Information processing device, dehydrator, moisture content estimation method, and moisture content estimation program

Abstract

To maintain estimated accuracy of a water content of a dewatering cake without updating an estimation model.SOLUTION: An information processor (1) includes: an acquisition part (101) for acquiring measurement data relating to operation of a dehydrator in an actual operation period; an estimation part (102) for estimating an estimated value of a water content of a dewatering cake discharged from the dehydrator at a measurement time point of the measurement data, using an estimation model (112); and a correction part (104) for correcting the estimated value using an off-set value. The estimation model (112) is an estimation model learned by using the measurement data in an operation period in which the operation condition is different from that in the actual operation period as an explanatory variable, and an actually measured value of the water content at the measurement time point as an objective variable. The off-set value is a difference between the estimated value of the water content estimated from the measurement data in the actual operation period within a past predetermined period, and the actually measured value of the water content at the measurement time point.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a technique for estimating the moisture content of a dehydrated cake obtained by dehydrating a liquid containing suspended solids in a dehydrator. [Background technology]

[0002] Sludge treatment carried out in wastewater treatment facilities such as sewage treatment plants includes a step of dewatering the sludge using a dehydrator. For efficient sludge treatment, it is important to maintain the moisture content of the dehydrated cake obtained by dehydration within a predetermined range. However, when dehydration treatment is performed under constant operating conditions of the dehydrator, the moisture content of the dehydrated cake fluctuates due to factors such as inconsistent properties of the supplied sludge, making it difficult to maintain the moisture content of the dehydrated cake within the predetermined range.

[0003] For this reason, development of technology for estimating the moisture content of dehydrated cake has been underway. If the current moisture content could be estimated in real time, it would be possible to maintain the moisture content within a predetermined range through various controls. For example, Patent Document 1 listed below discloses a technology for estimating the moisture content by generating a moisture content estimation model using multiple parameters, such as the amount of sludge supplied to a centrifugal dehydrator and values ​​related to the centrifugal effect of the dehydrator. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2020-114569 Summary of the Invention [Problem to be solved by the invention]

[0005] Generally, the sludge treatment equipment is test-run under a wide variety of operating conditions to determine the optimum operating conditions. During the test run, the moisture content estimation model is generated using a wide variety of measured values ​​of the multiple parameters measured under the wide variety of operating conditions as training data.

[0006] The apparatus is then operated under the determined operating conditions. During this operation, unless the moisture content estimation model is updated using the most recent measured values ​​of the parameters as training data, the accuracy of the estimation by the moisture content estimation model will decrease.

[0007] An object of one aspect of the present invention is to maintain the accuracy of estimating the moisture content of a dehydrated cake without updating the estimation model. [Means for solving the problem]

[0008] In order to solve the above-mentioned problems, an information processing device according to one aspect of the present invention includes an acquisition unit that acquires measurement data related to operation of a dehydrator that dehydrates a liquid discharged from a coagulation tank, to which a chemical agent for coagulating the suspended solids is added to the liquid containing the suspended solids, while transporting the liquid; an estimation unit that estimates an estimated value of the moisture content at the time of measurement of the measurement data acquired by the acquisition unit during the actual operation period using an estimation model that is trained using an explanatory variable of the measurement data acquired by the acquisition unit during an operation period whose operating conditions are different from those of an actual operation period of the dehydrator, and an actual value of the moisture content of dehydrated cake discharged from the dehydrator at the time of measurement of the measurement data; and a correction unit that corrects the estimated value using an offset value, the offset value being a combination of a past estimated value that is an estimated value of the moisture content estimated by the estimation unit using measurement data acquired by the acquisition unit during the actual operation period within a predetermined past period, and a correction value of the measurement data. This is the difference between the actual measured value of the moisture content at the time of measurement and the past measured value.

[0009] In addition, a moisture content prediction method according to another aspect of the present invention is a moisture content estimation method executed by one or more information processing devices, and includes: an acquisition step of acquiring measurement data related to the operation of a dehydrator that dehydrates a liquid discharged from a coagulation tank, to which an agent that coagulates the suspended solids is added to the liquid containing the suspended solids, while transporting the liquid; an estimation step of estimating an estimated value of the moisture content at the time of measurement of the measurement data acquired in the acquisition step during the actual operation period using an estimation model trained with the measurement data acquired in the acquisition step during an operation period whose operating conditions are different from those of the actual operation period of the dehydrator as explanatory variables and the actual moisture content of the dehydrated cake discharged from the dehydrator at the time of measurement of the measurement data as the objective variable; and a correction step of correcting the estimated value using an offset value, wherein the offset value is the difference between a past estimated value, which is an estimated value of the moisture content estimated in the estimation step using the measurement data acquired in the acquisition step during the actual operation period in a predetermined past period, and a past actual value, which is the actual measured value of the moisture content at the time of measurement of the measurement data. [Effects of the Invention]

