WO2019235604A1 - Dispositif, système et procédé de commande, et programme - Google Patents
Dispositif, système et procédé de commande, et programme Download PDFInfo
- Publication number
- WO2019235604A1 WO2019235604A1 PCT/JP2019/022671 JP2019022671W WO2019235604A1 WO 2019235604 A1 WO2019235604 A1 WO 2019235604A1 JP 2019022671 W JP2019022671 W JP 2019022671W WO 2019235604 A1 WO2019235604 A1 WO 2019235604A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- property
- time series
- control device
- electric motor
- unit
- Prior art date
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B02—CRUSHING, PULVERISING, OR DISINTEGRATING; PREPARATORY TREATMENT OF GRAIN FOR MILLING
- B02C—CRUSHING, PULVERISING, OR DISINTEGRATING IN GENERAL; MILLING GRAIN
- B02C15/00—Disintegrating by milling members in the form of rollers or balls co-operating with rings or discs
- B02C15/04—Mills with pressed pendularly-mounted rollers, e.g. spring pressed
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B02—CRUSHING, PULVERISING, OR DISINTEGRATING; PREPARATORY TREATMENT OF GRAIN FOR MILLING
- B02C—CRUSHING, PULVERISING, OR DISINTEGRATING IN GENERAL; MILLING GRAIN
- B02C25/00—Control arrangements specially adapted for crushing or disintegrating
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B02—CRUSHING, PULVERISING, OR DISINTEGRATING; PREPARATORY TREATMENT OF GRAIN FOR MILLING
- B02C—CRUSHING, PULVERISING, OR DISINTEGRATING IN GENERAL; MILLING GRAIN
- B02C4/00—Crushing or disintegrating by roller mills
- B02C4/02—Crushing or disintegrating by roller mills with two or more rollers
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
- H02P23/00—Arrangements or methods for the control of AC motors characterised by a control method other than vector control
- H02P23/14—Estimation or adaptation of motor parameters, e.g. rotor time constant, flux, speed, current or voltage
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
- H02P29/00—Arrangements for regulating or controlling electric motors, appropriate for both AC and DC motors
Definitions
- the present invention relates to a control device, a control system, a control method, and a program for a pulverizer that pulverizes an object to be pulverized.
- a coal fired thermal power plant is provided with a mill (pulverizer) that pulverizes coal to produce pulverized coal.
- the mill includes a rotary separator that is separated into fine and coarse coal particles by rotation.
- the fineness of coal supplied from the mill is controlled by the rotational speed of the rotary separator. That is, the fineness of the supplied coal increases as the rotational speed of the rotary separator increases.
- the rotational speed for obtaining the same fineness varies depending on the amount of coal to be input and the coal type of coal.
- Patent Document 1 discloses a technique for controlling the rotational speed of a rotary separator according to the amount of coal input and the type of coal.
- a control apparatus pinpoints the coal type of coal by inputting the coal type of coal which an operator inputs.
- the coal type of coal is specified in advance, it is difficult to accurately grasp what type of coal it is unless it is actually burned.
- the coals are related to the same coal type, there is some variation in their properties (hardness, etc.), so even if the same control is performed on coal of the same coal type, the fineness varies. It can happen.
- the objective of this invention is providing the control apparatus, control system, control method, and program which control a grinder automatically appropriately according to the property of coal.
- the control device is a control device for a pulverizer that pulverizes an object to be pulverized, and includes at least a measured value of a physical quantity related to electricity input to an electric motor of the pulverizer.
- a measurement value acquisition unit that acquires at a frequency equal to or higher than the frequency of the power supply current of the electric motor, and a control unit that outputs a control signal to the pulverizer based on the time series of the acquired measurement values.
- the control device stores a property storage unit that stores a time series of physical quantities related to electricity input to the electric motor according to the property of the object to be crushed. Based on the time series stored in the property storage unit and the time series of the acquired measurement values, the device further includes a property specifying unit that specifies the property of the object to be crushed, and the control unit is specified A control signal may be output to the pulverizer based on the property.
- control device further includes a frequency analysis unit that specifies a frequency spectrum by time-series frequency analysis of the measurement value, and the control unit includes the measurement
- a control signal may be output to the pulverizer based on a time-series frequency spectrum of values.
- the control device stores a property storage unit that stores a frequency spectrum of a physical quantity related to electricity input to the electric motor according to the property of the object to be crushed, Based on the frequency spectrum stored in the property storage unit and the frequency spectrum obtained by frequency analysis, the property storage unit further includes a property specifying unit that specifies the property of the object to be crushed, and the control unit specifies the specified A control signal may be output to the pulverizer based on properties.
- control device changes the property of the object to be crushed based on the time-series change of the measurement value over time. It may further include a property change determination unit for determining the presence or absence of the.
- the control device provides a time series of physical quantities related to electricity input to the electric motor and the electric power input to the electric motor according to the properties of the object to be crushed.
- a property storage unit that stores at least one of frequency spectra of physical quantities according to the above, a comparison between the frequency spectrum stored in the property storage unit and the frequency spectrum obtained by frequency analysis, or a time series stored in the property storage unit And a comparison with the time series of the acquired measurement values, a property specifying unit for specifying the properties of the object to be crushed based on any one of the above, the frequency spectrum obtained by frequency analysis, and the acquired
- a property change that determines the presence or absence of a property change of the object to be crushed based on a change over time of a time series of measurement values that are not used for specifying the property May be one further comprising a tough.
