Moisture prediction device for dryer, and dryer

Abstract

To bring the drying quality and / or treatment time closer to a desired level in a drying device.SOLUTION: The prediction device comprises a distribution acquiring unit configured to acquire a present moisture distribution of the drying objects, an evaluation unit configured to evaluate the characteristics of the present moisture distribution, and a prediction information determination unit configured to determine prediction information relating to the result of the drying treatment in the dryer according to the result of the evaluation obtained by the evaluation unit.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present invention relates to a moisture prediction technique for a dryer. [Background technology]

[0002] Circulation-type grain dryers (hereinafter simply referred to as "dryer") that dry grain while circulating it have been known for some time. For this type of dryer, there is a demand for reducing the variation in the moisture content of the processed grain (improving drying quality). Because drying quality and processing time are in a trade-off relationship, it is desirable to balance improved drying quality and reduced processing time to a desired degree depending on the situation. For this reason, a dryer such as that described in Patent Document 1 below has been developed. This dryer predicts the standard deviation of the grain after drying processing based on the average and standard deviation of the grain's initial moisture content and a prediction model, and sets the pause drying time after the drying operation based on the prediction result. This allows for the setting of an appropriate pause drying time according to the initial moisture content of the grain being processed, thereby improving drying quality while preventing the pause drying time from being set too long. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 6765742 Summary of the Invention [Problem to be solved by the invention]

[0004] However, the above-described dryer leaves room for improvement. For example, the above-described dryer does not perform any control based on the initial moisture value of the grain during drying operation, so there is room for improvement in moisture control during drying operation. Furthermore, because the above-described dryer focuses only on the average moisture value and standard deviation of the grain, the desired drying quality is not necessarily achieved. Specifically, even if the standard deviation is the same, the moisture distribution characteristics are not necessarily the same. However, the above-described dryer performs uniform control based on the predicted standard deviation when the standard deviation is the same. As a result, there is a risk that appropriate control based on the moisture distribution characteristics will not be achieved. For example, moisture control can be performed accurately when the moisture distribution is close to a normal distribution, but accuracy deteriorates when it is not. Furthermore, because the above-described dryer cannot predict the end time of operation, it cannot be operated to meet the user's desired time conditions. For these reasons, technology that can more closely approximate the desired drying quality and / or processing time is desired is desired. The above-described problems are not limited to circulation-type dryers, but are common to various types of dryers. Furthermore, the above-mentioned problems are not limited to those used for grains, but are common to dryers used to treat various objects. [Means for solving the problem]

[0005] The present invention has been made to solve at least part of the above-mentioned problems, and can be realized, for example, in the following forms.

[0006] According to a first aspect of the present invention, there is provided a prediction device for a dryer, comprising: a distribution acquisition unit configured to acquire a current moisture distribution of an object to be dried; an evaluation unit configured to evaluate characteristics of the current moisture distribution; and a prediction information determination unit configured to determine prediction information regarding a drying treatment result of the dryer based on the evaluation result by the evaluation unit.

[0007] The prediction device can be realized in any form, for example, in the form of a controller mounted on the dryer or in the form of an information processing device (computer). In this case, the information processing device may acquire the current moisture distribution of the object to be dried from the dryer via wired or wireless communication. Alternatively, when a user connects a storage medium storing the current moisture distribution of the object to the information processing device, the information processing device may acquire the current moisture distribution of the object to be dried from the storage medium. The dryer may be a circulation dryer. The "current moisture distribution of the object to be dried" may be the moisture distribution of the object to be dried before the drying process or the moisture distribution of the object to be dried during the drying process. The "drying process" refers to one or any combination of at least two of the drying operation, circulation operation, agitation operation, low-temperature finishing operation, and pause operation of the circulation dryer, and is typically the drying operation.

[0008] This prediction device can obtain predictive information regarding the results of the drying process in a dryer based on the current moisture distribution. The predictive information may be, for example, the operation end time, which is the time required for the moisture content of the material to be dried to reach a target moisture content, and / or the future moisture distribution, which is the moisture content after a predetermined time of drying process in the dryer. Furthermore, the prediction device evaluates the characteristics of the current moisture distribution and determines the predictive information based on the evaluation results, thereby enabling more accurate predictions based on the characteristics of the current moisture distribution compared to prediction methods that do not reflect the characteristics of the current moisture distribution. Based on the predictive information obtained in this manner, the operation settings and / or operation control of the dryer can be performed to bring the drying quality and / or processing time closer to the desired level.

[0009] According to the second aspect of the present invention, in the first aspect, the characteristics of the current moisture distribution include the magnitude of the average moisture value. Since the magnitude of the average moisture value has a significant effect on the drying process result, according to this aspect, more accurate prediction can be performed.

[0010] According to a third aspect of the present invention, in the first or second aspect, the characteristics of the current moisture distribution include a shape of the current moisture distribution. Since the shape of the moisture distribution has a significant effect on the drying process results, this aspect allows for more accurate prediction.

[0011] According to a fourth aspect of the present invention, in the third aspect, the evaluation of the shape of the current moisture distribution by the evaluation unit includes an evaluation of the magnitude relationship between the moisture value or moisture value range corresponding to the peak of the current moisture distribution and at least one reference value. Since the position of the peak of the current moisture distribution has a significant impact on the drying process results, this aspect allows for more accurate prediction.

[0012] According to a fifth aspect of the present invention, in the third or fourth aspect, the evaluation of the shape of the current moisture distribution by the evaluation unit includes an evaluation of whether the current moisture distribution is unimodal or multimodal. Whether the current moisture distribution is unimodal or multimodal has a significant impact on the results of the drying process, so this aspect allows for more accurate prediction.

[0013] According to a sixth aspect of the present invention, in any one of the first to fifth aspects, the evaluation unit is configured to classify the current moisture distribution into one of a plurality of predetermined classes. The prediction information determination unit is configured to determine prediction information based on the class classified by the evaluation unit. According to this aspect, by classifying the current moisture distribution, characteristics of the current moisture distribution can be easily and efficiently evaluated.

[0014] According to a seventh aspect of the present invention, in the sixth aspect, the predetermined classes are set based on differences in the magnitude of the average moisture value corresponding to the current moisture distribution and differences in the shape of the current moisture distribution. According to this aspect, classification is performed based on two factors that have a significant effect on the drying process result, so that highly accurate prediction can be made.

[0015] According to an eighth aspect of the present invention, in any one of the first to seventh aspects, the prediction information includes an operation end time, which is the time required for the moisture content of the object to be dried to reach a target moisture content. According to this aspect, the operation settings and / or operation control of the dryer can be performed so as to approach the desired operation end time.

[0016] According to a ninth aspect of the present invention, in any one of the first to eighth aspects, the prediction information includes a future moisture distribution, which is a moisture distribution after a predetermined time of drying processing by the dryer. According to this aspect, the operation setting and / or operation control of the dryer can be performed so that the moisture distribution of the object to be dried after the drying processing approaches a desired moisture distribution (in other words, drying quality).

[0017] According to a tenth aspect of the present invention, in any of the sixth and seventh aspects and the ninth aspect including the sixth aspect, the prediction information includes an operation end time, which is the time required for the moisture content of the object to be dried to reach a target moisture content. The prediction device includes a storage device that stores a prediction model generated by learning using artificial intelligence, the prediction model including a current moisture content distribution as an explanatory variable and a future moisture content distribution, which is the moisture content distribution after a predetermined time of drying processing by the dryer, as a dependent variable. The prediction information determination unit inputs the current moisture content distribution into the prediction model to obtain a future moisture content distribution, predicts a first operation end time, which is the time required for the moisture content of the object to be dried to reach the target moisture content, based on the future moisture content distribution, corrects the first operation end time based on the classified class to obtain a second operation end time, and determines the second operation end time as the operation information time of the prediction information. According to this aspect, by correcting the first operation end time based on the class classified by the evaluation unit, deterioration in prediction accuracy due to differences in the characteristics of the current moisture content distribution can be compensated for, thereby enabling the operation end time (second operation end time) to be predicted with high accuracy. Moreover, since there is no need to prepare a prediction model for each of a plurality of predetermined classes, the prediction device is easy to manufacture.

[0018] According to an eleventh aspect of the present invention, there is provided a dryer. The dryer includes the prediction device according to any one of the first to tenth aspects, and an operating condition determination unit that determines operating conditions or candidate operating conditions for the dryer based on the prediction information determined by the prediction information determination unit. This dryer can achieve the same effects as any one of the first to tenth aspects.

[0019] According to a twelfth aspect of the present invention, there is provided a method for predicting the result of a drying process performed by a dryer. This method includes acquiring a current moisture distribution of an object to be dried, evaluating characteristics of the current moisture distribution, and determining prediction information regarding the result of the drying process based on the evaluation result of the characteristics of the current moisture distribution. This method can achieve the same effects as the first aspect. A feature corresponding to any one of the second to tenth aspects may be added to the twelfth aspect.

[0020] According to a thirteenth aspect of the present invention, there is provided an operating parameter setting device for a dryer. The operating parameter setting device includes a distribution acquisition unit configured to acquire a current moisture distribution of an object to be dried, an evaluation unit configured to evaluate characteristics of the current moisture distribution, and a determination unit that determines operating parameters of the dryer or candidate operating parameters based on the evaluation results by the evaluation unit. This operating parameter setting device makes it possible to appropriately set operating parameters of the dryer based on the characteristics of the moisture distribution of the object to be dried. When the dryer is configured to be selectively operable in one of a plurality of operating modes with different drying speeds, the operating parameters may include information on which of the plurality of operating modes and / or at least one combined operating mode, which is a combination of at least two of the plurality of operating modes and is switched over over time, to operate in.

