System for determining operating conditions of a dewatering device and method for operating a dewatering device

Abstract

To provide a driving condition decision system capable of deciding an optimal driving parameter of a screw press.SOLUTION: A driving condition decision system 3 is assembled with: a candidate formation part 62 for formation of a plurality of candidates of a driving parameter of a dehydrator 1; a state prediction part 63 wherein each of the plurality of candidate is input in a state prediction model and a plurality of state prediction indicators are output from the plurality of state prediction models; and a select command part 64 where a candidate satisfying all of a plurality of target conditions predetermined to the plurality of state prediction indicators are selected and a driving parameter constituting the selected candidate is sent to a drive control part 2 of the dehydrator 1.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to the operation control of a dehydration device equipped with a screw press, and particularly to a technology for automatically determining optimal operating conditions and achieving stable operation of the dehydration device.

Background Art

[0002] Conventionally, a dehydration device that presses a suspension (e.g., sludge) discharged from a liquid treatment facility such as a sewage treatment plant, a night soil treatment plant, or an industrial wastewater treatment plant to separate water from the suspension (i.e., dehydrate) has been used. In the dehydration process, the suspension injected with a flocculant is stirred in a coagulation tank to form coagulation flocs and make it easier to perform solid-liquid separation, and then dehydration is performed by a concentration device or a dehydrator.

[0003] A screw press, which is an example of a dehydrator, is known as a sludge dehydrator. This screw press includes a filter cylinder formed from a screen (perforated plate) and a screw disposed inside the filter cylinder. By rotating the screw, the sludge introduced into the filter cylinder is pressed and dehydrated.

[0004] Cake (dehydrated sludge) is retained at the downstream opening end of the filter cylinder to form a plug (stopper) made of the cake. This plug applies back pressure to the cake fed later to further press the cake. The cake forming the plug is pushed by the subsequent cake and discharged little by little from the filter cylinder. In this way, a cake with a low moisture content is generated by the screw press.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Patent Document 2

Patent Document 3

[0006] Various proposals have been made to optimize the operating parameters of screw presses. For example, Patent Document 1 discloses a dewatering system that calculates a predicted state value of turbidity residue from suspension data and operating data. This dewatering system determines operating parameters that allow the predicted state value to fall within a target range. However, since the dewatering device is controlled based on the state of the turbidity residue, the state of the filtrate is not considered when determining the operating parameters.

[0007] Patent Document 2 discloses a machine learning device for a centrifugal separation system. This machine learning device learns a dataset of input data including the slurry concentration of the liquid to be processed, the water content of the dewatered solids, the concentration of the separated liquid, and the torque value of the screw conveyor, and output data including control parameters of the centrifugal separation system (one or more of the following: additive supply amount, centrifugal force, and differential velocity). However, the input data (explanatory variables) includes the water content of the dewatered solids and the concentration of the separated liquid, which need to be measured or estimated.

[0008] This invention has been made in view of the above-mentioned conventional problems, and its objective is to provide an operating condition determination system and operating method that can determine the optimal operating parameters of a screw press. [Means for solving the problem]

[0009] In one embodiment, a system for determining operating conditions for a dewatering apparatus is provided, comprising a coagulation mixing tank for injecting a coagulant into sludge and a screw press for separating the sludge into a dewatered cake and filtrate, the system comprising: a candidate generation unit that generates a plurality of candidate operating parameters including at least the injection rate of the coagulant into the sludge, the stirring speed of the sludge and the coagulant in the coagulation mixing tank, and the screw rotation speed of the screw press; a state prediction unit that inputs each of the plurality of candidates into a plurality of state prediction models and outputs a plurality of state prediction indices from each of the plurality of state prediction models; and a selection command unit that selects a candidate from the plurality of candidates that satisfies all of a plurality of predetermined target conditions for the plurality of state prediction indices, and sends the operating parameters constituting the selected candidate to the operation control unit of the dewatering apparatus.

[0010] In one embodiment, the plurality of state prediction models is characterized by including at least two of the following: an operation prediction model that outputs a state prediction index indicating the operation prediction result of the screw press; a moisture content prediction model that outputs a state prediction index indicating the predicted moisture content of the dewatered cake discharged from the screw press; and an SS recovery rate prediction model that outputs a state prediction index indicating the predicted SS recovery rate of the screw press. In one embodiment, the state prediction index output from the operation prediction model is characterized by being at least one of the following: a predicted value or predicted rate of change of the sludge level in the sludge inlet of the screw press, a predicted torque of the screw of the screw press, and a predicted result of whether or not co-rotation occurs between the screw of the screw press and the dewatered cake. In one embodiment, the plurality of target conditions include a first target condition defining stable operation of the screw press, a second target condition specifying that the predicted moisture content is within the moisture content target range, and a third target condition specifying that the predicted SS recovery rate is within the SS recovery rate target range.

[0011] In one embodiment, the selection command unit is configured to select a candidate that satisfies all of the plurality of target conditions and also satisfies predetermined constraint conditions regarding the operating parameters. In one embodiment, the operating parameters further include the amount of sludge to be processed by the screw press, and the predetermined constraints are conditions that restrict the correlation between the amount of sludge processed and the screw rotation speed. In one embodiment, the predetermined constraint is a condition that restricts the correlation between the injection rate of the flocculant and the stirring speed.

[0012] In one embodiment, the operating condition determination system further comprises a stable operation determination unit that determines, based on determination indicators, whether the actual value of the sludge level in the sludge inlet of the screw press or the actual rate of change of the level is within a predetermined allowable range, whether the actual torque of the screw is within a target range, or whether the screw and the dewatered cake are actually rotating together in the screw press, wherein the determination indicators are at least one of the measured value of the sludge level in the sludge inlet of the screw press, the detected value of the torque of the screw, the measured value of the pressure applied to the sludge in the filter cylinder of the screw press, and the state of the dewatered cake discharged from the filter cylinder. In one embodiment, the operating parameters further include the concentration of the sludge. In one embodiment, the selection command unit is configured to select from among a plurality of candidates that satisfy the plurality of target conditions the candidate that is most suitable for a preset operating mode of the dewatering device. In one embodiment, the selection command unit is configured to assign a degree of deviation from a plurality of actual operating data included in the training data used for machine learning of the plurality of state prediction models to a plurality of candidates that satisfy the plurality of target conditions, and to select candidates whose degree of deviation is less than or equal to a predetermined standard.

[0013] In one embodiment, a method for operating a dewatering apparatus is provided, comprising a coagulation mixing tank for injecting a coagulant into sludge and a screw press for separating the sludge into a dewatered cake and filtrate, characterized in that a plurality of candidate operating parameters are generated, each including at least the injection rate of the coagulant into the sludge, the stirring speed of the sludge and the coagulant in the coagulation mixing tank, and the screw rotation speed of the screw press; each of the plurality of candidates is input into a plurality of state prediction models; a plurality of state prediction indicators are output from each of the plurality of state prediction models; a candidate that satisfies all of a plurality of predetermined target conditions for the plurality of state prediction indicators is selected from the plurality of candidates; and the dewatering apparatus is operated with the operating parameters constituting the selected candidate.