- the property specifying unit includes a time series of the latest measurement values and a time series of the past measurement values.
- the property may be specified based on accumulation of deviations.
- the property storage in which the control device according to the fifth or sixth aspect stores the frequency spectrum of the physical quantity related to electricity input to the electric motor according to the property of the object to be crushed.
- the property change determination unit is configured to perform a pattern matching between the frequency spectrum obtained by frequency analysis and each of the frequency spectra classified by properties stored in the property storage unit, to change the property change of the object to be crushed. The presence or absence may be determined.
- the control device includes a correction parameter storage unit that stores a parameter related to correction of a control signal for each property of the pulverized object.
- a correction amount specifying unit that specifies a correction amount of a control signal based on the specified property and information stored in the correction parameter storage unit, and the control unit includes the specified correction amount
- the control signal corrected based on the above may be output to the pulverizer.
- the property storage unit includes information related to a time series of physical quantities related to electricity input to the electric motor, and the pulverization.
- the combination with the properties of the target object is used as teacher data, the information related to the time series is input, the learning model learned to be the output of the properties of the pulverized target object is stored, and the property specifying unit is The property of the object to be crushed may be specified by a learning model.
- the property storage unit includes information relating to a time series of physical quantities related to electricity input to the electric motor, and information on the grinding object.
- a learning model that is trained so that a combination with the presence or absence of property change is used as teacher data, information related to the time series is input, and whether or not the property change of the pulverized object is output is stored, and the property A change determination part may determine the presence or absence of the change of the property of the said grinding
- the measurement value acquisition unit is input to the first electric motor among the plurality of electric motors included in the pulverizer.
- the measured value of the physical quantity related to electricity is acquired, and the control unit is provided with a second electric motor provided on the downstream side in the flow direction of the object to be crushed from the first electric motor among the plurality of electric motors.
- the control signal may be output.
- the first electric motor is an electric motor that rotates a table of the crusher
- the second electric motor is the crusher. It may be an electric motor that rotates a rotary separator of the machine.
- the pulverizer is a fine pulverizer
- the electric motor is an induction motor
- the object to be ground may be at least one of coal and biomass.
- the control system is a control device for a pulverizer that pulverizes an object to be pulverized, wherein a measured value of an input current to the electric motor of the pulverizer is at least a power supply current of the electric motor.
- a measurement value acquisition unit that acquires at a sampling period corresponding to the frequency of the control unit, and a control unit that outputs a control signal to the pulverizer, and a remote control unit provided with the measurement value acquisition unit.
- an arithmetic unit that performs a calculation related to the control signal based on the time series of the measured values.
- the control device is one of a plurality of control devices that control a plurality of pulverizers, and the arithmetic device comprises the The calculation related to the control signal in each control device may be performed based on each time series of the measurement values acquired by a plurality of control devices.
- the control method is a control method for a pulverizer that pulverizes an object to be pulverized, wherein at least a measured value of a physical quantity related to electricity input to an electric motor of the pulverizer Acquiring at a sampling period corresponding to the frequency of the power supply current of the electric motor, and outputting a control signal to the pulverizer based on the acquired time series of the measured values.
- the program obtains a measurement value of a physical quantity related to electricity input to the electric motor of the pulverizer at least at a sampling period corresponding to the frequency of the power supply current of the electric motor. And a step of outputting a control signal to the pulverizer based on the acquired time series of the measured values.
- control device can automatically and appropriately control the pulverizer according to the properties of the coal.
- FIG. 1 is a diagram illustrating a configuration of a roller mill according to the first embodiment.
- the roller mill 100 includes a coal supply port 102 that accepts supply of coal from a coal bunker (not shown) and a coal outlet 103 through which pulverized coal is discharged.
- the roller mill 100 includes a table 104 on the bottom of the casing 101 on which coal supplied from the coal supply port 102 is placed.
- the table 104 is rotated by a table motor 105.
- the table motor 105 is provided with an ammeter 111 that measures an input current.
- a roller 106 that presses the table 104 is provided on the table 104.
- the roller 106 is rotated by a roller motor 107. Coal placed on the table 104 is pulverized by being caught between the rotating table 104 and the roller 106.
- the roller mill 100 includes an air inlet 108 through which primary air is blown below the table 104.
- the primary air blown from the air blowing port 108 passes through the gap between the table 104 and the casing 101 and blows up inside the casing 101.
- the pulverized coal is raised by the primary air that blows up inside the casing 101.
- the roller mill 100 includes a rotary separator 109 that classifies coal pulverized by rotation between the table 104 and the coal outlet 103.
- the rotary separator 109 is rotated by a separator motor 110.
- the coal blown up by the primary air is guided to the rotary separator 109, and is classified into relatively coarse particles (coarse particles) and relatively fine particles (fine particles) by the rotation of the rotary separator 109.
- Coarse particles fall on the table 104 and are pulverized again by the rollers 106.
- the fine particles pass through the rotary separator 109 and are discharged from a discharge port 103 to a combustion device (not shown).
- the roller mill 100 is an example of a fine pulverizer.