[0021] The present invention is not limited to the above-described embodiments and can be embodied in any form. For example, the present invention can be realized as a method for predicting the moisture content of an object to be dried, a method for predicting the operation end time, a method for setting operation parameters, a moisture prediction program, a program for predicting the operation end time, or a program for setting operation parameters, or as a storage medium that stores any of these programs in a computer-readable manner. [Brief explanation of the drawings]

[0022] [Figure 1] 1 is an external perspective view of a circulation type grain dryer according to one embodiment. FIG. [Figure 2] FIG. 2 is a diagram showing a schematic internal structure of a circulation type grain dryer. [Figure 3] FIG. 2 is a functional block diagram of a controller. [Figure 4] 10 is a flowchart illustrating an example of an operation mode determination process. [Figure 5] FIG. 10 is a diagram illustrating a plurality of classes for evaluating moisture distribution characteristics by classification. [Figure 6] FIG. 10 is a diagram illustrating a method for evaluating the characteristics of the shape of the moisture distribution. [Figure 7] FIG. 10 is a diagram illustrating a method for evaluating the characteristics of the shape of the moisture distribution. [Figure 8] FIG. 10 is a diagram illustrating a method for evaluating the characteristics of the shape of the moisture distribution. [Figure 9] FIG. 10 is a diagram illustrating an example of a GUI used in the operation mode determination process. [Figure 10] 1 is a diagram showing various examples of moisture distribution in grain. [Figure 11] 10 is a flowchart illustrating an example of an automatic operation mode change process. [Figure 12] 10 is a flowchart illustrating an example of a control parameter setting process. [Figure 13] 10 is a flowchart illustrating an example of a control parameter setting process. DETAILED DESCRIPTION OF THE INVENTION

[0023] FIG. 1 is an external perspective view of a circulation-type grain dryer 10 (hereinafter simply referred to as the dryer 10) according to one embodiment. FIG. 2 is a diagram showing a schematic internal structure of the dryer 10. In the following example, the dryer 10 is a device for drying brown rice as an example of grain. As shown in FIG. 1, the dryer 10 includes a storage chamber 20 for storing grain, a drying section 30 disposed below the storage chamber 20 and configured to dry the grain, and an elevator device 60. As shown in FIG. 2, the drying section 30 includes a hot air chamber 31, grain flow-down chambers 32 and 33, and an exhaust air chamber 34. Hot air generated by a combustion burner (not shown) is blown into the hot air chamber 31. The grain flow-down chambers 32 and 33 are vertically connected to the storage chamber 20 and extend downward in a long, narrow manner on both sides of the hot air chamber 31. The exhaust air chamber 34 is located outside the grain flow-down chambers 32 and 33. The hot air flowing through the hot air chamber 31 is guided by the action of a suction fan (not shown) through small holes in the casing that forms the grain flow-down chambers 32, 33, through the grain flow-down chambers 32, 33, and into the air exhaust chamber 34. At this time, the hot air comes into contact with the grain in the grain flow-down chambers 32, 33, drying the surface of the grain.

[0024] As shown in FIG. 2, a rotary valve 40, which is an example of a supply device for discharging grain from the drying section 30, is disposed at the lower end of the grain flow-down chambers 32, 33. The rotary valve 40 discharges grain downward from the drying section 30. Specifically, the rotary valve 40 has a cylindrical shape, and its interior functions as a storage chamber for storing grain. An opening 41 is formed in its outer circumferential surface, thereby forming a C-shaped opening in a portion of the circumferential direction. The rotary valve 40 is configured to be rotatable forward and backward by a motor (not shown). In its initial position, the rotary valve 40 is oriented so that the opening 41 faces downward, as shown in FIG. 2.

[0025] When the rotary valve 40 rotates forward (clockwise in FIG. 2) from the initial position, the opening 41 of the rotary valve 40 and the grain flow-down chamber 33 communicate with each other, and the grain in the grain flow-down chamber 33 flows into the rotary valve 40, filling the rotary valve 40 with grain. When the rotary valve 40 rotates further and returns to the initial position, the grain inside the rotary valve 40 falls from the opening 41 and is sent downward (to the bottom of the air exhaust chamber 34).

[0026] When the rotary valve 40 rotates in the reverse direction (counterclockwise in FIG. 2) from the initial position, the opening 41 of the rotary valve 40 and the grain flow-down chamber 32 communicate with each other, and the grain in the grain flow-down chamber 32 flows into the rotary valve 40, filling the rotary valve 40 with grain. When the rotary valve 40 rotates further and returns to the initial position, the grain inside the rotary valve 40 falls from the opening 41 and is sent downward (to the bottom of the air exhaust chamber 34).

[0027] As shown in FIG. 2, a lower screw conveyor 50 is disposed at the bottom of the air exhaust chamber 34. The lower screw conveyor 50 transports the grain discharged by the rotary valve 40 horizontally to the lifting device 60. The lifting device 60 is in the form of a bucket conveyor and is disposed so as to extend vertically as shown in FIG. 1. The lifting device 60 transports the grain upward after drying in the drying section 30 to return the grain to the storage chamber 20. The grain transported to near the top of the dryer 10 by the lifting device 60 is transported to the storage chamber 20 by an upper screw conveyor (not shown). The grain returned to the storage chamber 20 with its surface dried is further dried by the tempering action within the storage chamber 20.

[0028] With this configuration, grain fed into the loading hopper 21 (see Figure 1) passes through the lower screw conveyor 50, the lifting device 60, the storage chamber 20, and the drying section 30 before entering the rotary valve 40. When the rotary valve 40 rotates, the grain returns to the lower screw conveyor 50 and is then circulated along this route until it is dried to a predetermined moisture level and then discharged outside the machine through the discharge port 22. The rotary valve 40, the lower screw conveyor 50, and the lifting device 60 function as a circulation device that sends grain out of the drying section 30 and returns it to the storage chamber 20.

[0029] As shown in FIG. 1 , a moisture meter 70 configured to measure the moisture content of sampled grain is attached to the base of the lifting device 60. The moisture meter 70 includes a rotatable sampling roll (not shown). The sampling roll includes a storage chamber for storing grain to be collected as a sample. A portion of the grain scooped up into the bucket of the lifting device 60 enters the storage chamber through a communication hole (not shown) formed in the lifting device 60 and the moisture meter 70. When the sampling roll is rotated forward, the grain stored in the storage chamber is taken into the measurement unit of the moisture meter 70. When the sampling roll is rotated reversely, the grain stored in the storage chamber is discharged into the lifting device 60. In this embodiment, when the sampling timing arrives, the sampling roll rotates reversely to empty the storage chamber, then returns to its initial position, waits for a predetermined time (the time required for the grain to fill the storage chamber), and then rotates forward to supply the grain stored in the storage chamber to the measurement unit.

[0030] In this embodiment, the moisture meter 70 measures the initial moisture content of the sampled grain when the grain is loaded from the hopper 21 (i.e., the moisture content of the grain before the drying process). The moisture meter 70 also measures the moisture content of the sampled grain (i.e., the moisture content of the grain during drying) at predetermined times (e.g., periodically) after the drying process begins. In this embodiment, the moisture meter 70 measures the moisture content of the grain every 60 minutes. However, the moisture measurement period can be set as desired. In this embodiment, the moisture meter 70 also measures the moisture content of 100 grains at one measurement time. However, the number of grains to be measured can be set as desired.

[0031] In this embodiment, the moisture meter 70 is an electrical resistance type, which requires the destruction of grains to measure the moisture content. As a result, raw material loss occurs in the amount of grain used for the moisture measurement. However, the type of moisture meter 70 is not particularly limited, and may be any type that requires the destruction of grains, or may be a type that does not require the destruction of grains.

[0032] As shown in FIG. 1, the dryer 10 further includes a controller 80 that controls the overall operation of the dryer 10. FIG. 3 is a functional block diagram of the controller 80. As shown in FIG. 3, the controller 80 includes a CPU 81 and a memory 88. Each function of the controller 80 is realized by the CPU 81 executing a predetermined program stored in the memory 88. However, each function of the controller 80 may be realized by a dedicated circuit instead of or in addition to the CPU 81. Furthermore, the functions of the controller 80 may be distributed among multiple controllers. Furthermore, some of the functions of the controller 80 may be realized by an external device (e.g., an information terminal) communicatively connected to the controller 80 in a wired or wireless manner, or may be realized by a standalone external device (e.g., a computer).

[0033] The CPU 81 also functions as a distribution acquisition unit 82, an evaluation unit 83, a prediction information determination unit 84, and an operating condition determination unit 85 by executing programs stored in the memory 88. Details of these functions will be described later. The memory 88 also stores a prediction model 89. Details of the prediction model 89 will also be described later.

[0034] 1, the dryer 10 further includes a display 90. In this embodiment, the display 90 also functions as a touch screen type operation panel (in other words, a user interface).

[0035] The controller 80 controls the drying process of the dryer 10. The drying process includes at least a drying operation. The drying process may also include at least one of a circulation operation, an agitation operation, a low-temperature finishing operation, and a pause operation. For example, the drying process may be a process in which a circulation operation, an agitation operation, a drying operation, and a pause operation are performed in this order.

[0036] In circulation operation, the rotary valve 40 is rotated in only one direction (forward or reverse) with the combustion burner stopped, and the lower screw conveyor 50 and the lifting device 60 are driven. In circulation operation, air blowing by driving the suction fan may or may not be performed. In mixing operation, the rotary valve 40 is rotated alternately forward and reverse with the combustion burner stopped, and the lower screw conveyor 50 and the lifting device 60 are driven. In mixing operation, air blowing by driving the suction fan may or may not be performed. In drying operation, the combustion burner is ignited, and the rotary valve 40 is rotated alternately forward and reverse with the lower screw conveyor 50 and the lifting device 60 driven. Low-temperature finishing operation can be optionally employed when the moisture content falls below a predetermined value (when the moisture content approaches the target finishing moisture content by a predetermined amount or more). In low-temperature finishing operation, the combustion burner is turned down in power compared to drying operation, the rotary valve 40 rotates alternately in forward and reverse directions, and the lower screw conveyor 50 and the lifting device 60 are driven. In low-temperature finishing operation, finish drying is performed while lowering the grain temperature, allowing the user to move on to the subsequent process (hulling) quickly. In pause operation, the combustion burner, rotary valve 40, lower screw conveyor 50, and lifting device 60 are stopped, but the grain remains inside the dryer 10.