[0014] In one embodiment, the plurality of state prediction models is characterized by including at least two of the following: an operation prediction model that outputs a state prediction index indicating the operation prediction result of the screw press; a moisture content prediction model that outputs a state prediction index indicating the predicted moisture content of the dewatered cake discharged from the screw press; and an SS recovery rate prediction model that outputs a state prediction index indicating the predicted SS recovery rate of the screw press. In one embodiment, the state prediction index output from the operation prediction model is characterized by being at least one of the following: a predicted value or predicted rate of change of the sludge level in the sludge inlet of the screw press, a predicted torque of the screw of the screw press, and a predicted result of whether or not co-rotation occurs between the screw of the screw press and the dewatered cake. In one embodiment, the plurality of target conditions include a first target condition defining stable operation of the screw press, a second target condition specifying that the predicted moisture content is within the moisture content target range, and a third target condition specifying that the predicted SS recovery rate is within the SS recovery rate target range.

[0015] In one embodiment, the method is characterized by selecting a candidate that satisfies all of the aforementioned target conditions and also satisfies predetermined constraint conditions regarding the operating parameters. In one aspect, the operating parameter further includes the amount of sludge to be fed into the screw press, and the predetermined constraint condition is a condition that restricts the correlation between the amount of sludge to be treated and the screw rotation speed. In one aspect, the predetermined constraint condition is a condition that restricts the correlation between the injection rate of the flocculant and the stirring speed.

[0016] In one aspect, it further includes determining whether the actual value of the sludge level in the sludge inlet of the screw press or the actual change rate of the level is within a predetermined allowable range, whether the actual torque of the screw is within a target range, or whether the rotation of the screw and the dewatered cake actually occurs within the screw press, based on a determination index, and the determination index is at least one of a measured value of the sludge level in the sludge inlet of the screw press, a detected value of the torque of the screw, a measured value of the pressure applied to the sludge in the filter cylinder of the screw press, and the state of the dewatered cake discharged from the filter cylinder. In one aspect, the operating parameter further includes the concentration of the sludge. In one aspect, among the plurality of candidates that satisfy the plurality of target conditions, the candidate most suitable for the preset operating mode of the dewatering device is selected. In one aspect, a deviation degree from a plurality of actual operating data included in the learning data used for machine learning of the plurality of state prediction models is respectively assigned to the plurality of candidates that satisfy the plurality of target conditions, and the candidate with the deviation degree below a predetermined standard is selected.

Advantages of the Invention

[0017] According to the present invention, an optimal operating parameter that can satisfy all of the plurality of target conditions is determined, and the dewatering device is operated using this operating parameter. Therefore, an optimal and stable operation of the dewatering device including a screw press is realized.

Brief Description of the Drawings

[0018] [Figure 1] It is a diagram showing an embodiment of a dehydration system. [Figure 2] It is a table showing an example of a plurality of candidates for operating parameters. [Figure 3] It is a table showing an example of a plurality of candidates for operating parameters and a state prediction index obtained by an operation prediction model. [Figure 4] It is a flowchart for explaining an embodiment of a method for determining an operating parameter by an operation condition determination system. [Figure 5] It is a table showing an example of a plurality of candidates for operating parameters, an operation prediction result obtained by an operation prediction model, a predicted moisture content obtained by a moisture content prediction model, and a predicted SS recovery rate obtained by an SS recovery rate prediction model. [Figure 6] It is a diagram showing an embodiment of a dehydration system provided with a biaxial screw press.

Mode for Carrying Out the Invention

[0019] Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a diagram showing an embodiment of a dehydration system. The dehydration system includes a dehydration device 1 that dehydrates sludge, which is an example of a suspension, that is, separates the sludge into a dehydrated cake and filtrate, an operation control unit 2 that controls the operation of the dehydration device 1, and an operation condition determination system 3 that determines optimal operation parameters for the dehydration device 1. The dehydration device 1 of the present embodiment includes a screw press 6 that pressurizes and dehydrates sludge, and a coagulation mixing tank 7 that sends sludge to the screw press 6.

[0020] The dehydrated cake is a low-moisture substance remaining after removing liquid from the sludge. Specific examples of the sludge include sludge generated during the treatment of sewage or night soil, suspensions containing industrial waste generated during the production of industrial products such as food products, cosmetics, and paper, or slurries.

[0021] The coagulation and mixing tank 7 is configured to inject a coagulant into the sludge, agitate the sludge and coagulant, and send the sludge mixed with the coagulant to the screw press 6. The coagulation and mixing tank 7 includes a container 10 for containing the sludge and a stirrer 11 for agitating the sludge in the container 10. The stirrer 11 includes a stirring blade 14 placed inside the container 10 and a stirring motor 15 connected to the stirring blade 14. The container 10 is connected to a sludge introduction pipe 20, and a pump 22 is provided in the sludge introduction pipe 20. The sludge is transferred to the coagulation and mixing tank 7 through the sludge introduction pipe 20 by the pump 22.

[0022] The flow rate of sludge into the coagulation and mixing tank 7 can be adjusted by operating the pump 22. The pump 22 is connected to the operation control unit 2, and the operation of the pump 22, i.e., the flow rate of sludge into the coagulation and mixing tank 7, is controlled by the operation control unit 2. A sludge concentration meter 24 is attached to the sludge inlet pipe 20, and the concentration of the sludge introduced into the coagulation and mixing tank 7 (i.e., the proportion of solids or suspended matter in the sludge) is measured by the sludge concentration meter 24. The measured value of the sludge concentration is input to the operation condition determination system 3. The sludge concentration meter 24 may be a sampling concentration meter that collects a certain amount of sludge, dries it, and calculates the sludge concentration from its dry weight.

[0023] The amount of sludge processed by the coagulation and mixing tank 7 and the screw press 6, i.e., the amount of sludge processed, is the amount of solid matter (suspended solids) in the sludge per unit time. The operating condition determination system 3 is configured to calculate the amount of sludge processed by multiplying the sludge concentration by the sludge flow rate. The flow rate of sludge fed into the screw press 6 may be estimated from the operating speed of the pump 22, or it may be measured by a sludge flow meter (not shown).

[0024] A coagulant is injected into the sludge in the coagulation mixing tank 7. The sludge is mixed with the coagulant by the agitator 11. By mixing the coagulant and sludge in the coagulation mixing tank 7, coagulated flocs are formed, which are aggregates of solid matter (suspended solids) in the sludge. The coagulation mixing tank 7 shown in Figure 1 is a single-stage tank, but the coagulation mixing tank 7 may have multiple stages. The intensity of sludge mixing can be adjusted by the rotational speed of the agitator 11 (hereinafter referred to as the mixing speed). The agitator 11 is connected to the operation control unit 2, and the operation of the agitator 11, i.e., the mixing speed, is controlled by the operation control unit 2.

[0025] The coagulation and mixing tank 7 is connected to the sludge inlet 28 of the screw press 6 by a sludge transfer pipe 27. The sludge mixed with the coagulant in the coagulation and mixing tank 7 is transferred to the screw press 6 through the sludge transfer pipe 27. In one embodiment, a concentrator may be provided between the coagulation and mixing tank 7 and the screw press 6.

[0026] The screw press 6 comprises a filter cylinder 30, a screw 32 concentrically arranged within the filter cylinder 30, and a screw motor 38 that rotates the screw 32 to send sludge toward the discharge chamber 41. The screw 32 has a screw shaft 35 and screw blades 36 fixed to the outer surface of the screw shaft 35. The filter cylinder 30 is made of a perforated plate such as perforated metal. One end of the filter cylinder 30 is sealed by a sealing wall 40, and the other end of the filter cylinder 30 is connected to the discharge chamber 41. The filter cylinder 30 has a sludge inlet 28 adjacent to the sealing wall 40.

