Computer program, information processing method, information processing device, and information processing system

Abstract

To provide a computer program or the like that can enhance accuracy in estimating moisture content of biomass fuel.SOLUTION: A computer program is configured to make a computer perform processing of: acquiring time-series measurement data of sound or vibration generated at a supply device of a biomass combustion facility, where biomass fuel is supplied to a combustion chamber through the supply device; and estimating moisture content of the biomass fuel on the basis of intensity of one or more frequency components contained in the acquired time-series measurement data.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present invention relates to a computer program, an information processing method, an information processing device, and an information processing system. [Background technology]

[0002] Biomass power generation is a well-known method of generating electricity by using biological resources (biomass) as fuel. In biomass power generation, the calorific value of the biomass and the temperature of the combustion chamber change depending on the moisture content of the biomass fuel being burned in the combustion chamber.

[0003] As an example of technology related to such biomass power generation, Patent Document 1 discloses a biomass incineration system having a drying chamber, a combustion chamber, a fluidized bed containing a fluidized medium that flows between the lower part of the drying chamber and the lower part of the combustion chamber, and a partition wall that separates the space above the fluidized bed in the drying chamber from the space above the fluidized bed in the combustion chamber and has a lower end located within the fluidized bed. In the biomass incineration system of Patent Document 1, the moisture content of the biomass in the drying chamber is confirmed using a steam flow meter that measures the steam flow rate of the drying exhaust gas discharged outside the drying chamber, and based on the measurement result, a first flow control valve that adjusts the flow rate of the drying exhaust gas discharged outside the drying chamber and a second flow control valve that adjusts the flow rate of the combustion exhaust gas discharged outside the combustion chamber are adjusted. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2017-187199 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the technique described in Patent Document 1 has a problem in that the accuracy of estimating the moisture content of biomass fuel is insufficient.

[0006] An object of the present disclosure is to provide a computer program or the like that can improve the accuracy of estimating the moisture content of biomass fuel. [Means for solving the problem]

[0007] A computer program according to one aspect of the present disclosure acquires time-series measurement data of sound or vibration generated by a supply device in a biomass combustion facility in which biomass fuel is supplied to a combustion chamber via the supply device, and causes a computer to execute a process of estimating the moisture content of the biomass fuel based on the intensity of one or more frequency components in the acquired time-series measurement data.

[0008] An information processing method according to one aspect of the present disclosure involves a computer acquiring time-series measurement data of sound or vibration generated by a supply device in a biomass combustion facility in which biomass fuel is supplied to a combustion chamber via the supply device, and estimating the moisture content of the biomass fuel based on the intensity of one or more frequency components in the acquired time-series measurement data.

[0009] An information processing device according to one aspect of the present disclosure includes an acquisition unit that acquires time-series measurement data of sound or vibration generated by a supply device in a biomass combustion facility in which biomass fuel is supplied to a combustion chamber via the supply device, and an estimation unit that estimates the moisture content of the biomass fuel based on the intensity of one or more frequency components in the acquired time-series measurement data.

[0010] An information processing system according to one aspect of the present disclosure includes a sensor installed in a biomass combustion facility in which biomass fuel is supplied to a combustion chamber via a supply device, and an information processing device. The information processing device includes an acquisition unit that acquires time-series measurement data of sound or vibration generated by the supply device measured by the sensor, and an estimation unit that estimates the moisture content of the biomass fuel based on the intensity of one or more frequency components in the acquired time-series measurement data. [Effects of the Invention]

[0011] According to the present disclosure, it is possible to improve the accuracy of estimating the moisture content of biomass fuel. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a schematic diagram of an estimation system. [Figure 2] FIG. 1 is a block diagram illustrating an example of the configuration of an estimation device. [Figure 3] FIG. 10 is an explanatory diagram illustrating a method for estimating the moisture content. [Figure 4] FIG. 10 is a diagram showing an example of the contents of correspondence information. [Figure 5] 10 is a flowchart illustrating an example of a processing procedure executed by the estimation device. [Figure 6] FIG. 10 is an explanatory diagram showing an overview of a learning model in the second embodiment. [Figure 7] 10 is a flowchart illustrating an example of a processing procedure executed by an estimation device according to a third embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0013] The present disclosure will be specifically described with reference to the drawings showing embodiments thereof.

[0014] (First embodiment) FIG. 1 is a schematic diagram of an estimation system 100. The estimation system 100 includes a biomass combustion facility 1 and an estimation device 2. The biomass combustion facility 1 of this embodiment constitutes a biomass power generation plant. FIG. 1 shows the biomass combustion facility 1 as seen from the side. In the following, various configurations of the biomass combustion facility 1 will be described with the left side in FIG. 1 being the "front" and the right side being the "rear."

[0015] Biomass power plants are power generation facilities that generate electricity using steam obtained by directly burning biomass as fuel using a direct fuel method. Examples of biomass fuel (hereinafter also referred to as biomass fuel) include wood chips, bark, wood pellets, palm kernel shells (PKS), livestock manure, rice husk, corn meal, and combinations thereof.

[0016] The biomass combustion facility 1 mainly comprises a supply device 10, a combustion chamber 5, and a control device 6. The supply device 10 includes a conveying unit 3 and a supply unit 4. Biomass fuel stored in a storage facility (not shown) is supplied and transported to the combustion chamber 5 via the conveying unit 3 and the supply unit 4, and is burned in the combustion chamber 5. When multiple types of biomass fuel are used, they may be mixed and stirred in advance before being supplied to the conveying unit 3.

[0017] The transport unit 3 includes a receiving hopper that receives biomass fuel (not shown), a flight conveyor, etc. The biomass fuel put into the receiving hopper is transported at a predetermined transport speed by the flight conveyor, etc., and supplied to the supply unit 4.