[0037] Furthermore, the controller 80 is configured to be able to selectively execute one of a plurality of operation modes with different drying speeds as the drying operation. In this embodiment, the operation modes are set to a "fast mode" with the fastest drying speed, a "slow mode" with the slowest drying speed, and a "normal mode" with a drying speed between the "fast mode" and the "slow mode." Each of these operation modes has a predetermined rate of reduction in the moisture content of the grain per hour. The controller 80 sets (changes) control parameters related to heating power and air volume as needed to achieve the reduction rate corresponding to the selected operation mode.

[0038] The dryer 10 described above has a function for predicting the moisture distribution of grain after a predetermined drying time and a function for predicting the time required for the moisture content of the grain to reach a target moisture content from the predicted time (e.g., from the start of processing by the dryer 10) (hereinafter also referred to as the operation end time), and can perform various processes using these functions. Such processes are described below. FIG. 4 is a flowchart showing an example of an operation mode determination process. In this example, the operation mode determination process is a process for determining whether the drying process should be performed in the "fast mode," "normal mode," or "slow mode" described above before the start of the drying process. The operation mode determination process is initiated by a user operating the GUI displayed on the display 90. The operation mode determination process is typically performed after the user loads grain from the hopper 21 and the initial moisture content of the grain is measured by the moisture meter 70. However, the operation mode determination process may also be performed after, for example, a circulation operation or a stirring operation is performed by a user.

[0039] As shown in FIG. 4, when the operation mode determination process is started, the controller 80, as processing by the distribution acquisition unit 82, acquires the current moisture distribution of the grain (here, the initial moisture distribution) based on the moisture value measured by the moisture meter 70, and displays a histogram of the acquired moisture distribution on the display 90 (step S110).

[0040] Next, the controller 80 accepts a priority mode input by the user (step S120). In this embodiment, a time priority mode and a quality priority mode are provided as priority modes. The time priority mode is a mode that prioritizes reducing the time required for the drying process over improving the drying quality. In the time priority mode, the user can set a condition regarding the allowable time required from the start to the end of the drying process (hereinafter also referred to as a time condition). The quality priority mode is a mode that prioritizes improving the drying quality over reducing the time required for the drying process. In the quality priority mode, the user can set a condition regarding the allowable degree of variation in moisture value (hereinafter also referred to as a quality condition).

[0041] Next, the controller 80 determines whether the priority mode accepted in step S120 is the time priority mode or the quality priority mode (step S130). If it is determined in step S130 that the time priority mode has been accepted, the controller 80 accepts the time conditions (i.e., the maximum allowable operation time) and the target finishing moisture value input by the user (step S140).

[0042] Next, controller 80 predicts the moisture distribution after completion of the drying process for each operating mode (i.e., for each of "fast mode," "normal mode," and "slow mode") (step S150). This prediction is performed using a prediction model 89 stored in memory 88. Prediction model 89 is a model that includes the current moisture distribution and the elapsed time T from the present of the drying process as explanatory variables, and uses the future moisture distribution (more specifically, the moisture distribution after the drying process has progressed for time T from the present) as a dependent variable. To improve the prediction accuracy of prediction model 89, the number of samples taken by moisture meter 70 may be increased only when measuring the initial moisture.

[0043] Using this prediction model 89, the moisture distribution after the drying process is determined as follows. First, the prediction information determination unit 84 of the controller 80 inputs the current moisture distribution and the elapsed time T into the prediction model 89 to obtain the moisture distribution after the time T has elapsed. This process is performed while changing the elapsed time T. For example, the input to the prediction model 89 and the output from the prediction model 89 are repeated while increasing the elapsed time T by one hour. Then, the prediction information determination unit 84 determines, from among the multiple moisture distributions obtained in this manner, the moisture distribution whose average value reaches the target finishing moisture value as the moisture distribution after the drying process is completed. For example, the moisture distribution after the time T has elapsed is obtained while changing T=1, 2, 3,..., 11, 12, 13, 14. If the average value reaches the target finishing moisture value for the first time at T=13, the prediction information determination unit 84 determines the moisture distribution corresponding to T=13 as the moisture distribution after the drying process is completed. Furthermore, the prediction information determination unit 84 determines the elapsed time T (T=13 in the above example) corresponding to the moisture distribution after the drying process is completed as the first operation end time. The first operation end time is a provisional prediction value regarding the time required for the moisture content of the grain to reach the target moisture content after the process by the dryer 10 is started.

[0044] In this embodiment, the prediction model 89 is generated by artificial intelligence learning (e.g., machine learning, deep learning). In this embodiment, the prediction model 89 is generated by the controller 80 based on the measurement results of the moisture meter 70 during past operations of the dryer 10. That is, each time the dryer 10 is operated, the controller 80 performs learning by storing, as learning data, in the memory 88, the elapsed time T (including zero) and the moisture distribution of the moisture values measured by the moisture meter 70 at the elapsed time T. The controller 80 also performs re-learning based on the measurement results of the moisture meter 70 during a new operation of the dryer 10, thereby updating the prediction model 89. Therefore, the accuracy of the prediction model 89 can be improved as the dryer 10 is used. A prediction model generated in advance by experiment, etc., may be used until the amount of learning data required to generate the prediction model 89 is accumulated. In an alternative embodiment, the prediction model 89 may be a function defined based on experiments.

[0045] In this embodiment, the controller 80 generates a prediction model 89 for each operation mode (i.e., for each of the "fast mode," "normal mode," and "slow mode"). For example, when a drying operation is performed in the "fast mode," the measurement results of the moisture meter 70 obtained at that time are used to generate the prediction model 89 for the "fast mode." The prediction information determination unit 84 then predicts the moisture distribution using the prediction model 89 corresponding to the processing conditions (operation mode). For example, when predicting the moisture distribution when the "fast mode" is adopted, the prediction model 89 for the "fast mode" is used. This configuration can improve the prediction accuracy of the prediction model 89.

[0046] In an alternative embodiment, the prediction model 89 includes an operating mode as an explanatory variable. Specifically, the controller 80 generates the prediction model 89 based on learning data in which the type of operating mode is associated with the moisture distribution. The prediction information determination unit 84 then inputs the current moisture distribution, the elapsed time T, and the type of operating mode into the prediction model 89 to obtain a predicted value of the moisture distribution after the time T has elapsed.

[0047] In a further alternative embodiment, the prediction model 89 includes the target finish moisture value of the grain as an explanatory variable. Specifically, the controller 80 generates the prediction model 89 based on training data in which the target finish moisture value is associated with a moisture distribution. The prediction information determiner 84 then inputs the current moisture distribution, the elapsed time T, and the target finish moisture value into the prediction model 89 to obtain a predicted value of the moisture distribution after the time T has elapsed. Because the moisture distribution after drying is affected by the target finish moisture value, this configuration improves the prediction accuracy of the prediction model 89. In a further alternative embodiment, the controller 80 generates the prediction model 89 for each value of the target finish moisture value or for each predetermined range. The prediction information determiner 84 then predicts the moisture distribution using the prediction model 89 corresponding to the processing conditions (the value or range of the target finish moisture value).

[0048] In a further alternative embodiment, the prediction model 89 includes environmental information as an explanatory variable. Specifically, the controller 80 generates the prediction model 89 based on learning data in which environmental information and moisture distribution are associated with each other. The prediction information determination unit 84 then inputs the current moisture distribution, the elapsed time T, and the environmental information into the prediction model 89 to obtain a predicted value of the moisture distribution after the time T has elapsed. Such environmental information may be, for example, temperature, humidity, atmospheric pressure, wind speed, vibration, infrared intensity, ultraviolet intensity, light intensity, sound volume, weather information (rainy weather, sunny weather, etc.), etc. The controller 80 may acquire the environmental information from various sensors included in the dryer 10. Alternatively, the controller 80 may acquire the environmental information from a weather information service provided on the Internet. Because drying performance is affected by environmental factors, this configuration can improve the prediction accuracy of the prediction model 89. In a further alternative embodiment, the controller 80 generates the prediction model 89 for each value of the environmental information or for each predetermined range. The prediction information determination unit 84 then predicts the moisture distribution using a prediction model 89 that corresponds to the processing conditions (values or ranges of the environmental information). Note that the explanatory variables of the prediction model 89 are not limited to the above examples and may include any factors that affect drying performance, and may include, for example, grain variety information.

[0049] Next, the controller 80 evaluates the characteristics of the initial moisture distribution received in step S110, corrects the first operation end time, and determines the second operation end time (step S160). Specifically, the controller 80 first evaluates the characteristics of the initial moisture distribution as processing by the evaluation unit 83. In this embodiment, the characteristics to be evaluated include the magnitude of the average moisture value (hereinafter also referred to as the moisture value characteristic) and the shape of the initial moisture distribution (hereinafter also referred to as the shape characteristic), as will be described in detail later. However, the characteristics to be evaluated may include only one of these two characteristics, or may include any other characteristic in addition to or instead of one or both of these two characteristics.

[0050] In this embodiment, the evaluation of the characteristics is performed by classification. Specifically, the evaluation unit 83 of the controller 80 classifies the initial moisture distribution into one of a plurality of predetermined classes. In this embodiment, classes C1 to C12 shown in FIG. 5 are defined as such a plurality of classes. The classes C1 to C12 are set based on the differences in the moisture value characteristics and the differences in the shape characteristics.