[0027] The screw shaft 35 extends through the inside of the filter cylinder 30. The screw shaft 35 has a frustoconical shape, with its diameter gradually increasing towards the downstream side. The screw shaft 35 extends through the closing wall 40, and the end of the screw shaft 35 is connected to the screw motor 38. The screw motor 38 is connected to the operation control unit 2, and the operation of the screw motor 38, i.e., the rotational speed of the screw 32, is controlled by the operation control unit 2.

[0028] The screw press 6 is equipped with an inverter 39 that supplies a variable frequency current to the screw motor 38. The operation control unit 2 is configured to detect the torque of the screw 32 based on the current supplied from the inverter 39 to the screw motor 38. In one embodiment, the screw press 6 may be equipped with a torque detector that detects the torque of the screw 32.

[0029] The screw blade 36 is a single blade that extends spirally along the longitudinal direction of the screw shaft 35. A minute gap is formed between the inner surface of the filter cylinder 30 and the screw blade 36, allowing the screw blade 36 to rotate without contacting the filter cylinder 30. Sludge introduced into the filter cylinder 30 from the sludge inlet 28 is transported through the filter cylinder 30 toward the discharge chamber 41 by the rotating screw blade 36.

[0030] The space through which the sludge is transported within the filter cylinder 30 is formed by the inner surface of the filter cylinder 30, the screw blades 36, and the screw shaft 35. The volume of this space gradually decreases along the direction of sludge movement. Therefore, as the sludge is moved through this space by the screw blades 36, it is compressed and dewatered. The filtrate that has passed through the filter cylinder 30 is collected by a filtrate receiver 45 located below the filter cylinder 30 and then discharged.

[0031] An annular back pressure plate 50 is positioned opposite the downstream end of the filter cylinder 30. This back pressure plate 50 has a frustoconical shape with a tapered surface for receiving the dewatered sludge transported inside the filter cylinder 30. A through hole is formed in the center of the back pressure plate 50 through which the screw shaft 35 passes, and the back pressure plate 50 is positioned concentrically with the screw shaft 35. The back pressure plate 50 is not fixed to the screw shaft 35 and does not rotate.

[0032] The back pressure plate 50 is connected to a back pressure plate drive unit 51. This back pressure plate drive unit 51 is configured to move the back pressure plate 50 in the axial direction of the screw shaft 35. The gap between the back pressure plate 50 and the downstream end of the filter cylinder 30 is adjusted by the back pressure plate 50. The back pressure plate drive unit 51 is composed of, for example, a hydraulic cylinder or an electric cylinder. The back pressure plate drive unit 51 is connected to the operating condition determination system 3, and the operation of the back pressure plate drive unit 51, i.e., the axial position of the back pressure plate 50, is controlled by the operation control unit 2.

[0033] Next, the operation of the screw press 6 will be described. Sludge is fed into the filter cylinder 30 from the sludge inlet 28. The sludge is transported through the filter cylinder 30 toward the discharge chamber 41 by the rotating screw 32. As it moves through the filter cylinder 30, the sludge is compressed and dewatered. The filtrate that has passed through the filter cylinder 30 is collected by the filtrate receiver 45 and discharged. The sludge is dewatered within the filter cylinder 30 to form a dewatered cake.

[0034] The dewatered cake, having moved through the filter cylinder 30, is pressed against the back pressure plate 50. The dewatered cake is compressed as its movement is hindered by the back pressure plate 50. This compressed dewatered cake forms a plug 52 that seals the downstream end of the filter cylinder 30. The plug 52 reduces the moisture content of the dewatered cake in the filter cylinder 30 by applying back pressure to subsequent dewatered cakes. As the dewatered cake forms the plug 52 within the filter cylinder 30, it is pushed by subsequent dewatered cakes and gradually discharged into the discharge chamber 41 through the gap between the back pressure plate 50 and the downstream end of the filter cylinder 30. The dewatered cake is discharged from the discharge chamber 41 through a chute 53 located at the bottom of the discharge chamber 41. In this way, liquid is removed from the sludge, and a dewatered cake with a low moisture content is produced.

[0035] The axial position of the back pressure plate 50 changes the gap between the back pressure plate 50 and the downstream end of the filter cylinder 30, and as a result, the compressive force applied to the sludge inside the filter cylinder 30 changes. More specifically, when the gap between the back pressure plate 50 and the downstream end of the filter cylinder 30 becomes smaller, a greater force is required to push out the plug 52, so the compressive force applied to the sludge inside the filter cylinder 30 increases. Therefore, the compressive force applied to the sludge can be adjusted not only by the rotational speed of the screw 32, but also by the position of the back pressure plate 50 relative to the filter cylinder 30.

[0036] The operating condition determination system 3 comprises a machine learning unit 61, a candidate generation unit 62, a state prediction unit 63, and a selection command unit 64. The operating condition determination system 3 is connected to the operation control unit 2. The operating condition determination system 3 consists of at least one computer. The operating condition determination system 3 comprises a storage device 3a that stores programs and an arithmetic unit 3b that performs calculations according to the instructions contained in these programs. The storage device 3a stores a plurality of state prediction models, a program for determining the operating parameters of the dewatering unit 1 based on a plurality of state prediction indicators obtained from the plurality of state prediction models, and a program for performing machine learning to construct the plurality of state prediction models. The storage device 3a comprises a main memory such as random access memory (RAM) and an auxiliary storage device such as a hard disk drive (HDD) or solid state drive (SSD). Examples of arithmetic units 3b include a CPU (central processing unit) and a GPU (graphics processing unit). However, the specific configuration of the operating condition determination system 3 is not limited to these examples.

[0037] The operating condition determination system 3 may consist of multiple computers. For example, the operating condition determination system 3 may be a combination of an edge server and a cloud server. In one embodiment, the operating condition determination system 3 and the operation control unit 2 may consist of a single computer.

[0038] In one embodiment, the machine learning unit 61, candidate generation unit 62, state prediction unit 63, and selection command unit 64 described below are devices virtually constructed within the operating condition determination system 3, and are composed of a storage device 3a and an arithmetic unit 3b. The operating condition determination system 3 may include multiple storage devices 3a and multiple arithmetic units 3b, and its specific configuration is not particularly limited as long as it can achieve the intended function. For example, the machine learning unit 61, candidate generation unit 62, state prediction unit 63, and selection command unit 64 may each be composed of separate computers.

[0039] The multiple state prediction models include at least two of the following: an operation prediction model, a moisture content prediction model, and a SS recovery rate prediction model. In this embodiment, the multiple state prediction models include an operation prediction model, a moisture content prediction model, and an SS recovery rate prediction model. The operation prediction model, the moisture content prediction model, and the SS recovery rate prediction model are stored in the storage device 3a of the operation condition determination system 3. The operation prediction model, the moisture content prediction model, and the SS recovery rate prediction model are configured to output (calculate) multiple state prediction indicators that represent the operating state of the screw press 6.