[0010] According to one aspect of the present invention, the accuracy of estimating the moisture content of a dehydrated cake can be maintained without updating the estimation model. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram showing an example of a main configuration of an information processing device according to an embodiment of the present invention; [Figure 2] FIG. 10 is a diagram illustrating an example of the configuration of a control system including the information processing device. [Figure 3] 10 is a flowchart showing an example of a process for estimating the moisture content of a dehydrated cake in the information processing device. [Figure 4] 10 is a flowchart illustrating an example of an update process of an offset value in the information processing device. [Figure 5] FIG. 3 is a diagram showing, in a table format, combinations of explanatory variables and response variables of an estimation model in the first embodiment of the information processing device. [Figure 6] 10 is a graph showing the variation in estimated moisture content values ​​after correction using the estimation model and offset value in Example 1 relative to actual moisture content measurements made by operators. [Figure 7] 10 is a graph showing variations in estimated moisture content values ​​after correction using an estimation model and an offset value in Example 2 of the information processing device, relative to actual moisture content measurements made by operators. [Figure 8] 10 is a graph showing the variation in moisture content estimated by the comparative model in Comparative Examples 1, 2, and 3 relative to the moisture content actually measured by the operator. [Figure 9] FIG. 10 is a block diagram showing an example of a main configuration of an information processing device according to another embodiment of the present invention. [Figure 10] FIG. 2 is a model diagram illustrating a concept of creating training data for a prediction model in the information processing device. [Figure 11] 10 is a flowchart showing an example of a process for predicting the moisture content of a dehydrated cake in the information processing device. [Figure 12] 10 is a flowchart illustrating an example of a process of updating the prediction model in the information processing device. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, embodiments of the present invention will be described in detail. For the sake of convenience, the same reference numerals will be used to designate components having the same functions as those in the embodiments, and the description thereof will be omitted where appropriate.

[0013] [Embodiment 1] An embodiment of the present invention will be described with reference to FIGS.

[0014] [System Configuration] FIG. 2 is a diagram showing an example of the configuration of a control system 100 according to this embodiment. The control system 100 is a system used in a plant that adds an agent that coagulates suspended solids to a liquid to be treated in a coagulation tank to form flocs, and then performs solid-liquid separation of the liquid to be treated in which the flocs have formed. In the following, an example will be described in which the liquid to be treated is sludge, but the control system 100 can also be applied to plants that treat liquids other than sludge. Note that sludge is a liquid containing fine solids that is generated during wastewater treatment, etc., and can also be called slurry.

[0015] As will be explained in detail below, the control system 100 performs each process of the sludge treatment process, from the process of converting the sludge to be treated into flocculated sludge by flocculating solid suspended matter in the sludge to be treated to form flocs, to the process of dewatering the flocculated sludge to obtain dehydrated sludge (also called dehydrated cake) and dehydrated filtrate. As shown in Fig. 2, the control system 100 includes an information processing device 1, a control device 3, a flocculator 5, and a dehydrator 9.

[0016] The flocculator 5 is a device that adds a chemical agent that coagulates suspended solids to the liquid being treated in a coagulation tank and moderately stirs the liquid to form flocs. Specifically, the flocculator 5 uses sludge as the liquid being treated, coagulates suspended solids in the sludge to form flocs, and produces coagulated sludge. The flocculator 5 in FIG. 2 includes a coagulation tank 51, a stirring blade 52, a motor 53, and an inspection window 54. The flocculator 5 also includes a sludge inlet 55, a chemical agent inlet 56, and a discharge outlet 57.

[0017] Furthermore, a photographing device 72 and a lighting device 71 for photography are attached to the inspection window 54. The photographing device 72 may be any device capable of taking at least still images. It is preferable that the coagulation tank 51 is opaque so that the way light hits the flocs does not change while the control system 100 is in operation. Furthermore, it is preferable that the photographing device 72 and the lighting device 71 are housed in a light-blocking dark box with an opening on the inspection window 54 side, as in the illustrated example.

[0018] The dehydrator 9 is a device that performs solid-liquid separation of the liquid to be treated in which flocs have formed. Specifically, the dehydrator 9 is disposed downstream of the flocculator 5 and performs solid-liquid separation by dehydrating the coagulated sludge (liquid) discharged from the flocculator 5. The dehydrator 9 in FIG. 2 is a screw press type dehydrator equipped with an outer screen 91 and a screw 92. The dehydrator 9 is also provided with a sludge inlet 93, a filtrate outlet 94, and a dehydrated cake outlet 95. Although not shown, the dehydrator 9 also includes a motor for rotating the screw 92. Of course, the dehydrator 9 is not limited to the screw press type as long as it can dehydrate the coagulated sludge. For example, a centrifugal dehydrator, a filter press dehydrator, or a belt press dehydrator may also be used.

[0019] In the control system 100, the sludge to be treated is continuously or intermittently supplied from a sludge inlet 55 into the coagulation tank 51 of the flocculator 5 by a supply device (not shown). The supply rate of the sludge may be automatically controlled by the supply device or its control device 3 according to the sludge treatment rate by the flocculator 5 and the dehydrator 9.

[0020] Then, chemicals (including at least a flocculant) for flocculating the sludge are fed into the sludge in the coagulation tank 51 through chemical inlet 56. In this state, motor 53 is driven to rotate agitator blade 52, which mixes the sludge and chemicals and forms flocs. The flocculated sludge, which is a mixture of the formed flocs and the water contained in the sludge, is discharged from outlet 57.

[0021] Subsequently, this flocculated sludge is supplied into the outer body screen 91 from the sludge inlet 93 of the dehydrator 9. In the dehydrator 9, the flocculated sludge is dehydrated under pressure by the screw 92, and the filtrate is discharged from the filtrate outlet 94, while the dehydrated cake, which is a mass of dehydrated flocculated sludge, is discharged from the dehydrated cake outlet 95.

[0022] The sludge is supplied into the coagulation tank 51 from the sludge inlet 55, whereby the coagulated sludge is pushed out and discharged from the outlet 57 of the coagulation tank 51, and the discharged coagulated sludge is supplied to the dehydrator 9. Therefore, the flow rate of the sludge supplied to the coagulation tank 51 and the flow rate of the sludge supplied to the dehydrator 9 coincide with each other at the same time.

[0023] As will be described in detail below, the information processing device 1 acquires measurement data relating to the operation of the dehydrator 9. Then, the information processing device 1 estimates the moisture content of the dehydrated cake based on the acquired measurement data.