[0051] First, the classification based on the characteristics of the moisture value will be described. The characteristics of the moisture value are classified by an evaluation regarding the magnitude relationship between the average value AV of the moisture value (hereinafter also referred to as the moisture average value AV), the first evaluation reference value R1 and the second evaluation reference value R2 (R1 < R2) regarding the moisture value. Specifically, as illustrated in FIG. 5, the characteristics of the moisture value include a low moisture group (classes C1 to C4) where the moisture average value AV is less than the first evaluation reference value R1, a medium moisture group (classes C5 to C8) where the moisture average value AV is greater than or equal to the first evaluation reference value R1 and less than the second evaluation reference value R2, and a high moisture group (classes C9 to C12) where the moisture average value AV is greater than or equal to the second evaluation reference value R2. The first evaluation reference value R1 and the second evaluation reference value R2 can be experimentally set to arbitrary values. In an alternative embodiment, instead of the first evaluation reference value R1 and the second evaluation reference value R2, one evaluation reference value or three or more evaluation reference values may be set. The number of classifications based on the characteristics of the moisture value increases or decreases according to the number of evaluation reference values set.

[0052] Next, the classification based on the characteristics of the shape will be described. In the present embodiment, the characteristics of the shape are classified by an evaluation regarding the magnitude relationship between the moisture value or moisture value range corresponding to the peak of the initial moisture distribution, the third evaluation reference value R3 and the fourth evaluation reference value R4 (R3 < R4) regarding the moisture value. This evaluation assesses the bias in the position of the peak of the initial moisture distribution in the entire moisture distribution (all data intervals). The moisture value corresponding to the peak of the moisture distribution is, when each data interval of the moisture distribution is set to include a plurality of values, the representative value (for example, the average value, the median, etc.) of the individual moisture values included in the data interval with the maximum occurrence frequency (also referred to as the frequency. In other words, the number of grains) among all the data intervals of the moisture distribution. Also, the moisture value corresponding to the peak of the moisture distribution is equal to the single value corresponding to the data interval with the maximum occurrence frequency when each data interval of the moisture distribution is set as a single value. The moisture value range corresponding to the peak of the moisture distribution is the data interval with the maximum occurrence frequency among all the data intervals of the moisture distribution. In the following description, the moisture value or moisture value range corresponding to the peak of the moisture distribution is also referred to as the peak value.

[0053] Specifically, the shape features are classified into a "left-leaning" group whose peak value is less than the third evaluation standard value R3, a "normal" group whose peak value is equal to or greater than the third evaluation standard value R3 and less than the fourth evaluation standard value R4, and a "right-leaning" group whose peak value is equal to or greater than the fourth evaluation standard value R4, based on the magnitude relationship between the peak value of the moisture distribution peak P1 and the third and fourth evaluation standard values R3 and R4. Figure 6 shows an example of moisture distribution in the right-leaning group. Using this method, the shape features of the moisture content (more specifically, the features related to the bias) are classified into a "normal" group (classes C1, C5, C9), a "left-leaning" group (classes C2, C6, C10), and a "right-leaning" group (classes C3, C7, C11), as shown in Figure 5.

[0054] As shown in FIG. 6 , in this embodiment, the third evaluation reference value R3 and the fourth evaluation reference value R4 are set so that the average moisture content AV falls between the third evaluation reference value R3 and the fourth evaluation reference value R4. This setting allows for proper evaluation of the bias in the position of peak P1 across the entire moisture distribution. Furthermore, in this embodiment, the third evaluation reference value R3 is set greater than the first evaluation reference value R1, and the fourth evaluation reference value R4 is set smaller than the second evaluation reference value R2. In other words, the moisture range defined by the third evaluation reference value R3 and the fourth evaluation reference value R4 is located inside the moisture range defined by the first evaluation reference value R1 and the second evaluation reference value R2. This setting allows for proper evaluation of the bias in the position of peak P1 across the entire moisture distribution. For example, setting the moisture range defined by the third evaluation reference value R3 and the fourth evaluation reference value R4 too broad can prevent classification results from concentrating on a specific group. However, the magnitude relationship between the first evaluation reference value R1 and the second evaluation reference value R2 and the third evaluation reference value R3 and the fourth evaluation reference value R4 can be set arbitrarily. For example, R1=R2 and R3=R4. The third evaluation reference value R3 and the fourth evaluation reference value R4 can be experimentally set to arbitrary values.

[0055] In an alternative embodiment, one or two evaluation criteria values, or four or more evaluation criteria values, may be set instead of the third evaluation criteria value R3 and the fourth evaluation criteria value R4. The number of classifications based on shape characteristics (bias) increases or decreases depending on the number of evaluation criteria values set.

[0056] Furthermore, in this embodiment, the shape features are classified based on an evaluation of whether the initial moisture distribution is unimodal or multimodal. Specifically, this evaluation is performed based on the positional relationship between the first peak P2 and the second peak P3. The first peak P2 is a point showing an occurrence frequency corresponding to the data interval with the highest occurrence frequency among all data intervals of the moisture distribution. The second peak P3 is a point showing an occurrence frequency corresponding to the data interval with the highest occurrence frequency among data intervals that satisfy the multimodal determination conditions described below.

[0057] 7 and 8, in this embodiment, the multimodal determination condition is that the difference H in height (i.e., the frequency of occurrence) between the first peak P2 and the second peak P3 is smaller than a fifth evaluation criterion value R5, and the width W (i.e., the width of the moisture value) between the first peak P2 and the second peak P3 is greater than a sixth evaluation criterion value R6. This determination is made by extracting the second peak P3 having an occurrence frequency corresponding to the data section with the highest occurrence frequency among data sections whose width W is greater than the sixth evaluation criterion value R6, and determining whether the difference H in height between the extracted second peak P3 and the first peak P2 is smaller than the fifth evaluation criterion value R5. In this embodiment, if the multimodal determination criterion is met as a result of the determination, the evaluation unit 83 determines that the moisture distribution to be evaluated is a multimodal moisture distribution having a peak with the first peak P2 and a peak with the second peak P3. On the other hand, if the multi-modal criterion is not met, the evaluation unit 83 determines that the second peak P3 is not a mountain peak. In other words, the evaluation unit 83 determines that the moisture distribution being evaluated is unimodal. The height difference H and width W can be experimentally set to any values.

[0058] For example, as shown in FIG. 8, the second peak P3 satisfies the multimodal determination condition described above, and therefore the moisture distribution shown is determined to be a multimodal moisture distribution having a mountain with the first peak P2 and a mountain with the second peak P3. On the other hand, the third peak P4 does not satisfy the condition regarding the height difference H described above, and therefore is determined not to be a mountain peak. Similarly, the fourth peak P5 does not satisfy the condition regarding the width W described above, and therefore is determined not to be a mountain peak. FIG. 5 shows an example of the results of classification performed in this manner. In this example, classes C4, C8, and C12 are multimodal, and classes C1 to C3, C5 to C7, and C9 to C11 are unimodal.

[0059] Using this method, the shape characteristics of the moisture content (more specifically, the multimodal characteristics) are classified into a multimodal group (classes C4, C8, C12) and a unimodal group (classes C1 to C3, C5 to C7, C9 to 11), as shown in Figure 5.

[0060] Classes C1 to C12 shown in Fig. 5 are set by combining the classification based on differences in moisture value characteristics described above with classification based on differences in shape characteristics (bias characteristics and multimodality characteristics). However, the number of classes is not limited to the above example, and any arbitrary number of classes may be set. For example, multiple classes may be set based on only some of the various characteristics described above. Alternatively, multiple classes may be set based on other characteristics (for example, standard deviation of moisture value) instead of or in addition to at least some of the various characteristics described above.

[0061] Returning now to FIG. 4 for further explanation, in step S160, the evaluation unit 83 evaluates the characteristics of the moisture distribution by classifying the initial moisture distribution into one of a plurality of predetermined classes C1 to C12 in this manner. Then, as processing by the prediction information determination unit 84, the controller 80 corrects the first operation end time determined in step S150 based on the evaluation result by the evaluation unit 83, i.e., the class into which the initial moisture distribution has been classified (one of classes C1 to C12), to determine a second operation end time. The second operation end time is a final prediction value related to the time required for the moisture content of the grain to reach the target moisture value after processing by the dryer 10 has begun, and is hereinafter also referred to simply as the operation end time.

[0062] The correction of the first operation end time is performed by adding a correction value CV (unit: hours) to the first operation end time. The correction value CV may be a positive value, a negative value, or zero. Such a correction value CV may be experimentally determined in advance. That is, the correction value CV may be set by collecting data regarding the difference between the first operation end time determined in step S150 and the operation end time (hereinafter also referred to as the actual operation end time) when an operation is actually performed to process grain having the corresponding initial moisture distribution, and statistically analyzing the collected data. Alternatively, the controller 80 may update the correction value CV by learning artificial intelligence based on the operating history of the dryer 10. In this embodiment, the correction value CV is set as a value common to each operation mode. However, the correction value CV may also be set individually for each operation mode.

[0063] For example, if the initial moisture distribution is classified into class C5, which corresponds to the shape of an ideal moisture distribution, the correction value CV may be set to any value satisfying -1≦CV≦1 (including zero). Alternatively, if the initial moisture distribution is classified into class C1, the correction value CV may be set to any value satisfying 3≦CV. Alternatively, if the initial moisture distribution is classified into class C6 or C10, the correction value CV may be set to any value satisfying 1≦CV<3. Alternatively, if the initial moisture distribution is classified into any of classes C7 to C9, the correction value CV may be set to any value satisfying -3≦CV<-1. Alternatively, if the initial moisture distribution is classified into class C11, the correction value CV may be set to any value satisfying -3>CV. In this way, the correction value CV can be set arbitrarily depending on the actual error between the first operation end time and the actual operation end time.

[0064] Next, the controller 80 outputs and displays on the display 90 the operation modes that satisfy the time condition and the corresponding moisture distributions (the moisture distributions after the drying process that correspond to the operation modes that satisfy the time condition among the moisture distributions predicted in step S150) (step S170). An operation mode that satisfies the time condition is an operation mode whose operation end time (second operation end time) obtained in step S160 is the same as or shorter than the time condition. If there are multiple operation modes that satisfy the time condition, the moisture distribution after the drying process may be displayed for each of those operation modes. Alternatively, of the multiple operation modes that satisfy the time condition, an operation mode with a relatively slow drying speed and the corresponding moisture distribution after the drying process may be displayed.