[0040] The condition prediction index output from the operation prediction model is at least one of the following: the predicted value or predicted rate of change of the sludge level in the sludge inlet 28 of the screw press 6, the predicted torque of the screw 32 of the screw press 6, and the predicted result of whether or not co-rotation between the screw 32 and the dewatered cake occurs. The predicted value of the sludge level is the predicted height of the sludge level in the sludge inlet 28, and the predicted rate of change of the sludge level is the predicted amount of change in the sludge level per set time. Co-rotation between the screw 32 and the dewatered cake is a phenomenon in which the dewatered cake rotates together with the screw 32 due to the discharge resistance of the dewatered cake as the moisture content of the dewatered cake decreases. When such co-rotation occurs, the cake cannot be discharged and the dewatering process becomes impossible.

[0041] The condition prediction index output from the moisture content prediction model is the predicted moisture content of the dewatered cake discharged from the screw press 6. The condition prediction index output from the SS recovery rate prediction model is the predicted SS recovery rate of the screw press 6.

[0042] First, let's explain the operation prediction model. The machine learning unit 61 is configured to build an operation prediction model by performing machine learning using training data. The operation prediction model is a model that calculates the operation prediction result of the screw press 6 from the operation parameters of the dewatering unit 1.

[0043] The operation prediction model of this embodiment is configured to output a predicted value or predicted rate of change of the sludge level in the sludge inlet 28 of the screw press 6, a predicted torque of the screw 32 of the screw press 6, and a predicted result of whether or not co-rotation between the screw 32 and the dewatered cake occurs as state prediction indicators. In one embodiment, the operation prediction model may be configured to output one or two of the following as state prediction indicators: a predicted value or predicted rate of change of the sludge level in the sludge inlet 28 of the screw press 6, a predicted torque of the screw 32 of the screw press 6, and a predicted result of whether or not co-rotation between the screw 32 and the dewatered cake occurs.

[0044] The operating parameters of the dewatering device 1 include at least the injection rate of the coagulant into the sludge in the coagulation mixing tank 7, the stirring speed of the sludge and coagulant in the coagulation mixing tank 7 (rotation speed of the agitator 11), and the rotation speed of the screw 32 of the screw press 6.

[0045] In one embodiment, the operating parameters of the dewatering device 1 may further include the position of the back pressure plate 50 relative to the filter cylinder 30. Furthermore, in one embodiment, the operating parameters of the dewatering device 1 may further include the concentration of sludge sent to the dewatering device 1 and / or the flow rate of sludge sent to the dewatering device 1. The sludge concentration meter 24 measures the concentration of sludge before it is sent to the coagulation and mixing tank 7 and sends the measured value of the sludge concentration to the operating condition determination system 3. The flow rate of sludge sent to the dewatering device 1 can be estimated from the operating speed of the pump 22. Alternatively, the flow rate of sludge sent to the dewatering device 1 may be measured by a sludge flow meter (not shown).

[0046] The training data used for machine learning includes multiple actual operating data of operating parameters previously used in the dewatering device 1, and the correct labels corresponding to each of these multiple actual operating data, which include measured values ​​or the rate of change of measured values ​​of the sludge level in the sludge inlet 28 of the screw press 6, detected values ​​of the torque of the screw 32, and information on whether the screw 32 and the dewatered cake rotate together. The training data is stored in the storage device 3a. The various data used for training may be acquired by sensors, or by manual measurement or manual analysis.

[0047] The operating condition determination system 3 further includes a stable operation determination unit 65 that determines, based on multiple determination indicators, whether or not the screw 32 and the dewatered cake are actually rotating together inside the screw press 6. Examples of multiple determination indicators include the measured level of sludge in the sludge inlet 28, the measured pressure applied to the sludge in the filter cylinder 30, and the state of the dewatered cake discharged from the filter cylinder 30 to the discharge chamber 41. The sludge level in the sludge inlet 28 is measured by a level sensor 70 located above the sludge inlet 28. The pressure applied to the sludge in the filter cylinder 30 is measured by a pressure sensor 71 located inside the filter cylinder 30. The state of the dewatered cake discharged from the filter cylinder 30 to the discharge chamber 41 is detected based on an image generated by an imaging device 73.

[0048] In one embodiment, the stable operation determination unit 65 may be configured to determine whether or not the screw 32 and the dewatered cake are rotating together in the screw press 6, based on the measured value of the sludge level in the sludge inlet 28 acquired by the level sensor 70, the measured value of the sludge pressure acquired by the pressure sensor 71, and the image of the dewatered cake generated by the imaging device 73.

[0049] In one embodiment, the stable operation determination unit 65 may be configured to determine whether the actual value of the sludge level in the sludge inlet 28 or the actual rate of change of the sludge level is within a predetermined allowable range, based on the measured value of the sludge level in the sludge inlet 28 obtained by the level sensor 70. The allowable range is a range in which the fluctuation of the sludge level in the sludge inlet 28 is small and the sludge level in the sludge inlet 28 can be considered to be approximately constant.

[0050] In one embodiment, the stable operation determination unit 65 may be configured to determine whether the actual torque of the screw 32 is within the target range based on the detected torque value.

[0051] The driving prediction model constructed by machine learning using the above training data is a pre-trained model. The driving prediction model may be a regression model or a classification model. The pre-trained driving prediction model is stored in the memory device 3a.

[0052] Examples of machine learning methods include SVR (Support Vector Regression), PLS (Partial Least Squares), deep learning, random forests, and decision trees. For example, a driving prediction model is composed of a neural network built using deep learning.

[0053] The candidate generation unit 62 is configured to generate multiple candidates for the operating parameters of the dewatering device 1. Figure 2 is a table showing an example of multiple candidates for operating parameters. In the example shown in Figure 2, each candidate consists of operating parameters including the coagulant injection rate, the coagulant stirring speed (rotation speed of the agitator 11), the rotation speed of the screw 32, and the sludge concentration (measured by the sludge concentration meter 24). The algorithm for generating multiple candidates for operating parameters is not particularly limited. For example, the candidate generation unit 62 may generate multiple candidates by gradually changing a reference operating parameter, or it may generate multiple candidates by combining a set number of pre-set values ​​for the operating parameter.

[0054] The state prediction unit 63 inputs each of the multiple candidate operating parameters generated by the candidate generation unit 62 into the operation prediction model and outputs state prediction indices from the operation prediction model that indicate the operation prediction result of the screw press 6, namely the predicted value or predicted rate of change of the sludge level in the sludge inlet 28 of the screw press 6, the predicted torque of the screw 32, and the predicted result of whether or not co-rotation between the screw 32 and the dewatered cake will occur.

[0055] The form of the predicted value or predicted rate of change of sludge level output from the operational prediction model is not particularly limited. For example, the predicted value of sludge level may be expressed as an absolute value representing the sludge level, or on a 10-point scale, and the predicted rate of change of sludge level may be expressed as a percentage.

[0056] The form of the predictive result for co-rotation output from the driving prediction model is not particularly limited. For example, the driving prediction model may be configured to output a numerical value (e.g., 1) indicating that co-rotation will occur, or a numerical value (e.g., 0) indicating that co-rotation will not occur. In another example, the driving prediction model may be configured to output the probability of co-rotation occurring (e.g., 0 to 100%, or a 10-level scale). In this case, the selection command unit 64, described below, may be configured to compare the probability of co-rotation occurring with a threshold value, and to determine that a probability greater than the threshold value is a predicted result for co-rotation occurring, and a probability less than or equal to the threshold value is a predicted result for no co-rotation occurring.