[0018] The supply unit 4 includes a hopper 41 and a chute 42 provided below the hopper 41. The hopper 41 is a container that receives the biomass fuel transported and supplied from the transport unit 3. The hopper 41 includes an inlet 411 that opens upward, a storage unit 412, an outlet 413 that opens downward, a chain conveyor 414 disposed within the storage unit 412, and a rotor 415 for breaking up the fuel. The chute 42 is connected to the outlet 413.

[0019] Biomass fuel fed into the inlet 411 falls onto a chain conveyor 414 located below the inlet 411, and is transported from front to rear by the chain conveyor 414 and rotors 415. The biomass fuel transported to the end 414a on the downstream side in the transport direction falls from the end 414a due to gravity, passes through the hopper 41, and is supplied to the chute 42.

[0020] The hopper 41 may also be provided with a bridge release device (not shown) for releasing the bridge of the hopper 41. By driving the bridge release device, clogging in the hopper 41 due to the bridge phenomenon can be released.

[0021] The chute 42 is a guide path that uses gravity to guide the biomass fuel supplied from the hopper 41 to the combustion chamber 5. The chute 42 has a wall portion 421 and an extension portion 422 that extends obliquely downward from the wall portion 421.

[0022] The wall portion 421 has a rear wall 421a extending in a substantially vertical direction and a front wall 421b extending at a downward incline. The extension portion 422 has, for example, a prismatic shape with a rectangular cross section. The front wall 421b and / or one surface of the extension portion 422 extending from the front wall 421b constitute the bottom surface of the chute 42. The biomass fuel dropped and supplied from the hopper 41 passes through the internal space of the chute 42, falls near the bottom surface of the chute 42 vertically below the end portion 414a, and is then guided along the bottom surface. Note that the shape of the extension portion 422 is not limited to a prismatic shape as long as it is hollow, and it may be, for example, cylindrical.

[0023] The biomass fuel guided by the chute 42 is introduced into the combustion chamber 5 via a feeder 51 provided in the combustion chamber 5. The feeder 51 is equipped with, for example, an inlet through which the biomass fuel carried out from the chute 42 is introduced and an inlet through which combustion air is introduced (neither of which are shown), and uses the combustion air to feed the biomass fuel introduced into the feeder 51 together with the secondary air into the combustion chamber 5. The feeder 51 is configured so that the flow rate of the combustion air introduced into the feeder 51 can be varied. This makes it possible to adjust the position at which the biomass fuel is introduced into the combustion chamber 5. The combustion chamber 5 is also equipped with a blower 52 that sends out primary air, and the biomass fuel is combusted using the primary air sent from the blower 52 through an inlet (not shown) through which the primary air is introduced.

[0024] The control device 6 is a computer that controls the operation of the biomass combustion facility 1. The control device 6 receives settings of operating conditions for the biomass combustion facility 1, and controls the operation of the biomass combustion facility 1 based on the received operating conditions. Note that the configurations of the biomass combustion facility 1, the hopper 41, and the chute 42 are merely examples, and are not limited to the above example.

[0025] The sensor 7 detects sound or vibrations generated in the supply unit 4 in response to the transport of biomass fuel. The sensor 7 is provided, for example, on the outside of the bottom surface of the chute 42. In this embodiment, the sensor 7 is a sound sensor that measures the physical quantity of sound in the chute 42. A specific example of a sound sensor is a microphone. The physical quantity measured by the sound sensor is, for example, frequency, sound pressure, amplitude, etc. The sensor 7 is connected to the estimation device 2 and outputs the measurement data of the detected sound directly or indirectly to the estimation device 2. The time-series measurement data of the sound obtained by the sensor 7 is represented as waveform data that indicates the temporal change of the sound.

[0026] In this embodiment, the sound measured by sensor 7 is the sound generated when biomass fuel is dropped into chute 42 and collides with the chute 42. The sound measured by sensor 7 is expected to change depending on the moisture content of the biomass fuel. When biomass fuel with a high moisture content collides with chute 42, a relatively low sound, i.e., a sound in the low frequency range, is likely to be generated. On the other hand, when biomass fuel with a low moisture content collides with chute 42, a relatively high sound, i.e., a sound in the high frequency range, is likely to be generated.

[0027] The sensor 7 can be provided at any appropriate position as long as it can detect sound in the supply unit 4. From the viewpoint of accurately detecting sound generated by collision between the biomass fuel and the chute 42, the sensor 7 is preferably provided at the position where the biomass fuel falls in the chute 42 or in the vicinity of this position. The position where the biomass fuel falls refers to the point where the biomass fuel, dropped and supplied from the hopper 41, collides with the chute 42. The sensor 7 may be provided on the outside of the bottom of the chute 42, at a position opposite the position where the biomass fuel falls on the inside of the bottom of the chute 42 or in the vicinity of this position. The sensor 7 may be detachably attached to the supply unit 4 or may be fixed to the supply unit 4. Multiple sensors 7 may be provided to detect sound at multiple locations.

[0028] As described above, sensor 7 may be a vibration sensor that detects vibrations in chute 42. In this case, sensor 7 obtains waveform data that indicates temporal changes in vibration as time-series measurement data of the vibration. Examples of vibration sensors include acceleration sensors, speed sensors, displacement sensors, and AE (Acoustic Emission) sensors, with acceleration sensors being preferred. Sensor 7 may include both a sound sensor and a vibration sensor. When both a sound sensor and a vibration sensor are installed, the sensors may be located in the same position or in different positions.

[0029] The estimation device 2 is a device capable of various information processing and information transmission and reception, such as a server computer, a personal computer, or a quantum computer. The estimation device 2 estimates the moisture content of biomass fuel based on measurement data received from the sensor 7. The estimation device 2 is an example of an information processing device. Note that the estimation device 2 and the control device 6 are not limited to being separate devices, and the estimation device 2 and the control device 6 may be a single processing device.

