the threshold value, the process proceeds to step S103. If the measured turbidity is below the threshold value, the process returns to step S101.

[0100] In step S103, the control unit 21 of the control device 20 acquires waveform data 70 representing the received light waveform. Specifically, the control unit 21 acquires data representing the received light waveform before passing through the low-pass filter 26 as waveform data 70. The reason for acquiring data representing the received light waveform before passing through the low-pass filter 26 as waveform data 70 is that the received light waveform, which is not frequency-limited, is considered to more significantly reflect the different types of substances. Even though a waveform with frequency domain limitations after passing through the low-pass filter 26 can be used, from the viewpoint of sensitivity and response time, it is desirable to use the received light waveform before passing through the low-pass filter 26. The signal of the light before passing through the low-pass filter 26, for example, can be obtained from... Figure 3The extraction from the front side of the low-pass filter 26 in the current-to-voltage conversion amplifier circuit shown can also be achieved by separately configuring a circuit that excludes the low-pass filter 26 from the current-to-voltage conversion amplifier circuit. Furthermore, the extracted signal can be sampled into a digital signal using a separate dedicated AD converter, or it can be sampled from a conventional AD converter.

[0101] In step S104, the control unit 21 of the control device 20 takes the acquired waveform data 70 as input and performs identification processing to identify the types of substances mixed in the liquid being measured 41. Specifically, the control unit 21 of the control device 20 performs the following processing: if data representing the received waveform is input, it refers to a database DB that defines the correspondence between multiple waveform patterns and the types of substances mixed in the liquid through which the light corresponding to each of the multiple waveform patterns passes, and determines the type of substance corresponding to the waveform pattern of the received waveform represented by the input data.

[0102] In this embodiment, the waveform pattern can be defined, for example, as a combination of parameters such as the amplitude, period, and DC value of the received waveform. As the received waveform, when data representing both transmitted and scattered light is input, the control unit 21 of the control device 20 can determine the waveform pattern of each received waveform and combine them into a single pattern for comparison in the database DB. As an example, suppose "Pattern A" is determined for the received waveform of transmitted light included in a certain received waveform RW, and "Pattern B" is determined for the received waveform of scattered light. The control unit 21 can also compare the "waveform pattern AB," representing the combination of "Pattern A" and "Pattern B," as the waveform pattern of the received waveform RW in the database DB. In the database DB, numerical ranges for the parameters of the corresponding waveform patterns are predefined for each type of substance, such as fallen leaves, sludge, sand, and bubbles. Therefore, if the parameter values ​​of "waveform pattern AB" are within the numerical ranges of the parameters of the waveform patterns defined in the database DB for a certain type of substance, the type of substance is determined to be the type of substance corresponding to "waveform pattern AB."

[0103] In this embodiment, for example, dried bonito flakes are used instead of fallen leaves, kaolin clay is used instead of sludge, ground soil is used instead of sand, and carbonated water is used instead of bubbles. A waveform pattern is predetermined for the light received by the liquid containing each type of substance. Then, a table defined in advance, as a database DB, is prepared to associate the determined waveform pattern with the type of substance, and used in step S104.

[0104] As a variation of this embodiment, the control unit 21 of the control device 20 may further acquire images of the measured liquid 41 captured when each waveform is observed, and display the acquired images to the user. The images can be acquired using any imaging device. For example, the control unit 21 may also acquire images using a device such as... Figure 8 The network camera 42, installed in the main body 40 or within the piping, captures data that can be remotely observed via a network. Capture can also be performed under normal water pressure. The captured data can be either still or moving images. The data can be either black and white or color. By displaying the captured data when the waveform is observed to the user, the user can visually confirm the type of substance identified based on the waveform and verify the accuracy of the comparison results in the database DB.

[0105] In this embodiment, the database DB is pre-built in the storage unit 22 of the control device 20. Alternatively, the database DB may be built in the cloud. The control unit 21 of the control device 20 refers to the database DB to determine the type of substance corresponding to the waveform pattern of the input light-receiving waveform.

[0106] In step S105, the control unit 21 of the control device 20 acquires the result of the identification process performed in step S104 as the identification result. Specifically, the control unit 21 acquires data indicating the type of substance determined in step S104 as the identification result.

[0107] In step S106, the control unit 21 of the control device 20 determines the type of substance mixed into the test liquid 41 by referring to the identification result obtained in step S105. Specifically, the control unit 21 determines that the substance of the type indicated by the data obtained in step S105 has mixed into the test liquid 41.

[0108] In step S107, the control unit 21 of the control device 20 notifies the user of the determination result of the type of substance from step S106. Specifically, the control unit 21 displays the determination result on the display, which is the output unit 25. The control unit 21 may also output the determination result as sound through the speaker, which is the output unit 25. The control unit 21 may also send the notification via a network. For example, the notification may be sent to a terminal device installed in the management office or to a terminal device carried by the user, i.e., the manager.