[0057] Figure 3 is a table showing several candidate operating parameters and an example of a state prediction index obtained by the operating prediction model. In the example shown in Figure 3, the operating parameters constituting each candidate include the coagulant injection rate, the coagulant stirring speed (rotation speed of the agitator 11), the rotation speed of the screw 32, and the sludge concentration. For each candidate, a state prediction index is obtained by the operating prediction model. By including the sludge concentration in the operating parameters, the screw torque, co-rotation, and the sludge level in the sludge inlet 28 can be predicted with greater accuracy.

[0058] In the example shown in Figure 3, the predicted value of the sludge level in the sludge inlet 28 is calculated as the state prediction index. However, instead of the predicted value of the sludge level, the predicted rate of change of the sludge level may be calculated as the state prediction index. Furthermore, in one embodiment, the operation prediction model may be configured to output one or two of the following: the predicted value or predicted rate of change of the sludge level in the sludge inlet 28, the predicted torque of the screw 32, and the predicted result of co-rotation.

[0059] Next, the moisture content prediction model will be described. The operating condition determination system 3 has a moisture content prediction model in addition to the operating prediction model. The moisture content prediction model is stored in the storage device 3a. The moisture content prediction model is a model that predicts the moisture content of the dewatered cake discharged from the screw press 6 from the operating parameters of the dewatering device 1. The moisture content prediction model is configured to output the predicted moisture content as a state prediction index.

[0060] The machine learning unit 61 is configured to perform machine learning using training data to construct a moisture content prediction model. The training data used for machine learning includes multiple actual operating data of operating parameters previously used in the dewatering device 1, and the moisture content of the dewatered cake discharged from the screw press 6, which are the correct labels corresponding to each of these multiple actual operating data. The training data is stored in the storage device 3a. The types of operating parameters in the actual operating data included in the training data for constructing the moisture content prediction model may be the same as or different from the types of operating parameters in the actual operating data included in the training data for constructing the operation prediction model, but at least include the coagulant injection rate, the coagulant stirring speed (rotation speed of the agitator 11), and the rotation speed of the screw 32. The various data used for the training data may be acquired by sensors, or by manual measurement or manual analysis.

[0061] The moisture content prediction model constructed by machine learning using the above training data is a pre-trained model. The moisture content prediction model may be a regression model or a classification model. The pre-trained moisture content prediction model is stored in the storage device 3a.

[0062] Next, the SS recovery rate prediction model will be described. The operating condition determination system 3 has an SS recovery rate prediction model in addition to the operating prediction model and the moisture content prediction model. SS stands for Suspended Solids. The SS recovery rate prediction model is stored in the memory device 3a. The SS recovery rate prediction model is a model that predicts the SS recovery rate of the screw press 6 from the operating parameters of the dewatering device 1. The SS recovery rate prediction model is configured to output the predicted SS recovery rate as a state prediction index.

[0063] The machine learning unit 61 is configured to perform machine learning using training data to construct a SS recovery rate prediction model. The training data used for machine learning includes multiple actual operating data of operating parameters previously used in the dewatering device 1, and the SS recovery rate of the screw press 6, which is the correct label corresponding to each of these multiple actual operating data. The training data is stored in the storage device 3a. The types of operating parameters in the actual operating data included in the training data for constructing the SS recovery rate prediction model may be the same as or different from the types of operating parameters in the actual operating data included in the training data for constructing the operation prediction model, but at least include the coagulant injection rate, the coagulant stirring speed (rotation speed of the agitator 11), and the rotation speed of the screw 32. The various data used for the training data may be acquired by sensors, or by manual measurement or manual analysis.

[0064] The SS recovery rate prediction model constructed by machine learning using the above training data is a pre-trained model. The SS recovery rate prediction model may be a regression model or a classification model. The SS recovery rate prediction model, as a pre-trained model, is stored in the memory device 3a.

[0065] The state prediction unit 63 is configured to input each of the multiple candidates generated by the candidate generation unit 62 into the operation prediction model, the moisture content prediction model, and the SS recovery rate prediction model, output a state prediction index from the operation prediction model that shows the operation prediction result of the screw press 6, output a state prediction index from the moisture content prediction model that shows the predicted moisture content of the dewatered cake discharged from the screw press 6, and output a state prediction index from the SS recovery rate prediction model that shows the predicted SS recovery rate of the screw press 6.

[0066] The selection command unit 64 is configured to select from among multiple candidates a candidate in which the state prediction index output from the operation prediction model indicates stable operation of the screw press 6, the predicted moisture content is within the moisture content target range, and the predicted SS recovery rate is within the SS recovery rate target range. According to this embodiment, stable operation of the screw press 6, a low moisture content in the dewatered cake, and a high SS recovery rate can be achieved.

[0067] The operating condition determination system 3 may further include a moisture content determination unit 66 that determines whether the actual moisture content of the dewatered cake is within the target moisture content range. The moisture content of the dewatered cake is measured by a moisture meter (not shown) that measures the moisture content of the dewatered cake discharged from the screw press 6.

[0068] The operating condition determination system 3 may further include an SS recovery rate determination unit 67 that determines whether the actual SS recovery rate of the screw press 6 is within the target SS recovery rate range. The SS recovery rate is calculated from the measured SS concentration of the filtrate discharged from the screw press 6, which is measured by an SS concentration meter (not shown).

[0069] Figure 4 is a flowchart illustrating one embodiment of the method for determining operating parameters using the operating condition determination system 3. In step 1, the candidate generation unit 62 generates several candidate operating parameters for the dewatering device 1. The operating parameters include at least the injection rate of the coagulant into the sludge, the stirring speed of the sludge and coagulant in the coagulation mixing tank 7, and the rotation speed of the screw 32. In step 2, the state prediction unit 63 inputs the multiple candidates generated in step 1 into the operation prediction model and performs calculations according to the algorithm defined by the operation prediction model, which is a trained model, to determine (calculate) the operation prediction result of the screw press 6 for each candidate (i.e., at least one of the predicted value or predicted rate of change of the sludge level in the sludge inlet 28 of the screw press 6, the predicted torque of the screw 32, and the predicted result of whether or not co-rotation between the screw 32 and the dewatered cake occurs).

[0070] In step 3, the state prediction unit 63 inputs the multiple candidates generated in step 1 into the moisture content prediction model and performs calculations according to the algorithm defined by the trained moisture content prediction model to determine (calculate) the predicted moisture content of the dewatered cake discharged from the screw press 6 for each candidate. In step 4, the state prediction unit 63 inputs the multiple candidates generated in step 1 into the SS recovery rate prediction model and performs calculations according to the algorithm defined by the trained SS recovery rate prediction model to determine (calculate) the predicted SS recovery rate of the screw press 6 for each candidate. The order in which steps 2, 3, and 4 are performed is not particularly limited. Steps 2, 3, and 4 may be performed simultaneously.

[0071] In step 5, the selection command unit 64 selects from among the multiple candidates generated in step 1 a candidate that satisfies all of the predetermined target conditions for the predicted operating result of the screw press 6, the predicted moisture content, and the predicted SS recovery rate. Multiple target conditions are provided for the predicted operating result of the screw press 6, the predicted moisture content, and the predicted SS recovery rate, respectively. In one embodiment, the multiple target conditions include a first target condition that defines stable operation of the screw press 6, a second target condition that specifies that the predicted moisture content of the dewatered cake discharged from the screw press 6 is within the moisture content target range, and a third target condition that specifies that the predicted SS recovery rate of the screw press 6 is within the SS recovery rate target range.