[0024] The information processing device 1 can also control the operation of various devices (for example, the flocculator 5, the dehydrator 9, and a sludge and chemical supply device (not shown)) that are components of the control system 100 via the control device 3. The control device 3 is a device that controls the operation of various devices that are components of the control system 100. The control device 3 may be, for example, a PLC (Programmable Logic Controller). In the example of FIG. 2, the control device 3 controls the operation of the flocculator 5, the dehydrator 9, and a sludge and chemical supply device (not shown). The flocculator control device 3a controls the equipment related to the flocculator 5, and the dehydrator control device 3b controls the equipment related to the dehydrator 9.

[0025] [Device configuration] The configuration of the information processing device 1 will be described with reference to Fig. 1. Fig. 1 is a block diagram showing an example of the configuration of the main parts of the information processing device 1. As shown in the figure, the information processing device 1 includes a control unit 10 that controls each unit of the information processing device 1 in an integrated manner, and a storage unit 11 that stores various data used by the information processing device 1. The information processing device 1 also includes a communication unit 12 that enables the information processing device 1 to communicate with other devices, an input unit 13 that accepts input of various data to the information processing device 1, and an output unit 14 that enables the information processing device 1 to output various data.

[0026] The control unit 10 also includes an acquisition unit 101, an estimation unit 102, an update unit 103, and a correction unit 104. The update unit 103 will be described later in the section "Regarding the Update Unit." The correction unit 104 will be described later in the section "Regarding the Correction Unit."

[0027] Memory unit 11 includes actual measurement file 111, estimation model 112, estimation file 113, and offset value 114. Actual measurement file 111 includes actual measured values ​​of the moisture content of the dehydrated cake measured by an operator. Estimation file 113 includes estimated values ​​of the moisture content of the dehydrated cake estimated by estimation unit 102. Details of estimation model 112 will be described later in the section "Regarding the Estimation Model." Details of offset value 114 will be described later in the section "Regarding the Update Unit."

[0028] The acquisition unit 101 acquires measurement data related to the operation of the dehydrator 9. Details of the measurement data will be described later in the section "Regarding Estimation Model." The acquisition unit 101 also acquires an actual measurement value of the moisture content of the dehydrated cake measured by an operator, and stores the actual measurement value in an actual measurement file 111 in the storage unit 11.

[0029] The estimation unit 102 estimates the moisture content of the dehydrated cake from the measurement data acquired by the acquisition unit 101, using the estimation model 112 stored in the storage unit 11. This dehydrated cake is the dehydrated cake discharged from the dehydrator 9 at the time when the measurement data is measured. The estimation unit 102 estimates the moisture content of the dehydrated cake from the measurement data acquired by the acquisition unit 101. The estimated value of the moisture content from the measurement data by the constant model 112 is stored in the estimation file 113 of the storage unit 11 and is also sent to the correction unit 104.

[0030] The estimation unit 102 performs estimation using the estimation model 112 every time the acquisition unit 101 acquires new measurement data. Therefore, the estimation unit 102 can estimate the moisture content of the dehydrated cake in real time, that is, continuously at short time intervals (for example, every minute).

[0031] [About the estimation model] The estimation model 112 is an estimation model trained using the measurement data related to the operation of the dehydrator 9 acquired by the acquisition unit 101 as an explanatory variable and the moisture content at the time of measurement of the measurement data as a response variable. In this embodiment, the estimation model 112 is trained during a trial operation period before the start of the actual operation period of the dehydrator 9. Various methods such as multivariate regression analysis and Random Forest Regressor can be used for the above training.

[0032] The measurement data is at least one of the operating time of the dehydrator 9, the rotation speed of the screw of the dehydrator 9, the drive current value or torque value of the screw, the back pressure by a back pressure plate that is provided at a dehydrated cake discharge outlet 95 (discharge section) of the dehydrator 9 and compresses the liquid, and the opening between the back pressure plate and the dehydrated cake discharge outlet 95, and is acquired from the dehydrator control device 3b. Among these, measurement data that have a large contribution (high importance) to the estimation accuracy of the estimation model 112 include the drive current value of the screw and the opening (particularly when the dehydrator 9 is controlled so that the back pressure is constant).

[0033] [About the update section] The update unit 103 updates the offset value 114 in the storage unit 11 every time a predetermined time (e.g., one day) has elapsed. The offset value 114 is the difference between a past estimated value, which is an estimated value of the moisture content estimated by the estimation unit 102 using the measurement data acquired by the acquisition unit 101 during an actual operation period within a past predetermined period (e.g., the most recent two weeks), and a past actual measured value, which is the actual measured value of the moisture content at the time of measurement of the measurement data.

[0034] Specifically, the update unit 103 reads out the past estimated value estimated using the measurement data during the actual operation period from the estimation file 113 in the storage unit 11, and also reads out the past actual measurement value at the time of measurement of the measurement data from the actual measurement file 111 in the storage unit 11. Then, the update unit 103 updates the offset value 114 in the storage unit 11 by using the difference between the read past estimated value and the past actual measurement value as a new offset value.

[0035] If there are multiple pairs of the past estimated value and the past measured value, the new offset value may be a statistical value such as the average, maximum, minimum, or median of the difference for each pair. Alternatively, the new offset value may be the difference between the past estimated value and the past measured value when the end point of the predetermined period is set as the update point. By using such a most recent offset value, the accuracy of the estimated value corrected by the estimation model 112 and the offset value 114 can be reliably maintained.

[0036] [Regarding the correction section] The correction unit 104 corrects the estimated value from the estimation unit 102 by using the offset value 114 in the storage unit 11. Specifically, the correction unit 104 sets the value obtained by adding or subtracting the offset value to or from the estimated value as the correction value.