[0109] In this embodiment, if the control unit 21 of the control device 20 determines that the type of substance is bubbles in step S106 based on the identification result, it notifies the user that the type of substance is bubbles in step S107. Alternatively, if the control unit 21 of the control device 20 determines that the type of substance is not bubbles based on the identification result in step S106, it may also notify the user in step S107 of the specific type of substance, such as fallen leaves, sludge, or sand.

[0110] Because air bubbles can introduce errors into turbidity measurements, they have traditionally been physically removed during turbidity measurements using degassing devices such as degassing tanks. According to this embodiment, the user is notified when air bubbles are mixed into the measured liquid 41, making it easy for the user to recognize that the turbidity meter 10's measurement contains errors. Furthermore, the user can correct turbidity errors and adjust the measurement results.

[0111] As described above, in this embodiment, the control unit 21 of the control device 20 measures turbidity based on the light-receiving waveform received via the test liquid 41. The control unit 21 acquires waveform data 70, which represents the light-receiving waveform received when the measured turbidity exceeds a threshold and rises. The control unit 21 uses the acquired waveform data 70 as input and performs identification processing to identify the types of substances mixed in the test liquid 41, thereby determining the types of substances. Therefore, according to this embodiment, the cause of the change in turbidity can be determined even without visually observing the state of the test liquid 41.

[0112] (Second Implementation)

[0113] Reference Figure 7 The general outline of this embodiment will be described below. Hereinafter, the differences between this embodiment and the first embodiment will be mainly explained.

[0114] In the first embodiment described above, the control unit 21 of the control device 20 performs identification processing using a database DB as step S104. In contrast, in this embodiment, the control unit 21 of the control device 20 performs identification processing using a learned model for identifying the types of substances mixed in the test liquid 41 as step S104. Specifically, the control unit 21 performs processing using a learned model that takes data representing the received light waveform as input and outputs data representing the types of substances mixed in the liquid through which the light corresponding to the received light waveform represented by the input data passes. The control unit 21 inputs the waveform data 70 obtained in step S103 into the learned model for identifying the types of substances mixed in the test liquid 41, and obtains the identification result of the types of substances from the learned model.

[0115] The learned model can be generated using any learning method, but in this embodiment, it is generated using the following learning method.

[0116] Reference Figure 7 The generation order of the learned model is explained. This order corresponds to the learning method involved in this implementation.

[0117] In step S201, the control unit 21 of the control device 20 acquires waveform data. In this embodiment, the waveform data is data representing the received waveform of light received via a liquid mixed with various types of substances. The waveform data may include audio data representing sound corresponding to the received waveform.

[0118] In step S202, the control unit 21 of the control device 20 generates teacher data. Specifically, for each type of substance, the control unit 21 generates teacher data by associating the waveform data acquired in step 201 with type data representing the substance type. For example, the control unit 21 associates the acquired waveform data with type data representing the substance type by using the substance type as a tag to establish a connection between the waveform data and the type data.

[0119] Assuming that it was obtained in step S201 Figure 4 The data representing each waveform is shown, and the order in which the teacher's data was generated is explained in detail. In this embodiment, the control unit 21 of the control device 20 receives input from the user via the input unit 24 regarding the type of substance introduced when each waveform is observed, and associates the type data representing the input type with the first waveform 71, the second waveform 72, the third waveform 73, and the fourth waveform 74. Specifically, the control unit 21 associates data representing "fallen leaves" as type data with the first waveform 71. The control unit 21 associates data representing "sludge" as type data with the second waveform 72. The control unit 21 associates data representing "sand" as type data with the third waveform 73. The control unit 21 associates data representing "bubbles" as type data with the fourth waveform 74.

[0120] As a variation of this embodiment, instead of requiring the user to input the type of substance, the control unit 21 of the control device 20 can acquire images of the measured liquid 41 captured when each waveform is observed and automatically identify the type of substance by analyzing the acquired images using known methods. Image analysis can be performed, for example, by AI (Artificial Intelligence). Images can be acquired using any imaging device. For example, the control unit 21 can also acquire imaging data captured by a network camera 42 installed in the main body 40 or piping, thereby enabling remote observation of the imaging data via a network.

[0121] In step S203, a learned model is generated. Specifically, machine learning is performed using the teacher data created in step S202 to generate a learned model that takes data representing the light-receiving waveform as input and outputs data representing the types of substances mixed in the liquid through which the light corresponding to the light-receiving waveform represented by the input data passes. Machine learning can be performed using known machine learning algorithms such as neural networks or deep learning.

[0122] In this embodiment, after the model has completed learning, it outputs the label corresponding to the category that matches the type of substance as the recognition result. Specifically, it outputs the label corresponding to the category that matches the type of substance shown in the category data of the four categories: "fallen leaves", "sludge", "sand", and "bubbles".