- the rotation of the separator motor 110 is controlled by the control device 200.
- the control device 200 outputs a control signal to the separator motor 110 based on the input current of the table motor 105 measured by the ammeter 111.
- the input current of the table motor 105 is an example of a physical quantity related to electricity input to the electric motor.
- the input current sampling frequency by the ammeter 111 is preferably at least equal to or higher than the input current frequency.
- the sampling frequency is more preferably twice or more the frequency of the input current. This is because, as a result of the experiment, the influence of vibration due to coal pulverization appears on the input current of the roller mill 100, the influence appears mainly on the power supply frequency of the input current, and the waveform of the input current is caused by this influence.
- Separator motor 110 is realized by, for example, an induction motor.
- FIG. 2 is a diagram showing an example of the frequency analysis result of the input current of the table motor.
- a solid line and a broken line indicate frequency spectra of input currents when different types of coal are pulverized.
- the frequency spectrum spreads around the frequency of the power supply current depending on the coal type of the coal to be crushed. Therefore, the coal type of the coal being pulverized can be specified by comparing the frequency spectrum of the input current equal to or higher than the power supply current frequency with the standard frequency spectrum of each coal type.
- the coal type of the coal being pulverized can be specified by comparing the input current waveform obtained at the sampling frequency equal to or higher than the power supply current frequency with the standard current waveform of each coal type.
- Coal is an example of the object to be crushed, and the coal type of coal is an example of the properties of the object to be crushed.
- FIG. 3 is a schematic block diagram illustrating the configuration of the control device according to the first embodiment.
- the control device 200 includes a measurement value acquisition unit 201, a measurement value storage unit 202, a frequency analysis unit 203, a property storage unit 204, a property specification unit 205, a property change determination unit 206, a correction value determination unit 207, a torque command acquisition unit 208, A correction unit 209 and a control unit 210 are provided.
- the control device 200 is an example of a control system.
- the measurement value acquisition unit 201 acquires a measurement value of the input current of the table motor 105 from the ammeter 111.
- the measurement value acquisition unit 201 records the acquired measurement values in the measurement value storage unit 202 as a time series.
- the measured value storage unit 202 stores a time series of measured values related to a sampling frequency equal to or higher than the power supply current frequency.
- the measurement value acquisition unit 201 may record the current value of each phase in the measurement value storage unit 202 or record the average value of the current values. May be.
- the frequency analysis unit 203 frequency-converts a partial time series related to the latest analysis target time (for example, 1 second or more) among the time series of measurement values stored in the measurement value storage unit 202 to obtain a frequency spectrum.
- the property storage unit 204 stores a correction function for specifying the frequency spectrum of the input current of the table motor 105 and the torque command correction value when pulverizing the coal related to the coal type in association with the coal type.
- the correction function is, for example, information (such as an equation and a reference table) indicating the relationship between the deviation from the frequency spectrum as a dependent variable and the torque command correction value as a target variable.
- the property specifying unit 205 specifies the coal type of the coal input to the roller mill 100 based on the comparison between the frequency spectrum obtained by the frequency analysis unit 203 and the frequency spectrum related to each coal type stored in the property storage unit 204. To do. For example, the property specifying unit 205 performs pattern matching processing between each frequency spectrum related to a plurality of coal types stored in the property storage unit 204 and the frequency spectrum obtained by the frequency analysis unit 203, and the coal type having the highest similarity. Is identified as the coal type of the coal charged into the roller mill 100.
- the property change determination unit 206 determines whether or not the coal type of the coal input to the roller mill 100 has changed based on the time series of the measurement values stored in the measurement value storage unit 202.
- the property change determination unit 206 includes a first partial time series related to the latest analysis target time and a start point of the first partial time series among the time series of measurement values stored in the measurement value storage unit 202. Based on the comparison with the second partial time series related to the analysis target time, the presence or absence of a change in the coal type is determined.
- the correction value determination unit 207 determines whether or not to change the torque command correction value based on the determination result of the change of the coal type by the property change determination unit 206.
- the correction value determination unit 207 is a torque command correction value stored in the property storage unit 204 in association with the coal type specified by the property specification unit 205 when the property change determination unit 206 determines that the coal type has changed. Then, it is determined to change the torque command correction value.
- the torque command acquisition unit 208 acquires a basic torque command for the separator motor 110.
- the torque command acquisition unit 208 may calculate the torque value of the separator motor 110 based on the amount of coal supply of the roller mill 100, or based on the torque command output by another device that generates the torque command of the roller mill 100. It may be acquired as a torque command.
- the correction unit 209 corrects the torque command by adding the torque command correction value specified by the correction value determination unit 207 to the torque value indicated by the basic torque command acquired by the torque command acquisition unit 208.
- the control unit 210 outputs the torque command corrected by the correction unit 209 to the separator motor 110.
- FIG. 4 is a flowchart showing the operation of the roller mill control device according to the first embodiment.
- the measurement value acquisition unit 201 of the control device 200 acquires the measurement value of the input current of the table motor 105 from the ammeter 111 at each timing related to the sampling period, and associates this with the time to measure the measurement value. Record in the storage unit 202. Thereby, the measured value storage unit 202 stores a time series of measured values of the input current.
- the control device 200 performs correction amount determination processing described below for each analysis target time (for example, 1 second).