[0072] The first target condition defining stable operation of the screw press 6 includes at least one of the following: the predicted value or predicted rate of change of the sludge level in the sludge inlet 28 of the screw press 6 is within a predetermined allowable range; the predicted torque of the screw 32 is within a predetermined target range; and no co-rotation occurs between the screw 32 and the dewatered cake. The above allowable range for the predicted value or predicted change of the sludge level is a range in which the fluctuation of the sludge level in the sludge inlet 28 is small and the sludge level in the sludge inlet 28 can be considered to be approximately constant.

[0073] If the operation prediction model is configured to output a predicted value or predicted rate of change of the sludge level in the sludge inlet 28 of the screw press 6, a predicted torque of the screw 32, and a predicted result of whether or not co-rotation of the screw 32 and the dewatered cake will occur, then the first target condition is that the predicted value or predicted rate of change of the sludge level in the sludge inlet 28 of the screw press 6 is within a predetermined allowable range, the predicted torque of the screw 32 is within a predetermined target range, and co-rotation of the screw 32 and the dewatered cake will not occur. If the operation prediction model is configured to output only a predicted value or predicted rate of change of the sludge level in the sludge inlet 28 of the screw press 6, then the first target condition is that the sludge level in the sludge inlet 28 of the screw press 6 is within a predetermined allowable range. If the operation prediction model is configured to output only a predicted torque of the screw 32, then the first target condition is that the predicted torque of the screw 32 is within a predetermined target range. If the operation prediction model is configured to output only a prediction result of whether or not co-rotation occurs between the screw 32 of the screw press 6 and the dewatered cake, then the first target condition is that co-rotation does not occur between the screw 32 and the dewatered cake.

[0074] In step 6, the selection command unit 64 sends the operating parameters constituting the candidate selected in step 5 to the operation control unit 2. The operation control unit 2 operates the coagulation mixing tank 7 and screw press 6 of the dewatering device 1 using the operating parameters determined by the operating condition determination system 3.

[0075] Figure 5 is a table showing an example of several candidate operating parameters, the predicted operating results of the screw press 6 obtained by the operating prediction model, the predicted moisture content obtained by the moisture content prediction model, and the predicted SS recovery rate obtained by the SS recovery rate prediction model. In this example, the first target condition is that the predicted rate of change of the sludge level in the sludge inlet 28 of the screw press 6 is within a predetermined allowable range and that the screw 32 and the dewatered cake do not rotate together; the second target condition is that the predicted moisture content of the dewatered cake is within the moisture content target range of 83-85%; and the third target condition is that the predicted SS recovery rate is within the moisture content target range of 93-95%. The selection command unit 64 selects candidate operating parameter No. 4 that satisfies all of the first, second, and third target conditions.

[0076] There may be multiple candidates that satisfy all of the first, second, and third target conditions. In this case, the selection command unit 64 selects a candidate that satisfies all of the first, second, and third target conditions and also satisfies predetermined constraints regarding the operating parameters. The constraints are predetermined based on past operating data of the dewatering device 1 when all of the first, second, and third target conditions are satisfied.

[0077] In one embodiment, the above constraint is a condition that restricts the correlation between the amount of sludge processed by the screw press 6 and the rotational speed of the screw 32 of the screw press 6. The amount of sludge processed by the screw press 6 is the amount of solid matter (suspended solids) contained in the sludge per unit time. This amount of sludge processed can be calculated by multiplying the concentration of the sludge processed by the flow rate of the sludge. The flow rate of the sludge processed by the screw press 6 may be estimated from the operating speed of the pump 22, or it may be measured by a sludge flow meter (not shown). In a twin-screw type screw press 6 (described later), the above constraint is a condition that restricts the correlation between the amount of sludge processed and the rotational speed of the second screw on the rear shaft.

[0078] For example, the constraints are defined by the following function. Screw rotation speed = C1 × Sludge processing capacity + C2 Here, C1 is a predetermined coefficient, and C2 is a coefficient that fluctuates within a predetermined tolerance range (e.g., a standard value ±3). These coefficients C1 and C2 are predetermined based on past operating data of the dewatering device 1 when all of the first, second, and third target conditions are met.

[0079] In one embodiment, the above constraint is a condition that restricts the correlation between the injection rate of the coagulant into the sludge in the coagulation mixing tank 7 and the stirring speed of the sludge and coagulant in the coagulation mixing tank 7 (the rotational speed of the agitator 11). For example, the constraint is defined by the following function. Stirring speed = C3 x flocculant injection rate + C4 Here, C3 is a preset coefficient, and C4 is a coefficient that fluctuates within a preset tolerance range (e.g., a standard value ±10). These coefficients C3 and C4 are predetermined based on past operating data of the dewatering device 1 when all of the first, second, and third target conditions are met. Alternatively, the above constraints may be functions determined based on physicochemical formulas. Furthermore, the above constraints may be conditions that restrict the correlation between the coagulant injection rate, the stirring speed of the sludge and coagulant, and the amount of sludge processed by the screw press 6. By setting such constraints, the operating condition determination system 3 can determine the operating parameters with higher accuracy.

[0080] In one embodiment, the selection command unit 64 may be configured to select the candidate that best suits the preset operating mode of the dewatering device 1 from among a plurality of candidates that satisfy the plurality of target conditions described above (including the first target condition, the second target condition, and the third target condition). Examples of operating modes include a cost optimization mode that minimizes total costs such as coagulant costs and sludge removal costs, a moisture content priority mode that reduces the moisture content of the dewatered cake as much as possible, and a coagulant suppression mode that minimizes the coagulant injection rate.

[0081] In cost optimization mode, the total cost is calculated for each of the multiple candidates that satisfy the above-mentioned target conditions (including the first, second, and third target conditions), and the candidate corresponding to the lowest total cost is selected. In moisture content priority mode, the candidate corresponding to the lowest predicted moisture content is selected from among the multiple predicted moisture content calculated using the moisture content prediction model. In coagulant suppression mode, the candidate with the lowest coagulant injection rate is selected from among the multiple candidates that satisfy the above-mentioned target conditions.

[0082] In one embodiment, the selection command unit 64 may be configured to assign a degree of deviation from multiple actual operating data included in the training data used in machine learning for the operation prediction model, moisture content prediction model, and SS recovery rate prediction model to multiple candidates that satisfy the above-described multiple target conditions (including the first target condition, second target condition, and third target condition), and to select a candidate whose degree of deviation is less than or equal to a predetermined standard (in one example, the candidate with the smallest distance). The degree of deviation represents the reliability of the candidate operating parameter. That is, a small degree of deviation means that the candidate operating parameter is highly reliable. To more accurately determine the reliability of the candidate operating parameter, the degree of deviation may be the degree of deviation from actual operating data in which no co-rotation occurred within the screw press 6.