[0030] 2 is a block diagram showing an example configuration of the estimation device 2. The estimation device 2 includes a control unit 21, a storage unit 22, a communication unit 23, a display unit 24, an operation unit 25, and an input / output unit 26. The estimation device 2 may be a multi-computer consisting of multiple computers, or may be a virtual machine virtually constructed by software.

[0031] The control unit 21 includes one or more arithmetic processing devices such as a central processing unit (CPU) or a graphics processing unit (GPU). The control unit 21 controls each component unit and executes processing using built-in memories such as a read-only memory (ROM) or a random access memory (RAM), a clock, a counter, etc. The functional units of the control unit 21 may be realized by software, or some or all of them may be realized by hardware such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA).

[0032] The storage unit 22 includes a non-volatile memory such as a hard disk, flash memory, or SSD (Solid State Drive). The storage unit 22 may be an external storage device connected to the estimation device 2. The storage unit 22 stores various computer programs and data referenced by the control unit 21. The storage unit 22 of this embodiment stores a program 2P for causing a computer to execute processing related to moisture content estimation, and measurement information 221 and estimation information 222 used to execute the program 2P. The storage unit 22 may also store recommendation information, which will be described later.

[0033] The measurement information 221 includes measurement data received from the sensor 7. The measurement information 221 contains records in chronological order, linking together, for example, the measurement time, the measurement value, and the identification information of the sensor 7. The measurement information 221 may also store waveform data. Every time the estimation device 2 acquires a new measurement value from the sensor 7, the estimation device 2 stores the new measurement value in the measurement information 221. The measurement information 221 may also be linked to a moisture content estimation result, which will be described later.

[0034] The estimation information 222 is information necessary for estimating the moisture content, and includes, for example, correspondence information 2221 or a learning model 2222. Details of the estimation information 222 will be described later.

[0035] A computer program (program product) including the program 2P may be provided by a non-transitory recording medium 2A on which the computer program is readably recorded. The storage unit 22 stores the computer program read from the recording medium 2A by a reading device (not shown). The recording medium 2A may be, for example, a magnetic disk, an optical disk, or a semiconductor memory. Alternatively, the computer program may be downloaded from an external server connected to a communications network and stored in the storage unit 22. The program 2P may be a single computer program or may be composed of multiple computer programs, and may be executed on a single computer or on multiple computers interconnected by a communications network.

[0036] The communication unit 23 includes a communication device that realizes communication with an external device via a network such as the Internet. The control unit 21 transmits and receives various information to and from the external device via the communication unit 23. The communication unit 23 may be omitted.

[0037] The display unit 24 includes a display device such as a liquid crystal display, an organic EL (Electro Luminescence) display, etc. The display unit 24 displays information to be notified to the user in accordance with instructions from the control unit 21. The display unit 24 may be interpreted as a notification unit and may be a means for notifying the user by other means such as sound or lighting.

[0038] The operation unit 25 is an interface that accepts user operations. The operation unit 25 includes, for example, a keyboard, a mouse, a touch panel device with a built-in display, a speaker, a microphone, etc. The operation unit 25 accepts operation input from the user and sends a control signal according to the operation content to the control unit 21.

[0039] The input / output unit 26 includes an input / output interface for connecting to an external device, and receives input and output of various data related to the execution of processing related to moisture content estimation. The connection between the input / output unit 26 and the external device may be wired or wireless.

[0040] The input / output unit 26 is connected to the sensor 7. The input / output unit 26 receives a signal output from the sensor 7 and sends the received signal to the control unit 21. The input / output unit 26 may include an A / D conversion function and convert an analog signal obtained from the sensor 7 into a digital signal. The input / output unit 26 may also be connected to the control device 6. The control unit 21 outputs various control signals to the control device 6 via the input / output unit 26 and controls the operation of the biomass combustion facility 1.

[0041] The input / output unit 26 is equipped with a communication device that realizes communication via short-range wireless communication such as Bluetooth (registered trademark) or WiFi (registered trademark), or via a network such as the Internet or a LAN (Local Area Network), and may transmit and receive various data via communication between the sensor 7 and external devices such as the control device 6.

[0042] The estimation device 2 may be configured to receive operations via an externally connected computer and output information to be notified to the external computer. In this case, the estimation device 2 may not be equipped with the display unit 24 and the operation unit 25. The estimation device 2 may be installed in a location remote from the biomass combustion facility 1. In this case, the estimation device 2 may receive measurement data, for example, via the communication function of the sensor 7, or may receive measurement data via the control device 6 connected to the biomass combustion facility 1.

[0043] In biomass power plants configured as described above, a mixture of various raw materials of different types and shapes is often used as biomass fuel. The moisture content of the biomass fuel varies depending on the type and shape of the raw materials. The moisture content also varies depending on the storage condition of the raw materials, weather, etc. The moisture content of the biomass fuel affects the heat value of the biomass combustion facility 1 to which the biomass fuel is supplied. To ensure stable combustion and power generation, it is necessary to keep the heat input to the biomass combustion facility 1 constant. To keep the heat input constant, it is important to accurately determine the moisture content of the biomass fuel supplied to the biomass combustion facility 1 and control the supply state of the biomass fuel to maintain a constant moisture content.

[0044] One possible method for measuring the moisture content of biomass fuel is to extract a sample from the biomass fuel supplied to the transport unit 3 or supply unit 4 and measure the moisture content using a heat-drying measuring device, but this measurement requires time and effort. This makes it difficult to take immediate action based on the measurement results. Furthermore, when measuring moisture content using a near-infrared moisture meter, lighting conditions change depending on the time of day and weather, reducing measurement accuracy.