[0123] In this embodiment, the control unit 21 of the control device 20 performs the recognition process in step S104 using the learned model described above, and obtains data representing the type of substance as the recognition result.

[0124] As described above, in this embodiment, the control unit 21 of the control device 20 performs a recognition process using a learned model. This learned model takes data representing the received light waveform as input and outputs data representing the types of substances mixed in the liquid through which the light corresponding to the received light waveform represented by the input data passes. The control unit 21 acquires the data output from the learned model as the recognition result.

[0125] According to this embodiment, the control unit 21 of the control device 20 uses a learned model to determine the type of substance. Since the learned model is used to determine the type of substance, the recognition accuracy can be improved through learning.

[0126] This disclosure is not limited to the embodiments described above. For example, the multiple blocks shown in the block diagram can be integrated, or a single block can be divided. Alternatively, instead of executing the multiple steps described in the flowchart sequentially as described, the steps can be executed in parallel or in a different order, depending on the processing capability of the apparatus executing each step or as needed. Furthermore, modifications can be made without departing from the scope of this disclosure.

[0127] For example, as one embodiment of this disclosure, a turbidimeter 10 based on the transmission and scattering light method is illustrated. However, the above embodiments are not limited to any device that optically measures the state of the liquid 41 being measured. For example, it can be applied to any turbidimeter based on the transmission light method, the scattering light method (e.g., right angle scattering type), the transmission and scattering light method, the surface scattering light method, or the integrating sphere method.

[0128] For example, as one embodiment of this disclosure, a turbidity meter 10 is described, but the above embodiment is applicable to any device that performs optical measurement of the state of the liquid 41 being measured, and is not limited to the turbidity meter 10. For example, it can be applied to any device such as a residual chlorine meter, pH meter, differential pressure / pressure gauge, flow meter, or colorimeter.

Claims

1. A control device, characterized in that, Equipped with a control unit, The control unit, Turbidity is measured based on the received waveform of light after passing through a low-pass filter and being received by the liquid being tested. Acquire waveform data representing the received light waveform before passing through the low-pass filter when the measured turbidity is determined to exceed a threshold and rises. The acquired waveform data is used as input to perform identification processing to identify the types of substances mixed in the test liquid, thereby obtaining the identification results of the types of substances. Based on the obtained identification results, the type of the substance is determined.

2. The control device according to claim 1, The control unit performs the following processing as part of the identification process: if data representing a light-receiving waveform is input, it refers to a database that defines the correspondence between multiple waveform patterns and the types of substances mixed in the liquid through which the light corresponding to each of the multiple waveform patterns passes, determines the type of substance corresponding to the waveform pattern of the light-receiving waveform represented by the input data, and obtains data representing the type of substance determined by referring to the database as the identification result.

3. The control device according to claim 1, The control unit performs a process using a learned model as part of the recognition process. This learned model takes data representing the light-receiving waveform as input and outputs data representing the types of substances mixed in the liquid through which the light corresponding to the light-receiving waveform represented by the input data passes. The data output from the learned model is then obtained as the recognition result.

4. The control device according to any one of claims 1 to 3, The control unit notifies the user of the determination result of the type of substance.

5. The control device according to claim 4, The control unit notifies the user of the determination result by displaying the determination result on the display screen.

6. The control device according to claim 4, If the control unit determines that the substance is a bubble based on the identification result, it will notify the user that the substance is a bubble.

7. The control device according to claim 4, The control unit acquires the imaging data of the liquid being tested and notifies the user of the acquired imaging data along with the judgment result.

8. The control device according to any one of claims 1 to 3, The received light waveform represented by the waveform data includes the waveform of transmitted light that passes through the liquid being measured and the waveform of scattered light that is reflected and scattered on the substance.

9. A turbidimeter, comprising: The control device according to any one of claims 1 to 8; The light receiving device receives light that has passed through the liquid being measured and inputs data representing the received light waveform to the control device.

10. A determination method, characterized in that, include: Turbidity is measured based on the waveform of light received by the liquid being measured and passed through a low-pass filter. Acquire waveform data representing the received light waveform before passing through the low-pass filter when the measured turbidity is determined to exceed the threshold and rises; The acquired waveform data is used as input to perform identification processing to identify the types of substances mixed in the test liquid, thereby determining the types of substances.

11. A learning method, characterized in that, include: Turbidity is measured using multiple types of substances, based on the received light waveform after passing through a low-pass filter and being received by a liquid containing various types of substances. Acquire waveform data representing the received light waveform before passing through the low-pass filter when the measured turbidity is determined to exceed the threshold and rises; For each type of substance, teacher data is created by associating the acquired waveform data with the type data representing the substance. By using the generated teacher data for machine learning, the following learned model is generated. This learned model takes data representing the light waveform as input and outputs data representing the types of substances mixed in the liquid through which the light corresponding to the light waveform represented by the input data passes.

Images