[0083] In one example, the degree of deviation is expressed as the difference between the driving parameters of each candidate and the actual driving data included in the training data. More specifically, the selection command unit 64 is configured to calculate the distance between the driving parameters of each candidate that satisfy the above-mentioned multiple target conditions and multiple actual driving data, and to select the candidate whose distance is less than or equal to a predetermined standard (in one example, the candidate with the smallest distance). The distance can be the sum of the differences between the multiple driving parameters of each candidate and the corresponding driving parameters of each actual driving data. However, the method of calculating the distance is not limited to this. The driving parameters of candidates with small distances are expected to have a small degree of deviation from the actual driving data and therefore be highly reliable.

[0084] In other examples, the deviation degree may be a predetermined deviation degree. More specifically, the selection command unit 64 selects the representative data closest to each candidate that satisfies the multiple target conditions described above from a set of multiple representative data. Multiple predicted deviation degrees are pre-associated with each of these multiple representative data. The selection command unit 64 is configured to select a candidate corresponding to representative data whose predicted deviation degree is below a predetermined standard (in one example, the representative data with the smallest predicted deviation degree).

[0085] The selection command unit 64 can select more accurate candidate operating parameters based on the degree of deviation from actual operating data.

[0086] The embodiments described so far are applicable not only to the single-screw press 6 shown in Figure 1, but also to the twin-screw press described below. Figure 6 shows one embodiment of a dewatering system equipped with a twin-screw press. The configuration and operation of this embodiment, which are not specifically described, are the same as those of the embodiments described with reference to Figures 1 to 4, so redundant explanations are omitted.

[0087] As shown in Figure 6, the twin-screw press 75 includes a cylindrical filter cylinder 80, a first screw 81 and a second screw 82 arranged concentrically with the filter cylinder 80 to transport sludge in a predetermined transport direction D, a first screw motor 85 that rotates the first screw 81, and a second screw motor 86 that rotates the second screw 82 independently of the first screw 81. Sludge introduced into the filter cylinder 80 from the sludge inlet 88 is transported in a predetermined transport direction D within the filter cylinder 80 by the rotating first screw 81 and second screw 82. The operation of the first screw motor 85 and the second screw motor 86, i.e., the rotational speed of the first screw 81 and the rotational speed and direction of the second screw 82, are controlled by the operation control unit 2.

[0088] The second screw 82 is connected to the first screw 81 so that it can rotate independently of the first screw 81. The first screw 81 and the second screw 82 extend through the filter cylinder 80 and the discharge chamber 90, respectively. The discharge chamber 90 is connected to the filter cylinder 80. The dewatered cake is discharged from the filter cylinder 80 into the discharge chamber 90.

[0089] The axial length of the second screw 82 is shorter than the axial length of the first screw 81. The first screw 81 has a first screw shaft 81A which is frustoconical (tapered) in shape and whose diameter gradually increases toward the downstream side in the sludge transport direction D, and a first screw blade 81B fixed to the outer surface of the first screw shaft 81A. The second screw 82 has a cylindrical second screw shaft 82A and a second screw blade 82B fixed to the outer surface of the second screw shaft 82A.

[0090] The second screw shaft 82A of the second screw 82 is positioned concentrically with the first screw shaft 81A. The outer diameter of the second screw shaft 82A is the same as the maximum diameter of the first screw shaft 81A. The second screw shaft 82A extends through the discharge chamber 90.

[0091] The upstream end of the first screw shaft 81A, which extends through the closure wall 89, is rotatably supported by bearings 91 and 92. The first screw shaft 81A is connected to a first screw motor 85 for rotating the first screw 81. The second screw shaft 82A is connected to a second screw motor 86 for rotating the second screw 82.

[0092] In this embodiment, the winding direction (i.e., the helical direction) of the second screw blade 82B is opposite to the winding direction of the first screw blade 81B. Therefore, when sending the sludge introduced from the sludge inlet 88 to the discharge chamber 90, the second screw 82 is rotated in the opposite direction to the first screw 81, as shown in Figure 6.

[0093] The winding direction of the second screw blade 82B may be the same as the winding direction of the first screw blade 81B. In this case, when the sludge introduced from the sludge inlet 88 is sent to the discharge chamber 90, the second screw 82 will rotate in the same direction as the first screw 81.

[0094] As shown in Figure 6, the filter cylinder 80 is divided into a dewatering region 100A where the first screw 81 is located and a plug-forming region 100B where the second screw 82 is located. The space through which the sludge is transported in the dewatering region 100A is formed by the inner surface of the filter cylinder 80, the first screw blades 81B, and the first screw shaft 81A. The cross-sectional area of ​​this transport space gradually decreases along the sludge transport direction D, as shown in Figure 6. Therefore, as the sludge introduced from the sludge inlet 88 is transported through this transport space by the first screw blades 81B, the sludge is compressed and dewatered. The filtrate that has passed through the filter cylinder 80 is collected by a filtrate receiver 93 located below the filter cylinder 80. A drain 94 is connected to the filtrate receiver 93, and the filtrate collected by the filtrate receiver 93 is discharged from the screw press 75 via the drain 94.

[0095] The space through which sludge is transported in the plug formation region 100B is formed by the inner surface of the filter cylinder 80, the second screw blade 82B, and the second screw shaft 82A. As shown in Figure 6, the cross-sectional area of ​​this transport space is constant. In the plug formation region 100B, a plug cake is formed by the dewatered cake consisting of sludge dewatered in the dewatering region 100A.

[0096] The operation control unit 2 rotates the second screw 82 in the opposite direction to the rotation direction of the first screw 81, and the second screw blades 82B gradually send the plug cake to the discharge chamber 90 (i.e., discharge it). The plug cake on the second screw 82 is discharged to the discharge chamber 90 by the second screw blades 82B of the second screw 82, which is rotated by the second rotation mechanism 20, while applying back pressure to the subsequent dewatered cake.

[0097] The operation control unit 2 can adjust the amount of plug cake discharged into the discharge chamber 90 by changing the rotational speed of the second screw 82. More specifically, when the operation control unit 2 decreases the rotational speed of the second screw 82, the amount of plug cake discharged decreases, and when the operation control unit 2 increases the rotational speed of the second screw 82, the amount of plug cake discharged increases. When the amount of plug cake discharged decreases, the subsequent dewatered cake accumulates in the dewatering area 100A, and the back pressure applied to the subsequent dewatered cake increases. Therefore, the operation control unit 2 can reduce the moisture content of the subsequent dewatered cake by decreasing the rotational speed of the second screw 82. In one embodiment, the back pressure applied to the subsequent dewatered cake may be adjusted by performing intermittent operation, which alternately rotates and stops the second screw 82.

[0098] The second screw 82 may be rotated in the same direction as the first screw 81. When the second screw 82 is rotated in the same direction as the first screw 81, the plug cake formed in the plug forming region 100B is pushed toward the dewatering region 100A, and a greater back pressure can be applied to the dewatered cake in the dewatering region 100A. As a result, the dewatering efficiency of the sludge can be improved.

[0099] If the second screw 82 is rotated in the same direction as the first screw 81 for a long period of time, there is a risk that the dewatered cake in the dewatering area 100A will rotate together with the first screw 81. Therefore, after rotating the second screw 82 in the same direction as the first screw 81 for a certain period of time, the rotation direction of the second screw 82 is reversed to the opposite direction to the rotation direction of the first screw 81.

[0100] The motion control unit 2 changes the rotation speed and direction of the second screw 82, making it possible to dewater sludge to a low moisture content that cannot be achieved with conventional screw presses.