- the frequency analysis unit 203 reads a partial time series related to the latest analysis target time from the time series of measurement values stored in the measurement value storage unit 202 (step S1).
- the frequency analysis unit 203 performs frequency conversion on the read partial time series to obtain a frequency spectrum (step S2).
- the property specifying unit 205 performs pattern matching between each frequency spectrum stored in the property storage unit 204 in association with a plurality of coal types and the frequency spectrum obtained in step S2, and the coal type having the highest similarity. Is specified (step S3).
- the property change determination unit 206 includes the first partial time series related to the latest analysis target time and the analysis target before the start point of the first partial time series among the time series of measurement values stored in the measurement value storage unit 202.
- the second partial time series related to time is read (step S4). That is, when the current time is T0 and the analysis target time is t, the first partial time series is a time series of measured values from time T0-t to time T0, and the second partial time series is , A time series of measured values from time T0-2t to time T0-t.
- the first partial time series is equal to the partial time series to be analyzed by the frequency analysis unit 203.
- the property change determination unit 206 calculates the accumulation of deviations at each sampling timing between the first partial time series and the second partial time series (step S5). That is, the property change determination unit 206 calculates the difference area between the waveform specified from the first partial time series and the waveform specified from the second partial time series.
- the property change determination unit 206 determines whether or not the calculated deviation accumulation is greater than or equal to a predetermined threshold (step S6). If the property change determination unit 206 determines that the accumulated deviation is equal to or greater than the threshold (step S6: YES), the property change determination unit 206 determines that the coal type of the input coal has changed. On the other hand, when it is determined that the accumulated deviation is less than the threshold (step S6: NO), the property change determination unit 206 does not determine that the coal type of the input coal has changed.
- the property change The determination unit 206 does not determine that the coal type has changed when the difference between the waveforms of the previous and current measurement values is small.
- the correction value determination unit 207 associates the coal type identified by the property identification unit 205 with the property storage unit 204.
- the stored correction function is read (step S7).
- the correction value determination unit 207 calculates the accumulation of deviations between the frequency spectrum stored in the property storage unit 204 in association with the identified coal type and the frequency spectrum obtained in step S2 (step S8).
- the correction value determination unit 207 obtains a torque command correction value by substituting the calculated accumulated deviation into a correction function (step S9).
- step S6 determines that the accumulated deviation is less than the threshold value
- the correction value determination unit 207 uses the previously determined torque command correction value as the current torque command correction value. (Step S10).
- the control device 200 can determine the torque command correction value for the basic torque command acquired by the torque command acquisition unit 208. Then, every time the torque command acquisition unit 208 acquires the basic torque command, the correction unit 209 corrects the torque command by adding the determined torque command correction value to the basic torque command, and the control unit 210 is corrected.
- the roller mill 100 is controlled by outputting the torque command to the separator motor 110.
- the torque command is an example of a control signal.
- the control device 200 acquires the measurement value of the input current of the table motor 105 of the roller mill 100 at a frequency at least equal to the frequency of the power supply current of the table motor 105, and A control signal is output to the roller mill 100 based on the time series.
- the frequency spectrum of the input current of the table motor 105 spreads around the frequency of the power supply current of the table motor 105 and varies depending on the coal type of coal.
- control device 200 specifies a frequency spectrum by time-series frequency analysis of measured values, and outputs a control signal to the roller mill 100 based on the frequency spectrum.
- the frequency spectrum is not affected by instantaneous changes in the analysis process compared to the time series of measurement values.
- the time series of measurement values is easier to detect an instantaneous change in the analysis process than the frequency spectrum.
- control device 200 determines whether or not there is a change in the coal type of the coal based on a time-dependent change in the measurement value over time.
- the control apparatus 200 can change the coal type even when an accidental change in the input current waveform due to the inclusion of foreign matter such as stone occurs, if the difference from the waveform of the past measurement value is small. It is possible to prevent erroneous determination that the coal type of coal has changed.
- control device 200 identifies the coal type of coal based on the frequency spectrum, and determines whether or not there is a change in the coal type of coal based on the time series of the measured values. That is, the control device 200 uses different ones for the identification of the coal type and the determination of the change of the coal type among the time series of the frequency spectrum and the measurement value. Thus, the control apparatus 200 can estimate the coal type of coal more robustly by using both the frequency spectrum and the time series of the measured values.
- the control device 200 according to the first embodiment identifies a coal type based on comparison of frequency spectra, and determines whether or not there is a change in the coal type based on a time series of measurement values.
- the control device 200 according to the second embodiment specifies the coal type based on the time-series comparison of the measurement values, and determines whether there is a change in the coal type based on the frequency spectrum.
- the configuration of the control device 200 according to the second embodiment is the same as that of the first embodiment, but the processing content in each processing unit of the control device 200 and the information stored in the property storage unit 204 are the same as those in the first embodiment. Different.
- the control device 200 includes a measurement value acquisition unit 201, a measurement value storage unit 202, a frequency analysis unit 203, a property storage unit 204, a property specification unit 205, and a property change determination unit. 206, a correction value determination unit 207, a torque command acquisition unit 208, a correction unit 209, and a control unit 210.
- the property storage unit 204 relates the time series of the input current value of the table motor 105 when the coal according to the coal type is pulverized and the torque command correction value in association with the coal type. A correction function for specifying is stored.