[0045] In this embodiment, the moisture content of the biomass fuel is estimated using measurement data of the sound generated by the collision between the biomass fuel and the supply unit 4, thereby achieving highly accurate estimation.

[0046] 3 is an explanatory diagram for explaining a method for estimating moisture content, which will be used to explain the method for estimating moisture content executed by the estimation device 2 of this embodiment.

[0047] The estimation device 2 acquires time-series measurement values ​​of the sound detected by the sensor 7, and obtains time-series measurement value data. The time-series measurement value data can be represented as waveform data, as described above. The waveform data based on the measurement values ​​may be generated by either the sensor 7 or the estimation device 2. Note that the estimation device 2 may acquire time-series measurement value data by reading data from the measurement information 221.

[0048] 3A is a diagram showing an example of waveform data. As shown in FIG. 3A, the waveform data is represented by a graph with the vertical axis representing the measured value of the sound and the horizontal axis representing time.

[0049] It is believed that the raw waveform data based on the measurements of the sensor 7 contains components other than sound waves resulting from the collision between the biomass fuel and the chute 42. Therefore, it is preferable to remove such other components from the raw waveform data. Examples of such other components include sound waves resulting from the operation of various components (e.g., the chain conveyor 414, the rotor 415, etc.) that make up the biomass combustion facility 1, and sound waves resulting from environmental noise.

[0050] Considering the content of the various sounds generated by the chute 42, sound waves resulting from the collision between the biomass fuel and the chute 42 can be classified as non-harmonic sounds (percussion sounds), and other components can be classified as harmonic sounds (harmonic sounds). Therefore, by extracting only the percussion sound components from the raw waveform data, waveform data can be obtained that shows the sound wave components resulting from the collision between the biomass fuel and the chute 42, with the other components removed.

[0051] The estimation device 2 analyzes the acquired raw waveform data to separate the sound into harmonic sounds and percussive sounds, and extracts only the percussive sounds. While there are no limitations on the method for extracting the percussive sounds, for example, harmonic / percussive source separation (HPSS) is used. HPSS is a method for separating harmonic sounds and percussive sounds from a source signal. Using HPSS, waveform data showing only the percussive sound components can be obtained.

[0052] After performing the HPSS process, the estimation device 2 divides the waveform data containing only percussion sound components into individual sound segments to obtain waveform data for each sound. FIG. 3B shows an example of waveform data for one sound. While the method for obtaining waveform data for one sound is not limited, an onset detection technique may be used, for example. Specifically, the root mean square (rmS) of the sound data for each frame is calculated for a predetermined time span including the most recent measurement point, and the point at which the calculated rmS is equal to or greater than a predetermined threshold is detected as the onset, which is the start point of the sound. In FIG. 3A, the start point of the sound is indicated by a dashed line.

[0053] Waveform data for one sound is obtained by extracting the signal from the detected first start point to the second start point, which is the start point following the first start point. By performing the above-described process for each start point, waveform data for each sound included in the waveform data is obtained. Note that the waveform data for one sound may be obtained by removing, from the section from the first start point to the second start point, any section in which the sound measurement value is below a predetermined sound threshold for a predetermined period of time or more.

[0054] The estimation device 2 performs frequency analysis on waveform data for one sound to acquire signal strength for each frequency component in the waveform data. The estimation device 2 may acquire a frequency spectrum indicating signal strength for each frequency component. In this embodiment, as an example, the waveform data is subjected to a short-time Fourier transform (STFT) to acquire a spectrogram as a frequency spectrum.

[0055] Fig. 3C is a diagram showing an example of a spectrogram for one sound. A spectrogram is three-dimensional data showing time, frequency, and signal strength (amplitude). In the spectrogram shown in Fig. 3C, the horizontal axis is time (S), the vertical axis is frequency (Hz), and signal strength (dB) is represented by color. The spectrogram may be a Mel frequency spectrogram, in which the frequency on the vertical axis is on the Mel scale.

[0056] The frequency analysis method is not limited to STFT, and other analysis methods such as Fast Fourier Transform (FFT) may also be used. The frequency spectrum is not limited to a spectrogram, and may be two-dimensional data with the first axis representing frequency and the second axis representing signal intensity.

[0057] The estimation device 2 determines the signal strength of a predetermined frequency component from among all frequency components contained in the waveform data based on the results of the frequency analysis. The frequency value extracted from the waveform data may be set appropriately depending on the type of biomass fuel and the supply unit 4. The number of predetermined frequency components extracted from the waveform data may be one or more. The predetermined frequency component may be all frequency components contained in the waveform data.

[0058] The signal strength is preferably calculated by frequency analysis of the waveform data for each sound, but may also be calculated by frequency analysis of the entire waveform data over a predetermined time width.

[0059] The estimation device 2 estimates the moisture content of the biomass fuel corresponding to the signal strength of the determined predetermined frequency component. The estimation device 2 determines the moisture content corresponding to the signal strength of the predetermined frequency component based on, for example, correspondence information 2221 stored in advance.

[0060] FIG. 4 is a diagram showing an example of the contents of the correspondence information 2221. The correspondence information 2221 stores records linking, for example, frequency, signal strength, and moisture content as an example of moisture content. The moisture content may be expressed as a percentage (%). FIG. 4 shows an example of the correspondence information 2221 in a table format in which the moisture content is associated with the signal strength for each frequency, but the correspondence information 2221 may also be stored as a graph, a function formula, or the like. Note that the moisture content of biomass fuel is not limited to being represented by moisture content. The moisture content may be, for example, a moisture content level (e.g., level 1 to level 10) classified into multiple stages according to the moisture content, classification information (e.g., high, normal, low), or the like. The estimation device 2 acquires the above-mentioned predetermined correspondence information 2221 and stores it in the storage unit 22.