[0037] Similar to the estimation unit 102, the correction unit 104 performs the above correction each time the acquisition unit 101 acquires new measurement data. Therefore, the correction unit 104 can correct the moisture content of the dehydrated cake in real time, that is, continuously at short time intervals (for example, every minute).

[0038] As described above, the information processing device 1 of this embodiment includes an acquisition unit 101 that acquires measurement data related to the operation of the dehydrator 9 that dehydrates the liquid discharged from the coagulation tank 51, to which an agent that coagulates the suspended solids is added to the liquid containing the suspended solids, while transporting the liquid; an estimation unit 102 that estimates, using an estimation model 112, an estimate of the moisture content at the time of measurement of the measurement data acquired by the acquisition unit 101 during the actual operation period of the dehydrator 9; and a correction unit 104 that corrects the estimate using an offset value. The estimation model 112 is an estimation model trained using the measurement data acquired by the acquisition unit 101 during a trial operation period, which is an operation period under different operating conditions from the actual operation period, as explanatory variables, and the actual moisture content of the dehydrated cake discharged from the dehydrator 9 at the time of measurement of the measurement data as a response variable. The offset value is the difference between the past estimated value, which is the estimated value of the moisture content estimated by the estimation unit 102 using the measurement data acquired by the acquisition unit 101 during the actual operating period within a specified period in the past, and the past actual measured value, which is the actual measured value of the moisture content at the time of measurement of the measurement data.

[0039] However, when the accuracy of estimation by the estimation model 112 decreases, the difference between the estimated moisture content and the measured moisture content increases. As a result of examining this for each predetermined period in the past, the inventors of the present application found that this difference is approximately constant.

[0040] Therefore, with the above configuration, an offset value is calculated in advance as the difference between a past estimated value, which is an estimated value of the moisture content estimated by the estimation unit 102 using measurement data from an actual operating period within a predetermined past time period, and a past actual measured value, which is the actual measured value of the moisture content at the time the measurement data was measured. Next, the estimation unit 102 estimates the estimated value of the moisture content at the time the measurement data was measured during the actual operating period using the estimation model 112. Then, the correction unit 104 corrects the estimated value using the offset value. The corrected estimated value is closer to the actual measured value of the moisture content at the time of measurement. Therefore, by calculating the offset value, the accuracy of the moisture content estimation can be maintained without updating the estimation model 112.

[0041] [Estimation process] 3 is a flowchart showing an example of a process for estimating the moisture content of dehydrated cake (a moisture content estimation method) in the information processing device 1 having the above configuration. As described above, this estimation process is performed during actual operation. As shown in FIG. 3, first, the acquisition unit 101 collects (acquires) various measurement data (S11, acquisition step). When the acquisition unit 101 acquires a new actual measurement value of the moisture content measured by an operator (S12), the acquisition unit 101 stores the actual measurement value in the actual measurement file 111 of the storage unit 11 (S13).

[0042] Next, the estimation unit 102 calculates an estimated value of the moisture content at the time of measurement of the measurement data from the measurement data using the estimation model 112 in the storage unit 11 (S14, estimation step). At this time, the estimation unit 102 stores the estimated value in the estimation file 113 in the storage unit 11 (S15).

[0043] Then, the correction unit 104 corrects the estimated value of the moisture content using the offset value 114 in the storage unit 11 (S16, correction step). After that, the process returns to step S11 and repeats the above operations.

[0044] [Update process] 4 is a flowchart showing an example of the update process of the offset value 114 in the information processing device 1. As described above, this update process is executed every time a predetermined time period elapses.

[0045] As shown in FIG. 4, first, the update unit 103 sets the estimated value of the moisture content of the dehydrated cake, which is estimated using the measurement data during the actual operation period within a predetermined period in the past, as the past estimated value, and stores it in the storage unit 1. The update unit 103 then reads the actual measurement value of the moisture content at the time of measurement of the measurement data from the estimated file 113 of the storage unit 11 as a past actual measurement value (S22). Note that the order of execution of steps S21 and S22 is arbitrary.

[0046] Next, the update unit 103 calculates the difference between the read previous estimated value and the previous measured value as a new offset value (S23). Then, the update unit 103 updates the offset value 114 in the storage unit 11 with the calculated new offset value (S24). Thereafter, the update process ends.

[0047] [Modification] The execution entity of each process described in the above embodiment is arbitrary and is not limited to the above example. For example, each step of the moisture content estimation method shown in Fig. 3 can be shared among multiple information processing devices. In other words, the moisture content estimation method may be executed by one information processing device 1 or multiple information processing devices.

[0048] [Additional Notes] In the above embodiment, the estimation model 112 is not updated, but this is not limiting. For example, measurement data acquired by the acquisition unit 101 when an operating condition different from normal occurs during actual operation may be accumulated, and the estimation model 112 may be updated using the accumulated measurement data.

[0049] In the above-described embodiment, the trial operation period is used as an operation period in which the operating conditions are different from those of the actual operation period. However, the different operation period is not limited to the trial operation period, and may be any operation period in which measurement data can be obtained under a wide variety of operating conditions, including the range of operating conditions in the actual operation period. For example, the different operation period may be a test operation period performed after maintenance of the dehydrator 9.

[0050] Furthermore, in the above-described embodiment, the acquisition unit 101 needs to acquire measurement data related to the dehydrator 9, but does not need to acquire measurement data related to the flocculator 5. Therefore, any dehydrator 9 equipped with the information processing device 1 having the above-described configuration can achieve the above-described advantageous effects.

[0051] [Example] An example of the information processing device 1 having the above configuration and a comparative example will be described with reference to FIGS.