- the property specifying unit 205 is input to the roller mill 100 based on a comparison between a time series of measurement values stored in the measurement value storage unit 202 and a time series of current values related to each coal type stored in the property storage unit 204. Identify the coal type.
- the property specifying unit 205 includes a cumulative value in a time series of current values relating to a plurality of coal types stored in the property storage unit 204 and a cumulative value in a time series of measurement values stored in the measurement value storage unit 202. A difference is calculated
- the property change determination unit 206 determines whether or not the coal type of the coal input to the roller mill 100 has changed based on the frequency spectrum obtained by the frequency analysis unit 203. For example, the frequency analysis unit 203 analyzes the first partial time series related to the latest analysis target time and the first partial time series from the time series of measurement values stored in the measurement value storage unit 202. Frequency analysis is performed for each of the second partial time series related to the target time. The property change determination unit 206 determines whether there is a change in the coal type based on the area difference between the frequency spectrum related to the first partial time series and the frequency spectrum related to the second partial time series.
- FIG. 5 is a flowchart showing the operation of the roller mill control device according to the second embodiment.
- the control device 200 performs correction amount determination processing shown below for each analysis target time.
- the frequency analysis unit 203 reads a partial time series related to the latest analysis target time from the time series of measurement values stored in the measurement value storage unit 202 (step S21).
- the property specifying unit 205 calculates the difference between the time-series accumulated value of each current value stored in association with a plurality of coal types by the property storage unit 204 and the partial time-series accumulated value read in step S21. Then, the coal type having the smallest difference in accumulated values is specified (step S22).
- the property change determination unit 206 includes the first partial time series related to the latest analysis target time and the analysis target before the start point of the first partial time series among the time series of measurement values stored in the measurement value storage unit 202.
- the second partial time series related to time is read (step S23).
- the frequency analysis unit 203 frequency-converts each of the read first partial time series and second partial time series, thereby obtaining a first frequency spectrum and a second frequency spectrum, respectively (step S24).
- the property change determination unit 206 calculates the accumulation of deviations for each frequency between the first frequency spectrum and the second frequency spectrum (step S25). That is, the property change determination unit 206 calculates the difference area between the waveform of the first frequency spectrum and the waveform of the second frequency spectrum.
- the property change determination unit 206 determines whether or not the calculated deviation accumulation is greater than or equal to a predetermined threshold (step S26). When the property change determination unit 206 determines that the accumulated deviation is equal to or greater than the threshold (step S26: YES), the correction value determination unit 207 associates the coal type identified by the property identification unit 205 with the property storage unit 204. The stored correction function is read (step S27). The correction value determination unit 207 calculates the accumulated deviation between the time series of the current values stored in the property storage unit 204 in association with the identified coal type and the time series of the measurement values read in step S1 (step S28). ). The correction value determining unit 207 obtains a torque command correction value by substituting the calculated deviation accumulation into the correction function (step S29).
- step S30 when the property change determination unit 206 determines that the accumulated deviation is less than the threshold (step S26: NO), the correction value determination unit 207 uses the previously determined torque command correction value as the current torque command correction value. (Step S30).
- control device 200 can determine the torque command correction value for the basic torque command acquired by the torque command acquisition unit 208.
- the control device 200 acquires the measured value of the input current of the table motor 105 of the roller mill 100 at a frequency that is at least equal to the frequency of the power supply current of the table motor 105, and calculates the measured value.
- a control signal is output to the roller mill 100 based on the time series.
- the frequency spectrum of the input current of the table motor 105 spreads around the frequency of the power supply current of the table motor 105 and varies depending on the coal type of coal. By using this frequency spectrum, the pulverizer can be automatically and appropriately controlled according to the properties of the coal as in the first embodiment.
- the control apparatus 200 which concerns on 1st Embodiment identifies a coal type by comparing with the frequency spectrum of the standard input current in each coal type about the frequency spectrum of a measured value.
- the property storage unit 204 of the second embodiment identifies the coal type by comparing the time series of measured values with the standard time series of input currents for each coal type.
- the property storage unit 204 according to the third embodiment specifies a coal type by inputting a time series and a frequency spectrum of measured values to a learned machine learning model.
- FIG. 6 is a schematic block diagram illustrating the configuration of the control device according to the third embodiment.
- the control device 200 according to the third embodiment further includes a machine learning unit 211 in addition to the configuration according to the first embodiment.
- the control device 200 according to the third embodiment is different from the configuration of the first embodiment in the information stored in the property storage unit 204, the processing of the property specifying unit 205, and the property change determining unit 206.
- the machine learning unit 211 inputs a time series and a frequency spectrum of the input current of the table motor 105 as inputs, and learns a machine learning model so as to output a torque command correction value. For example, the machine learning unit 211 uses a combination of a torque command correction value and a time series and a frequency spectrum of the input current of the table motor 105 when the coal related to the coal type is crushed as supervised learning. A learning model can be learned. In addition, for example, the machine learning unit 211 determines that the rotational speed needs to be corrected based on fluctuations in the amount of coal supplied when coal associated with various types of coal is crushed with the rotational speed of the rotary separator 109 being constant.