[0101] In one embodiment, the operation prediction model is configured to predict whether the first screw 81 or the second screw 82 and the dewatered cake rotate together in the screw press 75, based on the operating parameters of the dewatering device 1. The operating parameters of the dewatering device 1 include at least the injection rate of the coagulant into the sludge in the coagulation mixing tank 7, the stirring speed of the sludge and coagulant in the coagulation mixing tank 7 (rotation speed of the agitator 11), the rotation speed of the first screw 81, and the rotation speed of the second screw 82. In one embodiment, the operating parameters of the dewatering device 1 may further include the rotation direction of the second screw 82.

[0102] The embodiments described above are intended to enable persons with ordinary skill in the art to implement the present invention. Various modifications of the above embodiments can be made naturally by those skilled in the art, and the technical idea of ​​the present invention can be applied to other embodiments as well. Therefore, the present invention is not limited to the embodiments described, but is to be interpreted in the broadest sense according to the technical idea defined by the claims. [Explanation of symbols]

[0103] 1 Dehydration device 2. Operation Control Unit 3. Operating Condition Determination System 6 Screw press 7 Coagulation mixing tank 10 containers 11 Stirrer 14. Agitation blades 15. Stirring motor 20 Sludge Inlet Pipe 22 pumps 24 Sludge concentration meter 27 Sludge transfer pipe 28 Sludge inlet 30 filter cylinders 32 Screw 35 Screw shaft 36 Screw blades 38 Screw motor 39 Inverter 40 Blocking wall 41 Discharge chamber 45 Filtrate receiver 50 Back pressure plate 51 Back pressure plate drive device 52 plugs 61 Machine Learning Department 62 Candidate generation section 63 State prediction unit 64 Selection Command Unit 65 Stable operation determination unit 66 Moisture content determination section 67 SS Recovery Rate Determination Unit 70-level sensor 71 Pressure Sensor 73 Imaging device 75 Screw Press 80 filter cylinder 81. Screw No. 1 81A First screw shaft 81B First screw blade 82. Second Screw 82A Second screw shaft 82B Second screw blade 85. First screw motor 86. Second screw motor 89 Blocking wall 90 Discharge chamber 91,92 Bearings 93 Filtrate receiver 94 Drain 100A dehydration area 100B Plug formation area

Claims

1. A system for determining operating conditions for a dewatering apparatus comprising a coagulation mixing tank for injecting a coagulant into sludge, and a screw press for separating the sludge into a dewatered cake and filtrate, A candidate generation unit that generates a plurality of candidate operating parameters, including at least the injection rate of the coagulant into the sludge, the stirring speed of the sludge and the coagulant in the coagulation mixing tank, and the screw rotation speed of the screw press, A state prediction unit inputs each of the aforementioned multiple candidates into multiple state prediction models and outputs multiple state prediction indicators from each of the aforementioned multiple state prediction models, A system for determining operating conditions for a dewatering device, characterized by comprising a selection command unit that selects from the plurality of candidates a candidate that satisfies all of the predetermined target conditions for the plurality of state prediction indicators, and sends the operating parameters constituting the selected candidate to the operation control unit of the dewatering device.

2. The dewatering apparatus operating condition determination system according to claim 1, characterized in that the plurality of state prediction models include at least two of the following: an operation prediction model that outputs a state prediction index indicating the operation prediction result of the screw press; a moisture content prediction model that outputs a state prediction index indicating the predicted moisture content of the dewatered cake discharged from the screw press; and an SS recovery rate prediction model that outputs a state prediction index indicating the predicted SS recovery rate of the screw press.

3. The operating condition determination system for a dewatering apparatus according to claim 2, characterized in that the condition prediction index output from the operating prediction model is at least one of the following: the predicted value or predicted rate of change of the sludge level in the sludge inlet of the screw press, the predicted torque of the screw of the screw press, and the predicted result of whether or not co-rotation occurs between the screw of the screw press and the dewatered cake.

4. The dewatering apparatus operating condition determination system according to claim 2, characterized in that the plurality of target conditions include a first target condition defining stable operation of the screw press, a second target condition determining that the predicted moisture content is within the moisture content target range, and a third target condition determining that the predicted SS recovery rate is within the SS recovery rate target range.

5. The dewatering apparatus operating condition determination system according to claim 1, characterized in that the selection command unit is configured to select a candidate that satisfies all of the plurality of target conditions and also satisfies predetermined constraint conditions regarding the operating parameters.

6. The aforementioned operating parameters further include the amount of sludge to be processed by the screw press, The dewatering apparatus operating condition determination system according to claim 5, characterized in that the predetermined constraint conditions are conditions that restrict the correlation between the amount of sludge processed and the screw rotation speed.

7. The dewatering apparatus operating condition determination system according to claim 5, characterized in that the predetermined constraint conditions are conditions that restrict the correlation between the injection rate of the coagulant and the stirring speed.

8. The operating condition determination system further includes a stable operation determination unit that determines, based on determination indicators, whether the actual value of the sludge level in the sludge inlet of the screw press or the actual rate of change of the level is within a predetermined allowable range, whether the actual torque of the screw is within a target range, or whether the screw and the dewatered cake are actually rotating together inside the screw press. The dewatering device operating condition determination system according to claim 1, characterized in that the determination indicator is at least one of the following: a measured value of the sludge level in the sludge inlet of the screw press, a detected value of the torque of the screw, a measured value of the pressure applied to the sludge in the filter cylinder of the screw press, and the state of the dewatered cake discharged from the filter cylinder.

9. The operating conditions determination system for a dewatering apparatus according to claim 1, characterized in that the operating parameters further include the concentration of the sludge.

10. The dehydration device operating condition determination system according to claim 1, characterized in that the selection command unit is configured to select from among a plurality of candidates that satisfy the plurality of target conditions the candidate that is most suitable for a preset operating mode of the dehydration device.

11. The system for determining operating conditions of a dewatering apparatus according to claim 1, characterized in that the selection command unit is configured to assign a degree of deviation from a plurality of actual operating data included in the training data used for machine learning of the plurality of state prediction models to a plurality of candidates that satisfy the plurality of target conditions, and to select a candidate whose degree of deviation is less than or equal to a predetermined standard.

12. A method for operating a dewatering apparatus comprising a coagulation mixing tank for injecting a coagulant into sludge, and a screw press for separating the sludge into a dewatered cake and filtrate, Multiple candidate operating parameters are generated, including at least the injection rate of the coagulant into the sludge, the stirring speed of the sludge and the coagulant in the coagulation mixing tank, and the screw rotation speed of the screw press. Each of the above multiple candidates is input into multiple state prediction models, and multiple state prediction indicators are output from each of the above multiple state prediction models. A candidate that satisfies all of the predetermined target conditions for the aforementioned multiple state prediction indicators is selected from the aforementioned multiple candidates. A method for operating a dewatering apparatus, characterized by operating the dewatering apparatus with the operating parameters that constitute the selected candidate.

13. The method for operating a dewatering apparatus according to claim 12, characterized in that the plurality of state prediction models include at least two of the following: an operation prediction model that outputs a state prediction index indicating the operation prediction result of the screw press; a moisture content prediction model that outputs a state prediction index indicating the predicted moisture content of the dewatered cake discharged from the screw press; and an SS recovery rate prediction model that outputs a state prediction index indicating the predicted SS recovery rate of the screw press.

Images