[0052] Example 1 In Example 1, the estimation unit 102 calculated an estimated value of the moisture content at the time of measurement of the measurement data acquired by the acquisition unit 101 using the estimation model 112, and the correction unit 104 corrected the value by adding an offset value 114.

[0053] Fig. 5 is a diagram showing, in tabular form, combinations of explanatory variables and response variables of estimation model 112 in Example 1. In the example of Fig. 5, the measurement data serving as explanatory variables of estimation model 112 are the operating time of dehydrator 9, the rotation speed of the screw of dehydrator 9, the drive current value (or drive torque) of the screw, the back pressure (back pressure of dehydrator 9) of a back pressure plate that is provided at dehydrated cake discharge outlet 95 of dehydrator 9 and compresses liquid, and the opening between the back pressure plate and dehydrated cake discharge outlet 95 (back pressure opening of dehydrator 9). The response variable of estimation model 112 is the actual measurement value of the moisture content of the dehydrated cake.

[0054] In Example 1, the estimation model 112 is an estimation model trained using, as training data, a combination of measurement data corresponding to the explanatory variables among measurement data obtained under various operating conditions during a test run period (3 months) and the actual measurement value of the moisture content, which is the objective variable. Furthermore, in Example 1, the estimation model 112 was not updated with new training data.

[0055] In addition, in Example 1, the update unit 103 calculates a new offset value using a combination of an estimated moisture content estimated using the estimation model 112 from measurement data for the most recent predetermined period and an actual moisture content value at the time of measurement of the measurement data, and updates the offset value 114 in the storage unit 11. In Example 1, the most recent predetermined period is eight days, and the frequency with which the worker actually measures the moisture content is once per day. Therefore, the number of combinations is (one set per day) × 8 days = 8 sets. For each combination, (the actual moisture content) - (the estimated moisture content) is calculated, and the average of the calculated eight sets of values ​​is set as the new offset value.

[0056] Fig. 6 is a graph showing the variation of the moisture content estimates after correction in Example 1 relative to the actual moisture content measurements made by operators. The graph in Fig. 6 uses measurement data and actual moisture content measurements from a seven-month actual operation period after training the estimation model 112. In the graph shown in Fig. 6, the mean absolute error (MAE) was 0.80%.

[0057] Example 2 Example 2 is different from Example 1 in that the most recent predetermined period is two weeks, and the frequency with which the worker measures the moisture content is once per week, but the rest is the same. Therefore, the number of combinations used by the update unit 103 is (1 set per week) x 2 weeks = 2 sets.

[0058] FIG. 7 is a graph showing the variation of the corrected moisture content estimates in Example 2 relative to the actual moisture content measurements by operator. The graph in FIG. 7 uses the same measurement data and actual moisture content measurements as in Example 1. The graph shown in FIG. 7 shows a mean absolute error (MAE) of 0.87%. This shows that the accuracy of the corrected moisture content estimates can be maintained even if the number of times an operator measures the moisture content is reduced from one set per day to one set per week.

[0059] (Comparative Example 1) In Comparative Example 1, an estimated value of the moisture content at the time of measurement of the measurement data was estimated from the measurement data using a comparison model trained in the same way as estimation model 112. In other words, no correction was made using offset value 114. Also, in Comparative Example 1, the comparison model was not updated using new training data.

[0060] The upper graph in Figure 8 shows the variation in moisture content estimates in Comparative Example 1 relative to the actual moisture content measured by operators. As with Example 1, all of the graphs in Figure 8 utilize measurement data and actual moisture content measurements from a seven-month operational period after the comparative model was trained. In the upper graph in Figure 8, the MAE was 1.95%. This indicates that the moisture content estimates in Comparative Example 1 are significantly worse than those in Examples 1 and 2.

[0061] (Comparative Example 2) Comparative Example 2 is similar to Comparative Example 1 except that the comparison model is updated with new training data during the shutdown period. The number of training data items was (6 per day) x 8 days = 48.

[0062] The middle part of Fig. 8 shows the estimated moisture content in Comparative Example 2 compared to the actual moisture content measured by the operator. 8 is a graph showing the variation. In the middle graph of Fig. 8, the MAE was 0.80%. From this, it can be understood that while the estimated moisture content in Comparative Example 2 was similar to that in Examples 1 and 2, the number of times that workers actually measured the moisture content was greater than in Examples 1 and 2, which resulted in an increased burden on workers and an increase in costs due to the increased number of workers.

[0063] (Comparative Example 3) Comparative Example 3 is similar to Comparative Example 2 except that the number of training data is (1 per day) x 8 days = 8. That is, the number of training data in Comparative Example 3 is approximately the same as the number of combinations for updating the offset value 114 in Example 1.

[0064] The lower graph in Fig. 8 is a graph showing the variation in the moisture content estimates in Comparative Example 3 relative to the actual moisture content measurements made by the operators. In the graph in the lower graph in Fig. 8, the MAE was 1.41%. This shows that the moisture content estimates in Comparative Example 3 were worse than those in Examples 1 and 2. These comparison results show that the corrected moisture content estimates in Examples 1 and 2 have good accuracy and can reduce the burden on the operators.

[0065] [Embodiment 2] Another embodiment of the present invention will be described with reference to Figures 9 to 12. A control system 100 of this embodiment is different from the control system 100 shown in Figures 1 to 4 in the configuration of the information processing device 1, but the other configurations are the same.

[0066] Fig. 9 is a block diagram showing an example of the configuration of the main parts of an information processing device 1 according to this embodiment. The information processing device 1 shown in Fig. 9 differs from the information processing device 1 shown in Fig. 1 in that the control unit 10 includes an acquisition unit 105 and a correction unit 106 instead of the acquisition unit 101 and the correction unit 104, and also newly includes a model update unit 107 and a prediction unit 108, and the storage unit 11 further includes a prediction model 115. The other configurations are the same. Details of the prediction model 115 will be described later in the section "Regarding the Prediction Model."