[0065] On the other hand, if it is determined in step S130 that the quality priority mode has been accepted, the controller 80 accepts the quality conditions and target finishing moisture value input by the user (step S180). The quality conditions are conditions related to the acceptable drying quality, and are typically index values that represent the acceptable degree of variation in moisture value. In this embodiment, the quality conditions are the standard deviation of the moisture value.

[0066] Next, the controller 80 predicts the moisture distribution after completion of the drying process for each operation mode (step S190). The process of step S190 is performed in the same manner as the process of step S150. At this time, the controller 80 determines the first operation end time in the same manner as step S150. Next, the controller 80 evaluates the characteristics of the initial moisture distribution received in step S110, corrects the first operation end time, and determines the second operation end time (step S200). This process is the same as the process of step S160.

[0067] Next, the controller 80 outputs and displays on the display 90 the operation modes that satisfy the quality conditions and the corresponding moisture distributions (the moisture distributions after drying that correspond to the operation modes that satisfy the quality conditions among the moisture distributions predicted in step S190) (step S210). If there are multiple operation modes that satisfy the quality conditions, the moisture distribution after drying may be displayed for each of those operation modes. Alternatively, of the multiple operation modes that satisfy the quality conditions, the operation mode with a relatively fast drying speed and the corresponding moisture distribution after drying may be displayed.

[0068] 9 shows an example of a graphical user interface (GUI) 91 used to display the moisture distribution in steps S170 and S210. In this example, the average moisture value and standard deviation corresponding to the initial moisture distribution acquired in step S110 are displayed as "moisture value" and "moisture variation," respectively, in a first area 92. The initial moisture distribution is also displayed in the form of a histogram in the first area 92.

[0069] Furthermore, the second area 93 displays the "operation end time," "moisture value," and "moisture variation," and also displays the moisture distribution after the drying process in the form of a histogram. The "moisture value" represents the target finishing moisture value accepted in step S140 or S180. If the time priority mode is selected in step S120, the "operation end time" represents the operation time (time condition) accepted in step S140, and the "moisture variation" represents the predicted value of the standard deviation of the moisture after the drying process. If the quality priority mode is selected in step S120, the "moisture variation" represents the standard deviation of the moisture value (quality condition) accepted in step S180, and the "operation end time" represents the second operation end time determined in step S200.

[0070] Returning to FIG. 4 for the explanation, after the moisture distribution is displayed on the display 90, the controller 80 then waits for the user to input an operation start command or input to reselect the priority mode (steps S220 and S230). Specifically, the user checks the information displayed on the display 90 in step S170 or step S210 (see FIG. 9). If the user wishes to check the prediction results for another priority mode, the user can reselect the priority mode by pressing the time priority button 94 or the quality priority button 95 (see FIG. 9). On the other hand, if the user accepts the moisture distribution after the drying process and other predicted values (the standard deviation of the moisture value in the time priority mode, or the operation end time in the quality priority mode) displayed on the display 90, the user can issue an operation start command by pressing the operation start button 96 (see FIG. 9).

[0071] Then, when controller 80 receives an input to reselect the priority mode (step S220: NO and step S230: YES), it returns the process to step S130. On the other hand, when controller 80 receives an input to start operation (step S220: YES), it determines to adopt the operation mode (in other words, the operation conditions) displayed in step S170 or step S210 as the process of the operation condition determination unit 85, and starts the drying process (step S240). In this way, the operation mode determination process ends.

[0072] According to such an operation mode determination process, the user can input desired conditions (for example, the time required from the start to the end of the drying process or the degree of moisture value variation) and check the future moisture distribution corresponding to the operation mode that satisfies the conditions before starting operation of the dryer 10. Therefore, the user can appropriately set the operation mode to achieve the desired conditions.

[0073] In addition, the user can easily check the characteristics of the initial moisture distribution and the moisture distribution after the drying process in the form of a histogram. Therefore, the drying quality obtained in each operating mode can be easily understood, and an appropriate operating mode can be selected. This point will be specifically explained with reference to FIG. 10. For drying quality, a moisture distribution that is unimodal and close to a normal distribution, as shown in pattern A in FIG. 10, is desirable, and the narrower the distribution width (i.e., the smaller the variation), the more desirable it is. Such a moisture distribution is less likely to produce grains with moisture values that deviate significantly from the average value, and is less likely to experience moisture fluctuations (equilibration) after discharge from the dryer 10. On the other hand, a multimodal moisture distribution such as pattern B, a comb-shaped moisture distribution such as pattern C, or a discrete moisture distribution such as pattern F is more likely to experience moisture fluctuations after discharge. Furthermore, moisture distributions that are biased to one side, such as patterns D and E, may experience moisture fluctuations in a direction other than the direction toward the average value after discharge. According to the above-described operation mode determination process, the user can easily grasp the differences in the shapes of the moisture distributions as described above from the histogram, and can therefore appropriately select an operation mode so that a moisture distribution having a shape similar to pattern A is obtained.

[0074] Furthermore, to assist the user, the controller 80 may display operation assistance information on the display 90 according to the moisture distribution prediction results. Such operation assistance information may be, for example, a message recommending the adoption of an operation mode corresponding to a moisture distribution that has been highly evaluated after evaluating the shape of the moisture distribution predicted for each operation mode. With this configuration, the user can operate the dryer 10 appropriately even if they have little knowledge about operating the dryer 10.

[0075] In this moisture distribution evaluation, for example, a unimodal moisture distribution may be given a high evaluation. In this case, a smoothing process may be performed on the histogram, and the number of peaks may be determined for the smoothed histogram. The smoothing process may be performed, for example, by widening the data intervals (also called bins) of the histogram, or by thinning out the larger or smaller of the frequencies of two adjacent data intervals. In this way, the overall shape of the histogram can be properly evaluated without being affected by small local features. Alternatively, the number of peaks in the moisture distribution may be determined using the method described above with reference to FIG. 7.

[0076] Alternatively, the moisture distribution may be evaluated by setting an ideal moisture distribution and determining the degree of similarity between the moisture distribution predicted for each operation mode and the ideal moisture distribution. In this case, the greater the degree of similarity, the higher the evaluation. Such a determination of the degree of similarity may be performed, for example, by a chi-square test of goodness of fit.

[0077] Alternatively, the moisture distribution may be evaluated by previously setting a plurality of classes as described above with reference to Fig. 5 and classifying the moisture distribution predicted for each operation mode into one of the plurality of classes. In this case, a merit or demerit of the evaluation may be set in advance for each of the plurality of classes.

[0078] In an alternative embodiment, in steps S150 and S190, the controller 80 may predict moisture distribution after completion of at least one drying process corresponding to at least one combined operation mode, instead of or in addition to each operation mode ("fast mode," "normal mode," and "slow mode"). A combined operation mode refers to a combination of at least two operation modes that are switched over over time, and is set in advance. For example, a combined operation mode that combines "slow mode" and "normal mode," or a combined operation mode that combines "normal mode" and "fast mode," etc., can be set. For each combined operation mode, the timing of switching the operation mode is also set, such as switching to "normal mode" after four hours in "slow mode."

[0079] In this alternative embodiment, in steps S150 and S190, the moisture distribution after the completion of the drying process corresponding to a combined operation mode in which, for example, the "slow mode" is switched to the "normal mode" after four hours has elapsed is obtained as follows. The controller 80 first inputs the initial moisture distribution and the elapsed time T=4 into the prediction model 89 corresponding to the "slow mode" to obtain the moisture distribution after four hours has elapsed. Then, the moisture distribution after four hours has elapsed and the elapsed time T are input into the prediction model 89 corresponding to the "normal mode" to obtain the moisture distribution at the elapsed time T. This process is performed while changing the elapsed time T. For example, the input to the prediction model 89 and the output from the prediction model 89 are repeated while increasing the elapsed time T by one hour. The controller 80 then determines, from among the multiple moisture distributions thus obtained, the moisture distribution whose average value reaches the target finishing moisture value as the moisture distribution after the completion of the drying process. The operation end time is corrected in the same way as in steps S160 and S200. The correction value CV may be set as a common value for each combined operation mode, or may be set individually for each combined operation mode.

[0080] In this alternative embodiment, in steps S170 and S210, among all the operation modes and / or combinations of operation modes, the operation modes and / or combinations of operation modes that satisfy the time condition or the quality condition are displayed together with the corresponding histograms. With this configuration, it is possible to set an operation mode that allows for more precise control of the drying quality.

[0081] In a further alternative embodiment, the input of the priority mode may not be accepted in step S120. In this case, the prediction information determination unit 84 may predict the moisture distribution after completion of the drying process for each of the operation modes and / or combination operation modes. The controller 80 may then simultaneously, sequentially, or selectively display each predicted moisture distribution on the display 90. This configuration allows the user to visually compare the moisture distributions after completion of the drying process corresponding to each of the operation modes and / or combination operation modes before starting operation of the dryer 10. This allows the user to easily set the optimal operation mode. In this alternative embodiment, operation assistance information may also be displayed on the display 90.

[0082] In a further alternative embodiment, in addition to the operation mode and / or the combined operation mode, a combination of at least one of the circulation operation, the agitation operation, the low-temperature finishing operation, and the pause operation with the drying operation may be set in advance, and the moisture distribution after the drying treatment may be predicted for each combination. In this case, the moisture distribution may be predicted in the same manner as in the combined operation mode, and a prediction model 89 may be prepared for each of the circulation operation, the agitation operation, the low-temperature finishing operation, the pause operation, and the drying operation.

[0083] In a further alternative embodiment, the controller 80 may output various moisture distribution information at any timing and in any format. For example, the controller 80 may output data so that at least two of the following can be simultaneously or sequentially displayed: a histogram of the current (latest) measured moisture distribution, a histogram of the predicted moisture distribution after the drying process, and a histogram of the predicted moisture distribution during the drying process. This configuration improves the user's convenience in checking the moisture distribution. For example, during the drying process, a histogram of the current (latest) measured moisture distribution and a moisture distribution histogram for the current time point predicted before the start of the drying process may be displayed. By comparing the two, the user can confirm whether the drying operation is progressing as predicted. Furthermore, a distribution frequency table may be output instead of or in addition to the histogram. Furthermore, the histogram is not limited to a one-dimensional form as shown in FIG. 10, but may also be in the form of a heat map that shows the temporal transition of the moisture distribution in two dimensions.