[0061] The estimation device 2 identifies the moisture content corresponding to the signal strength of the determined predetermined frequency component by referring to the correspondence information 2221. The estimation device 2 may calculate the signal strength using an interpolation method such as an interpolation method based on the correspondence information 2221. If multiple frequency components are extracted as the predetermined frequency component, the moisture content corresponding to the signal strength of each frequency component is determined.

[0062] The estimation of moisture content is not limited to using the signal intensity values ​​of frequency components, but may also be performed using a frequency spectrum indicating the signal intensity of multiple frequency components. The frequency spectrum may be a spectrum for all frequencies included in the waveform data, or a spectrum for a partial frequency band of all frequencies. In this case, information associating frequency bands, reference frequency spectra, and moisture content may be stored as the correspondence information 2221. The frequency spectrum is, for example, a spectrogram or a two-dimensional graph showing signal intensity versus frequency.

[0063] The estimation device 2 refers to the correspondence information 2221 to identify a reference frequency spectrum corresponding to the acquired frequency spectrum and to identify the moisture content corresponding to the identified reference frequency spectrum. The reference frequency spectrum corresponding to the frequency spectrum is identified by, for example, selecting the most similar reference frequency spectrum based on the similarity between the acquired frequency spectrum and each reference frequency spectrum. The similarity may be image similarity between spectrogram images or shape similarity between two-dimensional graphs.

[0064] The moisture content is calculated based on the waveform data through the above process. When estimating moisture content based on waveform data for a predetermined time width, if the waveform data contains multiple sounds, it is advisable to estimate the moisture content for each sound.

[0065] When multiple moisture content estimates for multiple sounds are obtained, the final moisture content may be determined by calculating a statistical value (e.g., average, median, etc.) of the moisture content for each sound. Similarly, when multiple moisture content estimates for multiple predetermined frequency components are obtained, the final moisture content may be determined by calculating a statistical value of the moisture content for each predetermined frequency component.

[0066] In this embodiment, the estimation device 2 further determines whether the estimated moisture content value is within a predetermined allowable range based on the moisture content estimation result. The allowable moisture content range is defined, for example, by an upper limit, a lower limit, or a combination of these, of the moisture content. If the estimated moisture content value is within the allowable range, it is presumed that the heat input is being maintained appropriately and combustion in the biomass combustion facility 1 will be stable. If the estimated moisture content value is outside the allowable range, it is presumed that the heat input is not being maintained appropriately and combustion will be unstable. Note that the allowable moisture content range used for the determination may be configured to be changeable at any time by an operator, etc.

[0067] If the estimated moisture content is outside the allowable range, the estimation device 2 outputs alert information to notify the operator that the moisture content is outside the allowable range. The alert information includes, for example, the estimated moisture content and a sentence indicating an abnormality in the moisture content, and is displayed on the display unit 24. The alert information may be output as an audio output via a speaker, by lighting a lamp, or the like.

[0068] The estimation device 2 may present the moisture content estimation result regardless of whether the estimated value of the moisture content is outside the allowable range. The moisture content estimation result may be displayed in the form of a trend graph, for example.

[0069] If the estimated moisture content is outside the allowable range, the estimation device 2 may derive suggested information regarding the operation of the biomass combustion facility 1. The suggested information includes, for example, recommended operating conditions for the biomass combustion facility 1. The recommended operating conditions are recommended operating conditions for restoring the moisture content to within the allowable range. The recommended operating conditions may also include a mixing method and mixing conditions when using multiple types (multiple lots) of biomass fuel. Depending on the moisture content, the recommended operating conditions are set to adjust the mixing ratio of low-moisture content biomass fuel and high-moisture content biomass fuel so as to increase or decrease the overall moisture content of the biomass fuel input to the biomass combustion facility 1.

[0070] The recommended operating conditions may include information for controlling the amount of biomass fuel supplied to at least one of the conveying section 3, the supply section 4, and the combustion chamber 5, and the amount of air in the combustion chamber 5. The recommended operating conditions are set so that the amount of supply is adjusted depending on the amount of moisture. The recommended operating conditions are set so that the amount of air is adjusted depending on the amount of moisture. The amount of supply can be adjusted by varying the rotation speed of the chain conveyor 414.

[0071] The estimation device 2 stores in advance in the storage unit 22 recommendation information that associates, for example, the deviation of the estimated moisture content from the allowable moisture content range with recommended operating conditions corresponding to the deviation. The deviation of the moisture content may be the difference between the upper or lower limit of the allowable moisture content range and the estimated moisture content. The recommended operating conditions may be recommended operating conditions or variations in the operating conditions. The estimation device 2 calculates the deviation of the moisture content based on the estimated moisture content value, and derives recommended operating conditions corresponding to the calculated deviation by referring to the recommendation information. The estimation device 2 displays the derived recommended operating conditions on the display unit 24.

[0072] Instead of or in addition to outputting the recommended operating conditions, the estimation device 2 may automatically transmit control information corresponding to the derived recommended operating conditions to the control device 6. The control information includes, for example, a recommended value or a fluctuation value for the rotation speed of the chain conveyor 414. The control device 6 controls the operation of the biomass combustion facility 1 in accordance with the control information received from the estimation device 2. The estimation device 2 may be configured to receive an operator's operation indicating whether or not to output control information, and to switch between transmitting and not transmitting the control information depending on the received indication of whether or not it is necessary.

[0073] 5 is a flowchart showing an example of a processing procedure executed by the estimation device 2. The processing in each of the following flowcharts is executed by the control unit 21 in accordance with a program 2P stored in the storage unit 22 of the estimation device 2.

[0074] The control unit 21 of the estimation device 2, functioning as an acquisition unit, acquires waveform data indicating the sound generated by the supply unit 4 (step S11).