[0067] The acquiring unit 105 acquires at least one of measurement data related to the liquid supplied to the coagulation tank 51, measurement data related to the chemicals supplied to the coagulation tank 51, measurement data related to the liquid in the coagulation tank 51, measurement data related to the operation of the coagulation tank 51, and measurement data related to the operation of the dehydrator 9. Details of each measurement data will be explained later in the section "Regarding the prediction model."

[0068] Correction unit 106 differs from correction unit 104 shown in FIG. 1 in that correction unit 106 further sends the corrected estimated value of the moisture content of the dehydrated cake to model update unit 107, but the other configurations are the same.

[0069] Prediction unit 108 uses prediction model 115 stored in memory unit 11 to predict the moisture content of the dehydrated cake from the measurement data acquired by acquisition unit 105. This dehydrated cake is the dehydrated cake that is discharged from dehydrator 9 at the point in time when the retention time of the flocculated sludge in dehydrator 9 has elapsed from the time of measurement of the measurement data (hereinafter referred to as the "elapsed time point"). Prediction unit 108 performs a prediction using prediction model 115 every time acquisition unit 105 acquires new measurement data. This allows prediction unit 108 to predict the moisture content of the dehydrated cake in real time, that is, continuously at short time intervals (for example, every minute). Note that a specific method for calculating the moisture content of the dehydrated cake will be described later in the section "Method for predicting moisture content using a prediction model."

[0070] [About the prediction model] The prediction model 115 uses the measurement data at the time of measurement acquired by the acquisition unit 105 as explanatory variables. , and is a prediction model trained using the moisture content of the dehydrated cake discharged from the dehydrator 9 at an elapsed time point (hereinafter referred to as "moisture content at the elapsed time point") as the objective variable. In this embodiment, the moisture content, which is the objective variable, is an estimated value after correction by the estimation model 112, the estimation unit 102, and the correction unit 106. The measurement data acquired by the acquisition unit 105 and used for prediction by the prediction model 115 includes the following:

[0071] (1) Measurement data regarding the sludge supplied to the coagulation tank 51. The measurement data is, for example, at least one of the supply flow rate of the sludge per unit time and the concentration of the sludge, and is measured before the sludge inlet 55.

[0072] (2) Measurement data relating to the chemical agent supplied to the coagulation tank 51. The measurement data is, for example, the supply flow rate of the chemical agent per unit time, and is measured before the chemical agent inlet 56.

[0073] (3) Measurement data on the sludge in the coagulation tank 51. The measurement data is at least one of the average gray value of the flocs and the average unit area of ​​the gaps between the flocs, and is obtained by image processing of the still images taken by the image capture device 72. The average gray value serves as an index of the color tone (lightness) of the sludge. The average unit area serves as an index of the floc diameter.

[0074] (4) Measurement data related to the operation of the coagulation tank 51. The measurement data is the rotation speed of the stirring blades in the coagulation tank 51, and is acquired from the control device 3.

[0075] (5) Measurement data related to the operation of the dehydrator 9. The measurement data is at least one of the operation time of the dehydrator 9, the rotation speed of the screw of the dehydrator 9, the flow rate of the coagulated sludge supplied to the dehydrator 9 per unit time, and the feeding pressure of the coagulated sludge fed to the dehydrator 9, and is acquired from the control device 3.

[0076] In this embodiment, various measurement data can be employed as shown in (1) to (5) above. Therefore, the moisture content of the dehydrated cake can be accurately predicted using multiple explanatory variables. Among the measurement data shown in (1) to (5) above, the measurement data that contribute significantly to the prediction accuracy of the prediction model 115 include the operation time, the screw rotation speed, and the supply flow rate shown in (5) above, and the average floc density value shown in (3) above.

[0077] [Prediction model training method] Next, we will explain the method of learning the moisture content using the prediction model 115. As described above, the prediction model 115 is a model that uses measurement data at the time of measurement to predict the moisture content at a time point in the future from the time of measurement.

[0078] Fig. 10 is a model diagram showing the concept of creating training data for the prediction model 115. The horizontal axis of Fig. 10 represents driving time (minutes). The vertical axis of Fig. 10 represents the measurement value of each measurement data, and each measurement data is shown on an arbitrary scale.

[0079] As shown in the upper part of FIG. 10, the measurement data at the time of measurement, which serve as explanatory variables for the prediction model 115, are the supply flow rate per unit time of the liquid (sludge), the concentration of the liquid (sludge), the average density value of the flocs, the average unit area of ​​the gaps between the flocs, the rotation speed of the agitator blades in the coagulation tank 51, the operation time of the dehydrator 9, the rotation speed of the screw of the dehydrator 9, the supply pressure of the liquid (coagulated sludge) fed into the dehydrator 9, and the ratio of the supply flow rate per unit time of the chemical agent to the supply flow rate of the liquid (sludge).

[0080] The objective variable of the prediction model 115 is the moisture content at the time point. When using actual measurements by workers, there is an upper limit to the number of measurements a worker can make in a day, as shown by the white circles in the lower part of Figure 10. This limits the amount of training data, which is a combination of explanatory variables and target variables, making it difficult to create a highly accurate prediction model.

[0081] Therefore, in this embodiment, instead of actual measurements by an operator, estimated values ​​by the estimation model 112 are used as the dependent variables of the training data. As a result, as shown by the gray circles in the lower part of Fig. 10, the number of dependent variables is no longer limited by the number of measurements that an operator can make, so that the amount of training data can be increased and a highly accurate prediction model 115 can be created.