[0084] In a further alternative embodiment, in step S180, controller 80 may accept, as a quality condition, an allowable time required from the start to the end of the drying process (hereinafter also referred to as the allowable drying time) instead of the standard deviation of the moisture value. Because the drying quality improves as the drying process time increases, the allowable drying time can also be treated as a quality condition. In this case, the allowable drying time that can be accepted in step S180 may be limited to a time longer than at least some of the time conditions that can be accepted in step S140. For example, the time condition that can be accepted in step S140 may be any time equal to or greater than 5 hours, and the drying process time that can be accepted in step S180 may be any time equal to or greater than 10 hours. In step S210, controller 80 may output and display on display 90 the moisture distribution after the drying process that corresponds to an operating mode that satisfies the allowable drying time accepted in step S180 as a quality condition (i.e., an operating mode that can complete the drying process within the allowable drying time).

[0085] Next, the automatic operation mode changing process will be described with reference to Fig. 11. The automatic operation mode changing process is a process in which, after the drying operation has been started in the operation mode determined by the operation mode determination process, the operation mode is automatically changed depending on the progress of grain drying. This process is repeatedly executed by the controller 80 during the drying operation. As shown in Fig. 11, when the automatic operation mode changing process is started, the controller 80 waits until the timing for predicting the moisture distribution (step S310). In this embodiment, the prediction timing is the timing when the moisture meter 70 has performed a moisture measurement.

[0086] Then, when the prediction timing arrives (step S310: YES), controller 80, as processing by distribution acquisition unit 82, acquires the latest moisture distribution based on the moisture value measured by moisture meter 70 (step S320). Next, controller 80 re-predicts the moisture distribution after completion of the drying process for each operation mode (step S330). This processing is the same as step S150 except that the latest moisture distribution acquired in step S320 is input into prediction model 89. At this time, controller 80 determines the first operation end time, as in step S150.

[0087] Next, the evaluation unit 83 of the controller 80 evaluates the latest moisture distribution characteristics acquired in step S320, corrects the first operation end time, and determines the second operation end time (step S340). This process is the same as step S160.

[0088] Next, in the operation mode determination process, controller 80 determines which priority mode was selected when the drying operation was started (step S350). If it is determined in step S350 that the drying operation was started when the time priority mode was selected, controller 80 determines, as processing by operation condition determination unit 85, whether an operation mode with a slower drying speed than the current operation mode exists and whether the time condition (see step S140) is satisfied even if the operation mode is changed to the operation mode with the slower drying speed (step S360). This determination is made based on the second operation end time determined in step S340, which corresponds to each of the moisture distributions re-predicted in step S330.

[0089] If the time condition is satisfied even when the operation mode is changed to a slower drying speed (step S360: YES), the operation condition determination unit 85 of the controller 80 changes the operation mode to a slower drying speed operation mode and terminates the automatic operation mode change process (step S370). This process makes it possible to improve the drying quality as much as possible while satisfying the time condition in accordance with the actual moisture distribution during the drying operation. On the other hand, if there is no operation mode with a slower drying speed than the current operation mode, or if changing to the operation mode with the slower drying speed would not satisfy the time condition (step S360: NO), the operation condition determination unit 85 of the controller 80 terminates the automatic operation mode change process without changing the operation mode.

[0090] On the other hand, if it is determined in step S350 that the drying process has started with the quality priority mode selected, the controller 80 determines whether an operating mode with a faster drying speed than the current operating mode exists and whether the quality conditions (see step S180) are satisfied even if the operating mode is changed to the operating mode with the faster drying speed (step S380). This determination is made based on the moisture distribution re-estimated in step S330.

[0091] If the quality condition is satisfied even when the operation mode is changed to a faster drying mode (step S380: YES), the controller 80 changes the operation mode to the faster drying mode and terminates the automatic operation mode change process (step S390). This process allows the drying operation time to be shortened as much as possible while satisfying the quality condition, depending on the actual moisture distribution during the drying operation. On the other hand, if there is no operation mode with a faster drying speed than the current operation mode, or if changing to the faster drying mode would not satisfy the quality condition (step S380: NO), the controller 80 terminates the automatic operation mode change process without changing the operation mode. In an alternative embodiment, steps S380 and S390 may be omitted. That is, even if the quality condition is satisfied even when the operation mode is changed to a faster drying mode, the controller 80 may maintain the current operation mode. This alternative embodiment allows the quality condition to be more reliably satisfied when the drying process is completed. In other words, the quality condition is more likely to be satisfied even if the moisture distribution does not progress as well as expected in the future.

[0092] In an alternative embodiment, when controller 80 receives the allowable drying time as a quality condition in step S180, controller 80 may determine in step S380, as processing by operating condition determination unit 85, whether an operating mode with a slower drying speed than the current operating mode exists and whether changing to the operating mode with the slower drying speed will satisfy the allowable drying time received in step S180. In this case, controller 80 may change the operating mode to the operating mode with the slower drying speed in step S390 if the allowable drying time received in step S180 will be satisfied even if the operating mode is changed to the operating mode with the slower drying speed, and may then terminate the automatic operating mode change processing.

[0093] Whether or not to execute such an automatic operation mode changing process may be set in advance by the user via a GUI displayed on the display 90. In an alternative embodiment, before executing step S370 or step 390, the controller 80 may output information for recommending the user to change the operation mode, as an example of the operation assistance information. This process is executed as a process of the operation assistance unit 86. The output destination of the operation assistance information may be, for example, the display 90 or an information terminal capable of communicating with the dryer 10.

[0094] The criteria for changing the operation mode (steps S360 and S380 in the above example) can be set as appropriate. Furthermore, instead of determining whether to change the operation mode based on the predicted moisture distribution, the determination may be made based on the latest moisture distribution acquired in step S320. In this case, the evaluation results of the moisture distribution shape using the various techniques described above may be used as the criteria. For example, if a highly evaluated moisture distribution shape has already been obtained during the drying process, the operation mode may be switched to an operation mode with a relatively fast drying speed.

[0095] For example, a criterion for changing the operating mode may be whether the moisture distribution predicted in step S330 after the most recent drying process is less desirable (worse) than the moisture distribution predicted at the start of the drying process. In this case, if the moisture distribution after the most recent drying process is undesirable, the controller 80 may change the current operating mode to one that achieves an improved moisture distribution, or may implement a change to introduce at least one of circulation operation, agitation operation, low-temperature finishing operation, and pause operation, or may output operation support information recommending such a change to the user. With this configuration, the operating mode can be changed so as not to deteriorate the drying quality even if the processing environment deteriorates, for example, if the weather changes to rain during the drying process and the outside humidity increases.

[0096] 11 shows a configuration in which the drying section 30 is controlled based on the predicted moisture distribution, but a circulation device may be controlled instead of or in addition to the drying section 30. For example, whether to introduce at least one of a circulation operation, an agitation operation, a low-temperature finishing operation, and a pause operation, and the duration thereof may be determined according to the predicted moisture distribution during the drying operation.

[0097] Next, the control parameter setting process will be described with reference to Figures 12 and 13. The control parameter setting process is a process for setting control parameters (heat power, air volume, etc.) for the drying process based on the moisture distribution of the grain. This process is executed when the drying operation starts. When the control parameter setting process is started, the controller 80 first acquires the latest moisture distribution as a process by the distribution acquisition unit 82 (step S410).

[0098] Next, the controller 80 predicts the moisture distribution after a predetermined time based on the moisture acquired in step S410 as processing by the prediction information determination unit 84 (step S420). The predetermined time here is equal to the cycle of moisture measurement by the moisture meter 70. In this embodiment, the cycle is one hour, so this predetermined time is also one hour. The processing in step S420 is the same as the processing in step S150.

[0099] Next, the controller 80 determines whether the prediction in step S420 is an initial prediction, that is, whether the prediction is based on an initial moisture distribution (step S430). If the determination result indicates that the prediction in step S420 is an initial prediction (step S430: YES), the controller 80 calculates statistics (e.g., mean value and standard deviation) of the moisture distribution predicted in step S420 and sets control parameters based on these statistics (step S460).

[0100] On the other hand, if the prediction in step S420 is the second or subsequent prediction (step S430: NO), the controller 80 evaluates the similarity between the latest moisture distribution acquired in step S410 and the moisture distribution predicted in step S420 (step S440). This process evaluates the similarity between the actually measured moisture distribution and the predicted moisture distribution at the same time point. In other words, step S440 evaluates whether the moisture distribution is progressing as predicted to a predetermined extent with respect to the prediction result as the drying process proceeds. The similarity evaluation in step S440 may be performed, for example, by a chi-square test of goodness of fit.

[0101] Next, controller 80 determines whether the evaluated similarity is equal to or greater than a threshold value (step S450). If the determination result is that the similarity is equal to or greater than the threshold value (step S450: YES), that is, if the moisture distribution is progressing somewhat as predicted, the predicted moisture distribution is considered to be more reliable (closer to the actual value of the entire grain being dried) than a moisture distribution actually measured using a limited number of samples. Therefore, controller 80 proceeds to the above-mentioned step S460, where it sets control parameters based on the statistics of the predicted moisture distribution (i.e., based on more reliable statistics).

[0102] Next, controller 80 waits until the timing of the next moisture measurement (step S470). Then, at the timing of the next moisture measurement, moisture meter 70 performs measurement (step S480). Next, controller 80 determines whether the moisture value (here, the average value) measured in step S480 has reached the target finishing moisture value (step S490). If the measured moisture value has reached the target finishing moisture value (step S490: YES), controller 80 ends the control parameter setting process. This also ends the drying operation.