- the machine learning model may be learned by unsupervised learning that derives a feature amount based on the current value at that time and analyzes the tendency.
- this machine learning model is referred to as a correction value specifying model.
- the machine learning unit 211 uses, as teacher data, a combination of the presence / absence of a change in the coal type of coal and the time series and frequency spectrum of the input current of the table motor 105 before and after the change of the input coal type.
- the machine learning model is trained to input the time series and frequency spectrum of the input current of the motor 105 and output the presence / absence of a change in the coal type of coal.
- this machine learning model is referred to as a change determination model.
- the correction value specifying model and the change specifying model are recorded in the property storage unit 204.
- the property storage unit 204 stores a correction value specifying model and a change specifying model.
- the property specifying unit 205 inputs a combination of the time series of the input current stored in the measurement value storage unit 202 and the frequency spectrum analyzed by the frequency analysis unit 203 to the correction value specifying model stored in the property storage unit 204. Thus, the torque command correction value is specified.
- the property change determination unit 206 inputs a combination of the time series of the input current stored in the measurement value storage unit 202 and the frequency spectrum analyzed by the frequency analysis unit 203 into the change determination model stored in the property storage unit 204. Then, it is determined whether or not the coal type of coal has changed.
- FIG. 7 is a flowchart showing the operation of the roller mill control device according to the third embodiment.
- the control device 200 performs correction amount determination processing shown below for each analysis target time.
- the frequency analysis unit 203 reads a partial time series related to the latest analysis target time from the time series of measurement values stored in the measurement value storage unit 202 (step S41).
- the frequency analysis unit 203 performs frequency conversion on the read partial time series to obtain a frequency spectrum (step S42).
- the property specifying unit 205 specifies the torque command correction value by inputting the partial time series and the frequency spectrum to the correction value specifying model stored in the property storage unit 204 (step S43).
- the property change determination unit 206 determines whether or not the coal type of coal has changed by inputting the partial time series and the frequency spectrum into the change determination model stored in the property storage unit 204 (step S44). ).
- step S44 determines that the coal type of coal has changed (step S44: YES)
- the correction value determination unit 207 sets the current torque command correction value to the torque command correction value specified in step S43. Determine (step S45).
- step S44: NO determines that the property change determination unit 206 determines that the accumulated deviation is less than the threshold (step S44: NO)
- the correction value determination unit 207 uses the previously determined torque command correction value as the current torque command correction value. (Step S46).
- the control device 200 can specify the torque command correction value according to the coal type of coal and the presence or absence of the change of the coal type based on machine learning.
- the control apparatus 200 which concerns on 3rd Embodiment specified the torque instruction correction value directly with the correction value specific model, it is not restricted to this.
- the control device 200 according to another embodiment specifies a coal type from a time series and a frequency spectrum of measured values using a coal type identification model that identifies the coal type, and then performs a correction function corresponding to the coal type.
- the torque command correction value may be directly specified by substituting the accumulated frequency spectrum deviation into the correction function.
- machine learning model according to the third embodiment receives a time series of measured values and a frequency spectrum as inputs
- the present invention is not limited to this.
- a machine learning model according to another embodiment may receive only one of a time series of measurement values or a frequency spectrum.
- control apparatus 200 which concerns on embodiment mentioned above specifies the coal type of coal as a property of a grinding
- control device 200 may specify the hardness, moisture content, HGI (Hardgrove Index), and the like of coal.
- control device 200 uses different ones for the specification of the coal type and the determination of the change of the coal type, among the time series of the frequency spectrum and the measurement value, It is not limited to this.
- control device 200 according to another embodiment may use a frequency spectrum for both the identification of the coal type and the determination of the change of the coal type, or may use a time series of measurement values.
- FIG. 8 is a schematic diagram illustrating a configuration of a control system according to another embodiment.
- a control system may be configured by the control device 200 provided in the vicinity of the roller mill 100 and the server device 300 provided remotely.
- the control device 200 includes at least the measurement value acquisition unit 201 and the control unit 210.
- the control device 200 or the server device 300 may include other control units and storage units. .
- a plurality of control devices 200 may be connected to one server device 300.
- control device 200 corresponding to each of the plurality of roller mills 100 acquires a measured value of the input current of the table motor 105 of each roller mill 100 and transmits it to the server device 300, acquires a torque command from the server device 300, and A torque command may be output to the roller mill 100.
- the plurality of roller mills 100 may be provided in the same plant, or may be provided in different plants.
- control target object of the control apparatus 200 which concerns on the above-mentioned embodiment is the roller mill 100
- the control target may be a pump that pumps liquid or a fan that blows gas.
- the roller mill 100 which is a fine pulverizer
- another pulverizer such as a coarse pulverizer, an intermediate pulverizer, or an attritor may be used as a control object.
- the separator motor 110 according to the above-described embodiment is realized by an induction motor, but may be realized by another motor such as a DC motor, a synchronous motor, or a commutator motor.
- FIG. 9 is a schematic block diagram illustrating a configuration of a computer according to at least one embodiment.
- the computer 90 includes a processor 91, a main memory 92, a storage 93, and an interface 94.
- the control device 200 described above is mounted on the computer 90.
- the operation of each processing unit described above is stored in the storage 93 in the form of a program.