[0075] The control unit 21, functioning as an extraction unit, separates the acquired waveform data into percussion sounds and harmonic sounds using the HPSS technique, and extracts only the percussion sound components (step S12). The control unit 21 detects onsets, which are the starting points of sounds, based on the waveform data containing only the extracted percussion sound components (step S13). The control unit 21 divides the waveform data into individual sounds based on the onset detection results, and thereby acquires waveform data for each sound included in the waveform data (step S14).

[0076] The control unit 21 performs frequency analysis on the waveform data for one sound to obtain the signal intensity for each frequency component, thereby obtaining the signal intensity of one or more predetermined frequency components among all the frequency components included in the waveform data for one sound (step S15). In step S15, the control unit 21 may obtain a frequency spectrum indicating the signal intensity of the plurality of predetermined frequency components, or may obtain a spectrogram by performing a short-time Fourier transform on the waveform data.

[0077] The control unit 21, functioning as an estimation unit, estimates the moisture content of the biomass fuel corresponding to the signal intensity or frequency spectrum of the predetermined frequency component based on the acquired signal intensity value or frequency spectrum of one or more predetermined frequency components (step S16). In step S16, the control unit 21 determines the moisture content corresponding to the signal intensity value or frequency spectrum of the predetermined frequency component based on the information stored in the correspondence information 2221. The control unit 21 estimates the moisture content by performing the above-mentioned process for each piece of waveform data for one sound included in the predetermined time width. The control unit 21 may determine the final moisture content based on multiple moisture content estimates.

[0078] Based on the moisture content estimation result, the control unit 21 determines whether the estimated value of the moisture content is outside a preset allowable range (step S17). If it is determined that the estimated value of the moisture content is within the allowable range (S17: NO), the control unit 21 proceeds to step S22.

[0079] If it is determined that the estimated moisture content is outside the allowable range (S17: YES), the control unit 21 generates alert information to notify that the moisture content is outside the allowable range (step S18).The control unit 21 displays the generated alert information on the display unit 24 (step S19).

[0080] The control unit 21 derives proposal information including recommended operating conditions for the biomass combustion facility 1 based on the moisture content estimation result (step S20). The control unit 21 displays the derived proposal information on the display unit 24 (step S21). Note that the control unit 21 may display the alert information and the proposal information together. The estimation device 2 may transmit control information according to the derived recommended operating conditions to the control device 6.

[0081] The control unit 21 determines whether or not to end the process (step S22). If it is determined not to end the process (S22: NO), the control unit 21 returns the process to step S11. It is preferable that the control unit 21 does not end the process while the biomass power generation plant is in operation, and repeatedly executes the above-described series of processes. If it is determined to end the process (S22: YES), the control unit 21 ends the process.

[0082] In the above, sound or vibration generated in the supply unit 4 of the supply device 10 is detected, but the detection position of the sound or vibration is not limited to the supply unit 4, and may be the conveying unit 3. For example, in a case where the conveying unit 3 includes a plurality of flight conveyors and a chute installed at a transfer point between the conveyors, sound or vibration generated in the chute in the conveying unit 3 may be detected.

[0083] According to this embodiment, the moisture content can be estimated easily and accurately using waveform data of the sound or vibration generated by the supply device. By frequency analyzing the waveform data, the moisture content can be estimated accurately by utilizing spectral changes corresponding to the moisture content. Since the moisture content can be continuously estimated in real time while the biomass power plant is operating based on the sound or vibration measured at predetermined intervals, immediate action can be taken based on the estimation results, leading to stable operation of the biomass power plant. Since sound or vibration can be measured using easily available sensors, the system can be easily introduced. Since lighting conditions do not affect the measurement results, accuracy is improved.

[0084] By separating the percussion sounds from the raw waveform data, the collision state between the biomass fuel and the chute can be properly reflected in the waveform data, improving the accuracy of estimating the moisture content. Estimating the moisture content based on the waveform data for each sound enables analysis that is in line with the sound generation situation, improving estimation accuracy.

[0085] By preparing the correspondence information in advance, the estimation process can be performed easily and accurately. By estimating the moisture content based on the frequency spectrum, the moisture content can be estimated taking into account multiple frequency components contained in the waveform data.

[0086] If the estimated moisture content is outside a predetermined range, alert information is output, thereby reliably informing the user of an abnormal moisture content. Furthermore, outputting proposal information according to the estimated moisture content makes it easier to take measures to stabilize operation, and enables the estimation results to be appropriately fed back to the operating conditions, thereby supporting more advanced and efficient plant operation.

[0087] By measuring the sound or vibration at or around the falling position of the biomass fuel in the chute, it is possible to grasp the state of the biomass fuel before it is fed into the combustion chamber.

[0088] (Second embodiment) In the second embodiment, a configuration for estimating moisture content using a learning model will be described. In the following embodiment, differences from the first embodiment will be mainly described, and components common to the first embodiment will be assigned the same reference numerals and detailed descriptions thereof will be omitted.

[0089] The estimation device 2 of the second embodiment stores a learning model 2222 as estimation information 222 in the storage unit 22. The learning model 2222 is a machine learning model that has learned predetermined training data. The learning model 2222 is expected to be used as a program module that constitutes part of artificial intelligence software.

[0090] Fig. 6 is an explanatory diagram showing an overview of the learning model 2222 in the second embodiment. The learning model 2222 receives the frequency spectrum of waveform data for one sound as input and outputs information indicating the moisture content of the waveform data. Fig. 6 shows an example in which the input data to the learning model 2222 is a spectrogram, but the input data may also be a two-dimensional graph showing signal intensity versus frequency.