[0082] Therefore, in this embodiment, instead of actual measurements by an operator, estimated values ​​corrected by the estimation model 112, the estimation unit 102, and the correction unit 106 are used as the dependent variables of the training data. As a result, as shown by the gray circles in the lower part of Fig. 10, the number of dependent variables is no longer limited by the number of measurements that an operator can make, so that the amount of training data can be increased and a highly accurate prediction model 115 can be created.

[0083] The estimated value of the moisture content at the elapsed time point is calculated from the measurement data at the elapsed time point using the estimation model 112. Therefore, the measurement data at the measurement time point, which is the explanatory variable of the prediction model 115, and the measurement data at the elapsed time point, which is the explanatory variable of the estimation model 112, are used to train the prediction model 115.

[0084] [Method for predicting moisture content using a prediction model] Prediction unit 108 predicts the moisture content of the dehydrated cake using the trained prediction model 115. That is, prediction unit 108 inputs at least one of the explanatory variables, measurement data on the liquid supplied to coagulation tank 51, measurement data on the chemicals supplied to coagulation tank 51, measurement data on the liquid in coagulation tank 51, measurement data on the operation of coagulation tank 51, and measurement data on the operation of dehydrator 9, into prediction model 115, thereby predicting the objective variable, the moisture content of the dehydrated cake at the completion of dehydration.

[0085] 1 , information processing device 1 of the present embodiment further includes prediction unit 108 that uses prediction model 115 to predict, from the measurement data acquired by acquisition unit 105, the moisture content of the dehydrated cake discharged from dehydrator 9 at a time point after the retention time during which the liquid remains in dehydrator 9 has elapsed from the time point at which the measurement data was acquired. Prediction model 115 uses the measurement data acquired by acquisition unit 105 as an explanatory variable, and is trained with the moisture content of the dehydrated cake discharged from dehydrator 9 at a time point after the retention time has elapsed from the time point at which the measurement data was acquired, estimated using estimation model 112, and corrected using offset value 114, as a dependent variable.

[0086] According to the above configuration, the estimation model 112 measures the explanatory variable, that is, the measurement time (at the completion of dehydration) of the measurement data related to the operation of the dehydrator 9, and the objective variable, that is, the moisture content of the dehydrated cake discharged from the dehydrator 9 (at the completion of dehydration). Therefore, even if there is a small amount of training data including the measurement data and the moisture content, the accuracy of the corrected estimate using the trained estimation model 112 is good. Therefore, the estimation model 112 can be trained without increasing the burden on the operator in manually analyzing the moisture content. Furthermore, the estimation model 112 can be used to calculate the corrected estimate of the moisture content for the number of times the measurement data is measured.

[0087] On the other hand, the prediction model 115 is estimated using the estimation model 112 from the measurement data related to the operation of the dehydrator 9 measured at the time when the retention time has elapsed from the time when the liquid is supplied (when the dehydration is completed) of the measurement data acquired by the acquisition unit 105, and the offset value 114 is used. It is possible to obtain a corrected estimate of the moisture content at an elapsed time point, which is corrected using the obtained corrected estimate value. Then, it is possible to create teacher data, in which the obtained corrected estimate value is used as the objective variable and the measurement data at the measurement time point obtained by the obtaining unit 105 is used as the explanatory variable, for the number of measurements of the measurement data obtained by the obtaining unit 105. Therefore, it is possible to use the prediction model 115 trained using the teacher data to accurately predict the moisture content at an elapsed time point when the retention time has elapsed from the measurement time point of the measurement data obtained by the obtaining unit 105.

[0088] [About the update section] The model update unit 107 performs a process of updating the prediction model 115 during the operation shutdown period. Specifically, the model update unit 107 updates the prediction model 115 during the operation shutdown period using, as training data, a set of measurement data acquired by the acquisition unit 101 during the operation period of the flocculation tank 51 and the dehydrator 9, and an estimated value corrected by the estimation model 112 and the offset value 114 regarding the moisture content of the dehydrated cake discharged from the dehydrator 9 at the time when the retention time has elapsed since the time when the measurement data was measured.

[0089] It is desirable to use measurement data from the most recent operating period to update the prediction model 115. In this case, predictions using the prediction model 115 can be adapted to the most recent conditions of the coagulation tank 51 and the dehydrator 9.

[0090] [Prediction processing] FIG. 11 is a flowchart showing an example of a process for predicting the moisture content of dehydrated cake (a moisture content prediction method) in the information processing device 1 configured as described above. This prediction process is executed during operation, as described above. As shown in FIG. 11, first, the acquisition unit 105 collects (acquires) various measurement data (S31, acquisition step). Next, the prediction unit 108 uses the prediction model 115 to calculate a predicted value of the moisture content from the measurement data at a point in time when a residence time has elapsed since the measurement of the measurement data (S32, prediction step). Thereafter, the process returns to step S31 and repeats the above operation.

[0091] [Model update process] 12 is a flowchart showing an example of the model update process of the prediction model 115 in the information processing device 1. As described above, this update process is executed for each shutdown period.

[0092] 12, the model update unit 107 calculates a corrected estimate of the moisture content during the operating period from the measurement data during the operating period using the estimation model 112 and the offset value 114 (S41). Next, the model update unit 107 updates the prediction model 115 using as training data a pair of the measurement data and the corrected estimate at a point in time when a retention time has elapsed since the measurement of the measurement data (S42). Thereafter, the model update process ends.

[0093] [Software implementation example] The functions of the information processing device 1 (hereinafter referred to as the "device") can be realized by a program for causing a computer to function as the device, and a program (moisture content estimation program) for causing a computer to function as each control block of the device (particularly each part included in the control unit 10).