- the processor 91 reads out a program from the storage 93, expands it in the main memory 92, and executes the above processing according to the program. Further, the processor 91 secures a storage area corresponding to each of the storage units described above in the main memory 92 according to the program.
- Examples of the storage 93 include HDD (Hard Disk Drive), SSD (Solid State Drive), magnetic disk, magneto-optical disk, CD-ROM (Compact Disc Read Only Memory), DVD-ROM (Digital Versatile Disc Read Only Memory). And semiconductor memory.
- the storage 93 may be an internal medium directly connected to the bus of the computer 90, or may be an external medium connected to the computer 90 via an interface 94 or a communication line. When this program is distributed to the computer 90 via a communication line, the computer 90 that has received the distribution may develop the program in the main memory 92 and execute the above processing.
- the storage 93 is a tangible storage medium that is not temporary.
- the program may be for realizing a part of the functions described above. Further, the program may be a so-called difference file (difference program) that realizes the above-described function in combination with another program already stored in the storage 93.
- difference file difference program
- the control device can automatically and appropriately control the pulverizer according to the properties of the coal.
Landscapes
- Engineering & Computer Science (AREA)
- Food Science & Technology (AREA)
- Power Engineering (AREA)
- Disintegrating Or Milling (AREA)
- Crushing And Grinding (AREA)
- Control Of Electric Motors In General (AREA)
Abstract
La présente invention concerne un dispositif de commande permettant d'acquérir une valeur mesurée d'une grandeur physique relative à l'entrée d'électricité dans un moteur électrique d'un broyeur, ladite acquisition étant effectuée à une fréquence au moins égale à la fréquence d'un courant d'une source d'alimentation du moteur électrique, et d'émettre un signal de commande au broyeur en fonction d'une série chronologique de la valeur mesurée acquise.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PH12020552104A PH12020552104A1 (en) | 2018-06-07 | 2020-12-04 | Control device, control system, control method, and program |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2018109827A JP7457452B2 (ja) | 2018-06-07 | 2018-06-07 | 制御装置、制御システム、制御方法および制御プログラム |
JP2018-109827 | 2018-06-07 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2019235604A1 true WO2019235604A1 (fr) | 2019-12-12 |
Family
ID=68769882
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2019/022671 WO2019235604A1 (fr) | 2018-06-07 | 2019-06-07 | Dispositif, système et procédé de commande, et programme |
Country Status (4)
Country | Link |
---|---|
JP (1) | JP7457452B2 (fr) |
PH (1) | PH12020552104A1 (fr) |
TW (1) | TWI764003B (fr) |
WO (1) | WO2019235604A1 (fr) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114308367A (zh) * | 2021-12-27 | 2022-04-12 | 江苏丰尚智能科技有限公司 | 物料细度的调节方法、装置、计算机设备和存储介质 |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2022070156A (ja) * | 2020-10-26 | 2022-05-12 | 株式会社アーステクニカ | 破砕機の破砕負荷制御装置及び方法 |
CN115400844B (zh) * | 2022-11-01 | 2023-01-06 | 合肥中亚建材装备有限责任公司 | 一种立式辊磨机内部自分级装置 |
CN116889922B (zh) * | 2023-07-06 | 2024-03-08 | 碎得机械(北京)有限公司 | 一种破碎机的控制方法、装置及*** |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0731898A (ja) * | 1993-07-26 | 1995-02-03 | Mitsubishi Heavy Ind Ltd | 回転式分級機付微粉炭機およびその運転方法 |
JPH11333319A (ja) * | 1998-03-24 | 1999-12-07 | Mitsubishi Heavy Ind Ltd | 石炭粉砕性自動推定装置 |
JP2001157851A (ja) * | 1999-12-02 | 2001-06-12 | Mitsubishi Heavy Ind Ltd | 石炭粉砕性自動推定方法とその装置 |
US7182283B1 (en) * | 2004-12-17 | 2007-02-27 | Engineering Consultants Group, Inc. | Pulverizer real-time monitoring system |
JP2007061727A (ja) * | 2005-08-31 | 2007-03-15 | Chugoku Electric Power Co Inc:The | ミル運転異常診断方法 |
JP2007523414A (ja) * | 2004-02-19 | 2007-08-16 | シーメンス アクチエンゲゼルシヤフト | 機械における摩耗を検出するための方法 |
JP2010288352A (ja) * | 2009-06-10 | 2010-12-24 | Nippon Steel Corp | 設備の異常診断方法 |