[0091] The learning model 2222 is, for example, a convolutional neural network (CNN), which is a type of neural network. The learning model 2222 includes an input layer to which a frequency spectrum is input, an output layer that outputs the moisture content, and an intermediate layer (hidden layer). The intermediate layer may include a convolutional layer, a pooling layer, a fully connected layer, etc. The intermediate layer has multiple nodes that extract features of the frequency spectrum, and passes the features extracted using various parameters to the output layer. When a frequency spectrum is input to the input layer, a calculation is performed in the intermediate layer using the learned parameters, and output information indicating the moisture content value is output from the output layer.

[0092] The learning model 2222 can be generated by preparing training data in which labels indicating moisture content are associated with frequency spectra, and using the training data to train an untrained neural network.

[0093] The estimation device 2 inputs multiple frequency spectra contained in the training data into the input layer of the pre-training neural network model, undergoes calculation processing in the intermediate layer, and obtains the moisture content output from the output layer. The estimation device 2 compares the moisture content output from the output layer with the moisture content contained in the training data, and optimizes parameters such as the weights between neurons using, for example, backpropagation so that the moisture content output from the output layer approaches the correct value. Note that the learning model 2222 may be constructed by an external device and deployed to the estimation device 2.

[0094] The learning model 2222 may be configured to input the signal intensity values ​​of predetermined frequency components instead of the frequency spectrum and output the moisture content. If multiple predetermined frequency components are extracted as the predetermined frequency components, the signal intensities of the multiple predetermined frequency components and information indicating the frequencies corresponding to each signal intensity may be provided to the learning model 2222.

[0095] The configuration of the learning model 2222 is not limited to the above example, and may be any model that can identify the water content from the frequency spectrum or signal intensity. The learning model 2222 may be a model constructed using other learning algorithms, such as a recurrent neural network (RNN), a graph neural network (GNN), a transformer, a support vector machine (SVM), logistic regression, or eXtreme Gradient Boosting (XGBoost).

[0096] The estimation device 2 estimates the moisture content using the above-described learning model 2222. In step S16 of the flowchart in Fig. 5 , the estimation device 2 inputs the acquired frequency spectrum or the signal intensity of a predetermined frequency component into the learning model 2222 and acquires the moisture content output from the learning model 2222, thereby estimating the moisture content.

[0097] According to this embodiment, it is possible to easily and accurately estimate moisture content using a learning model. By using a frequency spectrum as an input to the learning model, moisture content can be estimated based on various feature quantities obtained from the spectrum.

[0098] (Third embodiment) In the third embodiment, a configuration will be described in which the occurrence of a bridge in the supply unit 4 or an unsuitable fuel in the biomass fuel is detected based on measurement data measured by the sensor 7.

[0099] The estimation device 2 of the third embodiment detects the occurrence of a bridging phenomenon, in which biomass fuel charged into the hopper 41 clogs the hopper 41, based on sound measurement data measured by the sensor 7. The occurrence of the bridging phenomenon hinders the supply of biomass into the combustion chamber 5.

[0100] When a bridging phenomenon occurs, the amount of biomass fuel dropped and supplied to the chute 42 decreases, resulting in fewer sounds than usual being generated in the chute 42. In this embodiment, bridging is detected by calculating the number of sounds generated based on waveform data over a predetermined period of time.

[0101] Furthermore, the sound generated by the collision between the biomass fuel and the supply unit 4 also changes depending on whether or not there are unsuitable fuels in the biomass fuel. When unsuitable fuels are present, the specific frequency components caused by the unsuitable fuels become stronger. In this embodiment, the unsuitable fuels are detected by extracting the specific frequency components caused by the unsuitable fuels from the waveform data.

[0102] FIG. 7 is a flowchart showing an example of a processing procedure executed by the estimation device 2 of the third embodiment.

[0103] The control unit 21 of the estimation device 2 executes a process for detecting a bridge in the hopper 41 based on waveform data indicating the sound generated in the supply unit 4 (step S31).

[0104] The control unit 21 calculates the number of wave occurrences within a predetermined time period by counting the number of sound start points within a predetermined time period based on the detection results of the sound start points described in the first embodiment. The control unit 21 detects the occurrence of a bridge by determining whether the calculated number of sound occurrences is equal to or greater than a predetermined number threshold. If the calculated number of sound occurrences is less than the predetermined number threshold, it is determined that a bridge has occurred, and a bridge is detected. If the calculated number of sound occurrences is equal to or greater than the predetermined number threshold, it is determined that a bridge has not occurred, and a bridge is not detected. Note that the bridge detection method is not limited to the above example, and may be any method that utilizes sound measurement data.

[0105] The control unit 21 executes a process for detecting unsuitable fuel in the biomass fuel based on the waveform data indicating the sound generated in the supply unit 4 (step S32).

[0106] The control unit 21 stores in advance in the storage unit 22 unsuitable substance information, which associates, for example, the type of unsuitable fuel substance with the specific frequency component of the sound generated due to the unsuitable fuel substance. The control unit 21 calculates the signal intensity of the specific frequency component in the waveform data of the acquired sound based on the frequency analysis results described in the first embodiment. The control unit 21 detects the unsuitable fuel substance by determining whether the signal intensity of the calculated specific frequency component is equal to or greater than a predetermined intensity threshold. If the signal intensity of the calculated specific frequency component is equal to or greater than the predetermined intensity threshold, it is determined that the unsuitable fuel substance is included, and the unsuitable fuel substance corresponding to the specific frequency component is detected. If the signal intensity of the calculated specific frequency component is less than the predetermined intensity threshold, it is determined that the unsuitable fuel substance is not included, and the unsuitable fuel substance is not detected.