[0103] On the other hand, if the measured moisture value has not reached the target finishing moisture value (step S490: NO), the controller 80 returns the process to step S420. As a result, the controller 80 re-predicts the moisture distribution after a predetermined time (one hour after the previous prediction in step S420) without changing the moisture distribution input to the prediction model 89.

[0104] On the other hand, if the similarity is less than the threshold value in the determination of step S450 (step S450: NO), that is, if the moisture distribution does not change as predicted, the predicted result of the moisture distribution is considered to be unreliable. Therefore, the controller 80 sets the control parameters based on the statistics of the actually measured moisture distribution (the moisture distribution acquired in step S410) (that is, based on relatively reliable statistics) (step S500).

[0105] Next, the controller 80 waits until the timing of the next moisture measurement (step S510). Then, at the timing of the next moisture measurement, the moisture meter 70 performs a measurement (step S520). Next, the controller 80 determines whether the moisture value (here, the average value) measured in step S520 has reached the target finishing moisture value (step S530). If the measured moisture value has reached the target finishing moisture value (step S530: YES), the controller 80 ends the control parameter setting process. This also ends the drying operation.

[0106] On the other hand, if the measured moisture value has not reached the target finishing moisture value (step S530: NO), the controller 80 returns the process to step S410. As a result, in step S410, the controller 80 changes the moisture distribution input to the prediction model 89 to the latest moisture distribution, and in step S420, it re-predicts the moisture distribution after a predetermined time (one hour after the previous prediction in step S420). This increases the possibility that control based on the predicted moisture distribution will be possible thereafter.

[0107] Based on this control parameter setting process, when the predicted and actual measured values for moisture distribution at the same time point (i.e., after a predetermined drying time) are similar to each other to a predetermined degree or more, moisture control can be performed more accurately based on the more reliable predicted value. Alternatively, when the predicted and actual measured values are similar to each other to a predetermined degree or more, there is no need to increase the number of grains sampled to improve the reliability of the actual measured values, so the total number of grains sampled can be reduced. Therefore, when measuring the moisture content of grains using a moisture meter that requires destruction of the grain (a common, inexpensive moisture meter), raw material loss due to sampling can be reduced.

[0108] In an alternative embodiment, if the determination in step S450 indicates that the similarity is less than the threshold (step S450: NO), the controller 80 may increase the number of samples or the sampling frequency for obtaining the measured moisture distribution on which the control parameters are set until the similarity becomes equal to or greater than the threshold. This configuration can improve reliability even when moisture control is performed based on the measured value.

[0109] 12 and 13 show a configuration in which the drying section 30 (control parameters of the drying section 30) is controlled based on a predicted moisture distribution or an actually measured moisture distribution, but a circulation device may be controlled instead of or in addition to the drying section 30. For example, the possibility of introducing at least one of a circulation operation, an agitation operation, a low-temperature finishing operation, and a pause operation, and the duration of each operation may be determined. For example, if the similarity between the predicted moisture distribution and the actually measured moisture distribution is smaller than a predetermined level, it may be determined that uneven filling has occurred, and a decision may be made to introduce an agitation operation to alleviate the uneven filling.

[0110] The process of determining candidate operating parameters executed by the dryer 10 will be described below. The candidate operating parameters are, for example, information on which of at least two of the above-mentioned "fast mode," "normal mode," "slow mode," and combined operating mode the dryer will operate in. To determine candidate operating parameters, the evaluation unit 83 of the controller 80 evaluates the characteristics of the current moisture distribution acquired by the distribution acquisition unit 82. Then, the prediction information determination unit 84 of the controller 80 determines candidate operating parameters based on the evaluation results by the evaluation unit 83.

[0111] The evaluation method for the characteristics of the current moisture distribution is performed by classifying the current moisture distribution into one of a plurality of classes, similar to step S160. Furthermore, a rating (for example, in the form of rating points) is preset for each of the plurality of classes. The higher the rating, the more the operating condition determination unit 85 determines, as a candidate operating parameter, an operating mode or combination operating mode that provides a faster drying speed.

[0112] According to an alternative embodiment, as described above with reference to FIG. 10 , a unimodal moisture distribution may be given a higher rating. In this case, a smoothing process may be performed on the histogram, and the number of peaks in the smoothed histogram may be determined. For example, the operating condition determination unit 85 may determine the "normal mode" or the "slow mode" as a candidate operating parameter when the moisture distribution is unimodal. Furthermore, the operating condition determination unit 85 may determine the "slow mode" as a candidate operating parameter when the moisture distribution is multimodal. Alternatively, the moisture distribution may be evaluated by setting an ideal moisture distribution and determining the degree of similarity between the moisture distribution predicted for each operating mode and the ideal moisture distribution. In this case, the greater the degree of similarity, the higher the rating. Such a determination of similarity may be performed, for example, by a chi-square test of goodness of fit.

[0113] Once the candidate operating parameters have been determined in this manner, the controller 80 outputs the candidate operating parameters determined by the operating condition determination unit 85 to the display 90. Furthermore, the controller 80 accepts an instruction from the user via the display 90 to indicate whether or not to adopt the candidate operating parameters. Upon receiving an instruction to accept the adoption of the candidate operating parameters, the controller 80 starts operation of the dryer 10 with the candidate operating parameters. This controller 80 can appropriately set the operating parameters of the dryer 10 based on the characteristics of the moisture distribution of the material to be dried. The operating condition determination unit 85 may determine the operating parameters instead of determining the candidate operating parameters. In this case, the controller 80 may start operation of the dryer 10 with the determined operating parameters.

[0114] According to an alternative embodiment, the controller 80 may output operation assistance information to the display 90 or the like based on the moisture prediction result (prediction result of the moisture distribution after the drying process is completed) when the drying operation is performed in the "normal mode" instead of the evaluation result of the current moisture distribution characteristics. More specifically, based on the evaluation result of the predicted moisture distribution. The predicted moisture distribution can be evaluated using methods similar to various evaluation methods for the current moisture distribution characteristics. In this case, the operation assistance information may include information regarding suggested operating parameters to be set in the dryer. The controller 80 may then accept user input regarding whether or not to adopt the suggested operating parameters. The operating parameters may include a combination of one of a plurality of operating modes and a combination operating mode with at least one of circulation operation, agitation operation, low-temperature finishing operation, and pause operation.

[0115] For example, if the moisture distribution prediction result when drying operation is performed in the "normal mode" is very good (e.g., the evaluation of the class into which the predicted moisture distribution is classified is equal to or higher than a first threshold), the controller 80 may present operating parameters for performing drying in the "fast mode" followed by a low-temperature finishing operation, along with a predicted value (T1) of the time required until the corresponding drying process is completed. Furthermore, if the moisture distribution prediction result when drying operation is performed in the "normal mode" is good (e.g., the evaluation of the class into which the predicted moisture distribution is classified is equal to or higher than a second threshold that is lower than the first threshold), the controller 80 may present operating parameters for performing low-temperature finishing operation after drying in the "normal mode" along with a predicted value (T2) of the time required until the corresponding drying process is completed. Furthermore, if the moisture distribution prediction result when drying is performed in the "normal mode" is poor (for example, the evaluation of the class into which the predicted moisture distribution is classified is less than the second threshold), the controller 80 may present operating parameters for performing a low-temperature finishing operation after drying in the "slow mode" along with a predicted value (T3) of the time required to complete the corresponding drying process. The low-temperature finishing operation is introduced, for example, when the moisture content reaches near the target moisture content (target moisture content + 1%) during drying in any of the "fast mode," "normal mode," and "slow mode" operating modes (switching to the low-temperature finishing operation from any of the above three operating modes).

[0116] The above-described predicted values T1 to T3 of the required times may be determined based on a predicted value (T0) of the time required for the drying process to be completed when the drying process is performed only in the "normal mode." Typically, the predicted values T1 to T3 are calculated by multiplying the predicted value T0 by a correction coefficient of an arbitrary positive value. The correction coefficient may be determined in advance experimentally, or may be determined based on the operating history of the dryer 10 (the time actually required in past operations). For example, the correction coefficient for the predicted value T1 may be 1.0, the correction coefficient for the predicted value T2 may be 1.1, and the correction coefficient for the predicted value T3 may be 1.4.

[0117] Typically, the "normal mode" is used very frequently, and therefore there is a wealth of driving history for that mode. Therefore, a prediction model 89 for the "normal mode" generated by learning from such a wealth of driving history has a relatively high prediction accuracy. Alternatively, a prediction model 89 applicable to various driving modes, generated by learning from the driving history for various driving modes, has a high prediction accuracy for the "normal mode." According to the above configuration, the high prediction accuracy for the "normal mode" can be utilized to suggest to the user whether or not to adopt various driving parameters.

[0118] In particular, in the above example, the user is presented with candidate operating parameters that combine a drying operation in either the "slow mode," "normal mode," or "fast mode" with a low-temperature finishing operation. With the low-temperature finishing operation, the grain can be discharged from the dryer 10 at a low temperature at the end of the drying process. This reduces moisture transfer between grains after discharge (reducing moisture changes after finishing), preventing changes to higher moisture content. This makes it possible to obtain grain that is less susceptible to the effects of the outside air environment and is of consistent quality.

[0119] In an alternative embodiment, if there is an operating mode other than the "normal mode" that is used frequently, operating assistance information may be output to the display 90 or the like based on the moisture prediction results when drying operation is performed using that operating mode instead of the "normal mode."

[0120] According to a further alternative embodiment, the controller 80 may determine the operating parameters based on predetermined conditions related to the predicted information of the drying process result, and may automatically start the drying process based on the determined operating parameters. The determined operating parameters may be any of a plurality of operating modes and a combination of an operating mode, or a combination of any of the plurality of operating modes and a combination of at least one of a circulation operation, an agitation operation, a low-temperature finishing operation, and a pause operation. The predetermined conditions are input in advance by the user.