[0094] In this case, the device includes a computer having at least one control device (e.g., a processor) and at least one storage device (e.g., a memory) as hardware for executing the program. The control device and storage device execute the program, thereby realizing the functions described in each of the above embodiments.

[0095] The program may be non-transitory and may be recorded on one or more computer-readable recording media. The recording media may or may not be included in the device. In the latter case, the program may be supplied to the device via any wired or wireless transmission medium.

[0096] Furthermore, some or all of the functions of the control blocks can be realized by logic circuits. For example, an integrated circuit in which a logic circuit that functions as each of the control blocks is formed is also included in the scope of the present invention. In addition, the functions of the control blocks can also be realized by, for example, a quantum computer.

[0097] The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. Embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present invention. [Explanation of symbols]

[0098] 1. Information processing equipment 3. Control device 5 Flocculator 9 Dehydrator 10 Control Unit 11 Storage section 12 Communications Department 13 Input section 14 Output section 51 Coagulation tank 91 Outer body screen 92 Screw 94 Liquid outlet 95 Dehydrated cake outlet 100 Control System 101, 105 Acquisition Department 102 Estimation part 103 Update Department 104, 106 Correction unit 107 Model Update Department 108 Prediction Department 111 Measurement File 112 Estimation Model 113 Estimate File 114 Offset Value 115 Predictive Models

Claims

1. an acquisition unit that acquires measurement data related to the operation of a dehydrator that dehydrates a liquid discharged from a coagulation tank in which a chemical agent for coagulating the suspended solids is added to the liquid containing the suspended solids while transporting the liquid; an estimation unit that estimates an estimated value of the moisture content at the time of measurement of the measurement data acquired by the acquisition unit during the actual operation period, using an estimation model that is trained using the measurement data acquired by the acquisition unit during an operation period under operating conditions different from those of the actual operation period of the dehydrator as explanatory variables and an actual measurement value of the moisture content of a dehydrated cake discharged from the dehydrator at the time of measurement of the measurement data as a response variable; a correction unit that corrects the estimated value using an offset value, the offset value is a difference between a past estimated value, which is an estimated value of the moisture content estimated by the estimation unit using measurement data acquired by the acquisition unit during the actual operation period within a predetermined past period, and a past actual measured value, which is an actual measured value of the moisture content at the time of measurement of the measurement data, a prediction unit that predicts, using a prediction model, from the measurement data acquired by the acquisition unit, the moisture content of a dehydrated cake that is discharged from the dehydrator at a time when a retention time for the liquid to remain in the dehydrator has elapsed from the time when the measurement data was acquired, The prediction model is an information processing device in which the measurement data acquired by the acquisition unit is used as an explanatory variable, and the moisture content at the time when the residence time has elapsed from the time of measurement of the measurement data is estimated using the estimation model, and the estimated value corrected using the offset value is learned as the objective variable.

2. The information processing apparatus according to claim 1 , wherein the offset value is an average value of the differences for each pair of the past estimated value and the past measured value.

3. an updating unit that updates the offset value every time a predetermined time elapses; The information processing apparatus according to claim 1 , wherein the update unit sets a difference between the past estimated value and the past measured value when an end point of the predetermined period is set as an update point as the new offset value.

4. the dehydrator is a screw press type dehydrator, 2. The information processing device according to claim 1, wherein the measurement data is at least one of the operating time of the dehydrator, the rotation speed of the screw of the dehydrator, the drive current value or torque value of the screw, the back pressure caused by a back pressure plate provided in the discharge section of the dehydrator to compress the liquid, and the opening degree between the back pressure plate and the discharge section.

5. The information processing apparatus according to claim 1 , wherein the estimation unit and the correction unit respectively perform the estimation and the correction each time the acquisition unit acquires new measurement data.

6. A dehydrator comprising the information processing device according to any one of claims 1 to 5.

7. A moisture content estimation method executed by one or more information processing devices, comprising: an acquiring step of acquiring measurement data related to the operation of a dehydrator that dehydrates while transporting a liquid discharged from a coagulation tank in which a chemical agent for coagulating the suspended solids is added to the liquid containing the suspended solids; an estimation step of estimating an estimated value of the moisture content at the time of measurement of the measurement data acquired in the acquisition step during the actual operation period, using an estimation model trained with the measurement data acquired in the acquisition step during an operation period under operating conditions different from those of the actual operation period of the dehydrator as explanatory variables and the actual measured value of the moisture content of the dehydrated cake discharged from the dehydrator at the time of measurement of the measurement data as a response variable; a correcting step of correcting the estimated value using an offset value; the offset value is a difference between a past estimated value, which is an estimated value of the moisture content estimated in the estimation step using the measurement data acquired in the acquisition step during the actual operation period in a predetermined past period, and a past measured value, which is an actual measured value of the moisture content at the time of measurement of the measurement data, a prediction step of predicting, based on the measurement data acquired in the acquisition step, a moisture content of a dehydrated cake discharged from the dehydrator at a time when a residence time during which the liquid remains in the dehydrator has elapsed from the time of measurement of the measurement data, using a prediction model; A moisture content estimation method in which the prediction model uses the measurement data acquired in the acquisition step as an explanatory variable, estimates the moisture content at the time when the residence time has elapsed from the time of measurement of the measurement data using the estimation model, and is a prediction model learned using the estimated value corrected using the offset value as the objective variable.

8. A moisture content estimation program for causing a computer to function as the information processing device according to claim 1, the moisture content estimation program causing a computer to function as the acquisition unit, the estimation unit, the correction unit, and the prediction unit.

Images