[0107] The control unit 21 may detect unsuitable fuel substances using a machine learning technique. For example, a non-suitable fuel detection model that has been trained to output the presence or absence of an unsuitable fuel substance when a frequency spectrum is input is prepared in advance. The control unit 21 may detect an unsuitable fuel substance by inputting a frequency spectrum based on the acquired waveform data into the non-suitable fuel detection model and obtaining the presence or absence of an unsuitable fuel substance output from the non-suitable fuel detection model. The non-suitable fuel detection model may be configured to output the type of the unsuitable fuel substance along with the presence or absence of the unsuitable fuel substance. Note that the detection method for unsuitable fuel substances is not limited to the above example, and may be any method that utilizes sound measurement data.

[0108] The control unit 21 generates a screen showing the detection results of the bridge and the detection results of the unsuitable fuel, displays the generated screen showing the detection results on the display unit 24 (step S33), and ends the series of processes. The control unit 21 may display the detection results only when either the bridge or the unsuitable fuel is detected.

[0109] In the above-described processing, the estimation device 2 may execute only one of the bridge detection processing and the fuel unsuitable substance detection processing.

[0110] According to this embodiment, more diverse information can be presented based on the measurement data obtained by the sensors, which increases the utilization of the measurement data and improves the convenience of the system. By understanding the state of biomass fuel more accurately, it leads to further stabilization of plant operation.

[0111] In the second and third embodiments described above, examples have been described in which sound measurement data is used, but it goes without saying that vibration measurement data may also be used.

[0112] The embodiments disclosed herein should be considered to be illustrative in all respects and not restrictive. The technical features described in each embodiment can be combined with each other, and the scope of the present invention is intended to include all modifications within the scope of the claims and the scope equivalent to the claims. The sequences shown in each embodiment are not limited, and the order of each process may be changed within a range consistent with the present invention, and multiple processes may be executed in parallel. The entity that performs each process is not limited, and the process of each device may be executed by another device within a range consistent with the present invention.

[0113] The matters described in each embodiment can be combined with each other. In addition, the independent claims and dependent claims described in the claims can be combined with each other in any combination, regardless of the reference format. Furthermore, although the claims use a format in which a claim references two or more other claims (multiple claim format), this is not limited to this format. A multiple claim (multi-multi claim) that references at least one other multiple claim may also be used. [Explanation of symbols]

[0114] 100 Estimation System (Information Processing System) 2. Estimation device (information processing device) 21 Control section 22 Memory section 23 Communications Department 24 Display 25 Control section 26 Input / output section 2A Recording Media 2P Program 222 Estimated Information 2221 Compatibility Information 2222 Learning Model 1. Biomass combustion facility 10 Feeding device 3. Conveyor 4 Supply section 41 Hopper 42 shots 5 Combustion chamber 6. Control device

Claims

1. In a biomass combustion facility in which biomass fuel is supplied to a combustion chamber via a supply device, time-series measurement data of sound or vibration generated by the supply device is acquired, The moisture content of the biomass fuel is estimated based on the intensity of one or more frequency components in the acquired time-series measurement data. A computer program that causes a computer to perform a process.

2. acquiring waveform data as the time-series measurement data; A frequency analysis is performed on the acquired waveform data to determine the intensity of the one or more frequency components.

2. The computer program of claim 1.

3. extracting percussion sound components related to the one or more frequency components; The moisture content is estimated based on the intensity of the extracted percussion sound component.

3. A computer program according to claim 1 or claim 2.

4. The moisture content is estimated based on the intensity of the one or more frequency components for one sound in the time-series measurement data.

3. A computer program according to claim 1 or claim 2.

5. The moisture content corresponding to the acquired time-series measurement data is estimated based on correspondence information between the intensities of the one or more predetermined frequency components and the moisture content.

3. A computer program according to claim 1 or claim 2.

6. A learning model is used to output the moisture content of biomass fuel corresponding to the time-series measurement data when the intensity of one or more frequency components in the time-series measurement data is input, and the moisture content corresponding to the acquired time-series measurement data is estimated.

3. A computer program according to claim 1 or claim 2.

7. The moisture content of the biomass fuel is estimated based on a frequency spectrum indicating the intensities of a plurality of frequency components in the time-series measurement data.

3. A computer program according to claim 1 or claim 2.

8. If the estimated moisture content is outside a predetermined range, output information indicating that the moisture content is outside the predetermined range.

3. A computer program according to claim 1 or claim 2.

9. generating proposal information regarding operation of the biomass combustion facility based on the estimated moisture content; Output the generated proposal information 3. A computer program according to claim 1 or claim 2.

10. Based on the measurement data, occurrence of a bridge in the supply device or unsuitable fuel in the biomass fuel is detected.

3. A computer program according to claim 1 or claim 2.

11. The supply device includes a hopper and a chute through which the biomass fuel is dropped and supplied from the hopper. Acquire measurement data of sound or vibration generated at or around the falling position of the biomass fuel in the chute.

3. A computer program according to claim 1 or claim 2.

12. In a biomass combustion facility in which biomass fuel is supplied to a combustion chamber via a supply device, time-series measurement data of sound or vibration generated by the supply device is acquired, The moisture content of the biomass fuel is estimated based on the intensity of one or more frequency components in the acquired time-series measurement data. An information processing method in which processing is performed by a computer.

13. an acquisition unit that acquires time-series measurement data of sound or vibration generated by a supply device in a biomass combustion facility in which biomass fuel is supplied to a combustion chamber via the supply device; an estimation unit that estimates the moisture content of the biomass fuel based on the intensity of one or more frequency components in the acquired time-series measurement data. Information processing device.

14. The biomass combustion equipment includes a sensor and an information processing device, the sensor being provided in the biomass combustion equipment and supplying biomass fuel to a combustion chamber via a supply device; The information processing device includes: an acquisition unit that acquires time-series measurement data of the sound or vibration generated by the supply device measured by the sensor; an estimation unit that estimates the moisture content of the biomass fuel based on the intensity of one or more frequency components in the acquired time-series measurement data. Information processing system.

Images