[0121] The predetermined conditions may include conditions related to the drying quality (in other words, the moisture distribution finally obtained by the drying process). In this case, the controller 80 may calculate an evaluation value related to the drying quality for each of a plurality of predetermined operating parameter candidates. The predetermined conditions may be thresholds related to the evaluation value. The evaluation value related to the drying quality may be, for example, a similarity to an ideal moisture distribution. Specifically, for example, an ideal moisture distribution may be set, and the similarity between the moisture distribution prediction result and the ideal moisture distribution may be calculated. In this case, a moisture distribution with a higher similarity is given a higher evaluation. Such a similarity may be determined, for example, by a chi-square test of goodness of fit, a T-test, a KS test, or the like. Alternatively, the evaluation value related to the drying quality may be determined by classifying the moisture distribution into one of a plurality of classes as described above with reference to FIG. 5. In this case, an evaluation value may be set in advance for each of the plurality of classes.

[0122] Furthermore, the predetermined conditions may include a condition related to the drying end time (the second drying end time described above) instead of or in addition to the condition related to the drying quality. In this case, the controller 80 determines the drying end time for each of the plurality of predetermined operating parameter candidates. The predetermined condition may be a threshold value related to the drying end time.

[0123] Once the evaluation value for drying quality and / or the drying end time have been determined in this manner, the controller 80 selects an operating parameter candidate that satisfies a predetermined condition from among a plurality of predetermined operating parameter candidates, and automatically starts the drying process. In particular, setting both the evaluation value for drying quality and the drying end time as predetermined conditions can shorten the drying time and achieve efficient drying with minimal drying unevenness. In an alternative embodiment, the controller 80 may output the operating parameter candidate that satisfies the predetermined condition and the corresponding evaluation value for drying quality and / or the operation end time to the display 90 or the like before starting the drying process. The controller 80 may then receive a user input regarding whether or not to select the operating parameter candidate.

[0124] In an alternative embodiment, when low-temperature finishing operation is included in the candidate operating parameters, multiple cases in which the settings of the combustion burner's heat power (hot air temperature) and / or the rotary valve 40's drive cycle time during low-temperature finishing operation are different may each be individually set as candidate operating parameters.

[0125] According to the dryer 10 described above, the operation end time can be obtained as predicted information regarding the drying treatment result in the dryer 10 based on the current moisture distribution. The operation end time is determined by evaluating the characteristics of the current moisture distribution and based on the evaluation results. Therefore, compared to prediction methods that do not reflect the characteristics of the current moisture distribution, the operation end time can be predicted more accurately based on the characteristics of the current moisture distribution. Based on the prediction information obtained in this manner, the operation settings and / or operation control of the dryer 10 can be performed so as to bring the treatment time closer to the desired time.

[0126] Furthermore, in dryer 10, the characteristics of the current moisture distribution evaluated by evaluation unit 83 include the magnitude of the average moisture value and the shape of the current moisture distribution (in the above example, the bias of the peaks in the moisture distribution and whether the moisture distribution is unimodal or multimodal). These have a significant impact on the operation processing time as a result of the drying process, so this configuration allows for more accurate prediction of the operation processing time.

[0127] Furthermore, in the dryer 10, the evaluation unit 83 evaluates the characteristics of the current moisture distribution by classifying the current moisture distribution into one of a plurality of predetermined classes C1 to C12. Therefore, the characteristics of the current moisture distribution can be evaluated easily and efficiently.

[0128] Furthermore, in the dryer 10, the prediction information determination unit 84 corrects the first operation end time based on one of classes C1 to C12 classified by the evaluation unit 83, obtains a second operation end time, and determines the second operation end time as the operation end time (the operation end time as the final prediction result). Therefore, deterioration in prediction accuracy due to differences in the characteristics of the current moisture distribution can be compensated for, and the operation end time can be predicted with high accuracy. Moreover, since there is no need to prepare a prediction model 89 for each of multiple predetermined classes, the dryer 10 is easy to manufacture.

[0129] In an alternative embodiment, the prediction information determination unit 84 may determine a future moisture distribution, which is the moisture distribution after a predetermined time of drying processing by the dryer 10, as prediction information regarding the drying processing result in the dryer 10, based on the evaluation result of the characteristics of the current moisture distribution by the evaluation unit 83. In this case, a prediction model 89 is prepared in advance for each of a plurality of classes. The prediction information determination unit 84 determines the future moisture distribution using the prediction model 89 corresponding to the class into which the current moisture distribution is classified. With this configuration, it is possible to more accurately predict the operation end time for the future moisture distribution based on the characteristics of the current moisture distribution. The prediction information determination unit 84 may determine the operation end time based on the future moisture distribution thus obtained. With this configuration, it is possible to accurately determine the operation end time without correcting the first operation end time to the second operation end time.

[0130] Although the embodiments of the present invention have been described above, the above-described embodiments are intended to facilitate understanding of the present invention and are not intended to limit the present invention. The present invention may be modified or improved without departing from the spirit thereof, and the present invention includes equivalents thereof. Furthermore, any combination or omission of the components described in the claims and specification is possible within the scope of solving at least part of the above-described problems or achieving at least part of the effects.

[0131] For example, the functions of controller 80 (the functions of distribution acquisition unit 82, evaluation unit 83, and prediction information determination unit 84) may be realized in the form of a computer. The computer may be communicatively connected to controller 80 of dryer 10, or may be a stand-alone computer. In the former case, the computer (distribution acquisition unit 82) may acquire the measurement results of moisture meter 70 via communication. In the latter case, the user may connect a storage medium to controller 80 and store the measurement results of moisture meter 70 in the storage medium, and then connect the storage medium to the computer, after which the computer (distribution acquisition unit 82) may acquire the measurement results of moisture meter 70 from the storage medium.

[0132] Furthermore, the above-described flowchart is merely an example, and the order of the processes constituting the flowchart can be changed or can be changed to equivalent processes without departing from the spirit of the present invention.

[0133] Furthermore, the present invention is not limited to grain dryers, but can be realized as a dryer for drying any object to be dried (or a moisture measuring device or operating parameter determining device for such a dryer). Such objects to be dried may be, for example, seeds, wood, resin, or clothing. Furthermore, the present invention is not limited to circulation dryers, but may be realized as a dryer (or a moisture measuring device or operating parameter determining device for such a dryer) in which the object to be dried is stored without being circulated. In this case, a non-contact (e.g., infrared) moisture meter may be used, and the moisture meter may be configured to be movable (to change the measurement location) by an actuator. [Explanation of symbols]

[0134] 10...Circulating grain dryer 20...Storage chamber 21...Hopper 22...Exhaust port 30...Drying section 31...Hot air chamber 32,33...Grain flow chamber 34...Exhaust room 40...Rotary valve 41...Aperture 50...Lower screw conveyor 60...Lifting device 70...Moisture meter 80...Controller 81...CPU 82...Distribution acquisition part 83...Evaluation section 84...Prediction information determination unit 85...Operating condition determination unit 86...Driving Support Department 88...Memory 89...Predictive Model 90...Display 92...First Area 93...Second Area 94...Time Priority Button 95...Quality Priority Button 96...Start button C1~C12...classes

Claims

1. 1. A prediction device for a dryer, comprising: a distribution acquisition unit configured to acquire a current moisture distribution of the object to be dried; an evaluation unit configured to evaluate a characteristic of the current moisture distribution; a prediction information determination unit configured to determine prediction information regarding a drying treatment result in the dryer based on the evaluation result by the evaluation unit; A prediction device comprising:

2. The prediction device according to claim 1 , The characteristics of the current moisture distribution include the magnitude of the average moisture value. Prediction device.

3. 3. The prediction device according to claim 1 or 2, The characteristics of the current moisture distribution include a shape of the current moisture distribution. Prediction device.

4. The prediction device according to claim 3, The evaluation of the shape of the current moisture distribution by the evaluation unit includes an evaluation of the magnitude relationship between the moisture value or moisture value range corresponding to the peak of the current moisture distribution and at least one reference value. Prediction device.

5. The prediction device according to claim 3, The evaluation of the shape of the current moisture distribution by the evaluation unit includes evaluation of whether the current moisture distribution is unimodal or multimodal. Prediction device.

6. 3. The prediction device according to claim 1 or 2, the evaluation unit is configured to classify the current moisture distribution into one of a plurality of predetermined classes; The prediction information determination unit is configured to determine the prediction information based on the class classified by the evaluation unit. Prediction device.

7. The prediction device according to claim 6, The predetermined classes are set based on differences in the magnitude of the average moisture value corresponding to the current moisture distribution and differences in the shape of the current moisture distribution. Prediction device.

8. 3. The prediction device according to claim 1 or 2, The prediction information includes an operation end time, which is the time required for the moisture content of the object to be dried to reach a target moisture content. Prediction device.

9. 3. The prediction device according to claim 1 or 2, The prediction information includes a future moisture distribution, which is a moisture distribution after a predetermined time of drying processing by the dryer. Prediction device.

10. The prediction device according to claim 6, the prediction information includes an operation end time, which is a time required for the moisture content of the object to be dried to reach a target moisture content; the prediction device includes a storage device that stores a prediction model generated by learning of artificial intelligence, the prediction model including the current moisture distribution as an explanatory variable and a future moisture distribution, which is the moisture distribution after drying processing by the dryer for a predetermined time, as a dependent variable; The prediction information determination unit inputting the current moisture distribution into the predictive model to obtain the future moisture distribution; predicting a first operation end time relating to a time required for the moisture content of the object to be dried to reach a target moisture content based on the future moisture content distribution; Correcting the first driving end time based on the classified class to obtain a second driving end time, and determining the second driving end time as the driving end time of the prediction information. A prediction device configured as follows.

11. A dryer, The prediction device according to claim 1 or 2; an operating condition determination unit that determines operating conditions of the dryer or candidates for the operating conditions based on the prediction information determined by the prediction information determination unit; Equipped with Dryer.

12. 1. A method for predicting the outcome of a drying process performed by a dryer, comprising: Obtain the current moisture distribution of the object to be dried, assessing a characteristic of the current moisture distribution; determining a prediction regarding the outcome of the drying process based on the evaluation of the current moisture distribution characteristics; method.

Images