JP2016195524A (ja) * | 2015-04-02 | 2016-11-17 | 三菱電機株式会社 | 電動機の診断装置 |
WO2017154091A1 (fr) * | 2016-03-08 | 2017-09-14 | 株式会社日立製作所 | Dispositif de diagnostic de machine rotative et procédé de diagnostic |
JP2017209595A (ja) * | 2016-05-23 | 2017-11-30 | クボタ環境サ−ビス株式会社 | 破砕機の遠隔監視システム、破砕機監視装置、破砕機管理サーバ、及び、端末装置 |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS59216643A (ja) * | 1983-05-23 | 1984-12-06 | 太平洋セメント株式会社 | 粉砕機の運転制御方法 |
JP2872868B2 (ja) * | 1992-10-27 | 1999-03-24 | シャープ株式会社 | コーヒーミル |
JPH09239287A (ja) * | 1996-03-08 | 1997-09-16 | Mitsubishi Heavy Ind Ltd | 石炭粉砕用ミルに付属しているロータリセパレータの回転数制御方法とその装置 |
CN107442262B (zh) * | 2017-08-04 | 2019-04-26 | 东北大学 | 一种盘磨***的粉体粒度分布控制装置及方法 |
-
2018
- 2018-06-07 JP JP2018109827A patent/JP7457452B2/ja active Active
-
2019
- 2019-06-06 TW TW108119780A patent/TWI764003B/zh active
- 2019-06-07 WO PCT/JP2019/022671 patent/WO2019235604A1/fr active Application Filing
-
2020
- 2020-12-04 PH PH12020552104A patent/PH12020552104A1/en unknown
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0731898A (ja) * | 1993-07-26 | 1995-02-03 | Mitsubishi Heavy Ind Ltd | 回転式分級機付微粉炭機およびその運転方法 |
JPH11333319A (ja) * | 1998-03-24 | 1999-12-07 | Mitsubishi Heavy Ind Ltd | 石炭粉砕性自動推定装置 |
JP2001157851A (ja) * | 1999-12-02 | 2001-06-12 | Mitsubishi Heavy Ind Ltd | 石炭粉砕性自動推定方法とその装置 |
JP2007523414A (ja) * | 2004-02-19 | 2007-08-16 | シーメンス アクチエンゲゼルシヤフト | 機械における摩耗を検出するための方法 |
US7182283B1 (en) * | 2004-12-17 | 2007-02-27 | Engineering Consultants Group, Inc. | Pulverizer real-time monitoring system |
JP2007061727A (ja) * | 2005-08-31 | 2007-03-15 | Chugoku Electric Power Co Inc:The | ミル運転異常診断方法 |
JP2010288352A (ja) * | 2009-06-10 | 2010-12-24 | Nippon Steel Corp | 設備の異常診断方法 |
JP2016195524A (ja) * | 2015-04-02 | 2016-11-17 | 三菱電機株式会社 | 電動機の診断装置 |
WO2017154091A1 (fr) * | 2016-03-08 | 2017-09-14 | 株式会社日立製作所 | Dispositif de diagnostic de machine rotative et procédé de diagnostic |
JP2017209595A (ja) * | 2016-05-23 | 2017-11-30 | クボタ環境サ−ビス株式会社 | 破砕機の遠隔監視システム、破砕機監視装置、破砕機管理サーバ、及び、端末装置 |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114308367A (zh) * | 2021-12-27 | 2022-04-12 | 江苏丰尚智能科技有限公司 | 物料细度的调节方法、装置、计算机设备和存储介质 |
Also Published As
Publication number | Publication date |
---|---|
TW202000308A (zh) | 2020-01-01 |
PH12020552104A1 (en) | 2021-08-02 |
JP2019209296A (ja) | 2019-12-12 |
JP7457452B2 (ja) | 2024-03-28 |
TWI764003B (zh) | 2022-05-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2019235604A1 (fr) | Dispositif, système et procédé de commande, et programme | |
CN105264181B (zh) | 用于监视旋转叶片的健康状况的方法和*** | |
CN104374675A (zh) | 一种磨煤机煤粉细度在线监测方法 | |
KR102483606B1 (ko) | 분쇄기의 열화 진단 장치, 분쇄 시스템, 및 열화 진단 방법 | |
CN1910434A (zh) | 在机械中诊断机械问题的方法和装置 | |
JP2019209296A5 (ja) | 制御装置、制御システム、制御方法および制御プログラム | |
CN105268536B (zh) | 一种自适应在线校正的火电厂球磨机负荷软测量方法 | |
CN106563537A (zh) | 基于筒体表面抛落区和滑动区振动信号的磨机负荷检测方法 | |
Agrawal et al. | A unified thermo-mechanical model for coal mill operation | |
JP5132746B2 (ja) | 動作異常検出装置、動作異常検出方法 | |
Si et al. | Experimental investigation of load behaviour of an industrial scale tumbling mill using noise and vibration signature techniques | |
JPS6327652B2 (fr) | ||
Tavakoli et al. | A Machine Learning approach to fault detection in transformers by using vibration data | |
CN112742544B (zh) | 一种中速磨煤机动态分离器的调试方法及*** | |
CN110701085A (zh) | 一种风机振动在线监测及管理*** | |
CN207571518U (zh) | 矿用风机监测装置 | |
JP5223506B2 (ja) | 粉砕ミルの寿命評価方法 | |
CN108636586A (zh) | 确定磨煤机出力修正系数的方法及装置 | |
Lahdelma et al. | Intelligent condition monitoring for lime kilns | |
Jha et al. | Fault diagnosis of Coal Ring Crusher in Thermal Power plant: A case study | |
RU2358133C2 (ru) | Система контроля узлов газовой турбины и способ ее работы | |
Rakić | Early fault detection and isolation in coal mills based on self-organizing maps | |
JP2020107025A (ja) | データ選別装置及び方法、ならびに監視診断装置 | |
JPH1038257A (ja) | 石炭性状による安定燃焼の推定方法 | |
JP2019155211A (ja) | 粉砕機及びその監視方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 19815203 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 19815203 Country of ref document: EP Kind code of ref document: A1 |