Magnesium ammonium phosphate generation system and production method of the same

The system addresses inefficiencies in MAP recovery by using real-time monitoring and control mechanisms to adapt reaction conditions, ensuring efficient and compliant production of magnesium ammonium phosphate from sludge.

JP2024158513APending Publication Date: 2024-11-08SWING CORP
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
JP2023073769
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-04-27
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

Existing methods for recovering magnesium ammonium phosphate (MAP) from sludge face challenges in efficiently adapting to changes in sludge properties, leading to inefficiencies and difficulties in maintaining optimal reaction conditions due to the high viscosity and variability of organic wastewater or sludge, and compliance with regulations regarding chemical use.

Method used

A system and method that includes a measuring device to monitor phosphorus input load, a reaction tank with seed crystals, a drug supply device, a separation device, a cleaning device, and a transfer device, along with control units to adjust operating conditions based on real-time measurements, ensuring the Mg/P ratio and seed crystal concentration are maintained within optimal ranges.

Benefits of technology

This approach allows for rapid adjustment of reaction conditions to match sludge properties, stabilizing phosphorus production efficiency and ensuring compliance with regulatory standards by optimizing the crystallization process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024158513000001_ABST
    Figure 2024158513000001_ABST
Patent Text Reader

Abstract

To provide a magnesium ammonium phosphate generation system and a magnesium ammonium phosphate production method where the operation condition of a crystallization reaction vessel can be quickly adapted to a condition adapting to the property change of charged sludge and thereby, the production efficiency of phosphorus can be stably improved.SOLUTION: A magnesium ammonium phosphate generation system comprises: a reaction vessel where MAP particles are crystallized in a crystallization reaction liquid by mixing phosphorus-containing charged sludge in the crystallization reaction liquid in which the seed crystals of magnesium ammonium phosphate exist; an agent feeder to feed an agent containing magnesium into the reaction vessel; a drawing device to draw the crystallization reaction liquid containing the MAP particles from the reaction vessel; a separation device to separate the MAP particles from the crystallization reaction liquid; a washing device to store and wash the MAP particles separated by the separation device; and a measuring device where the variation of the phosphorus charge load amount of the charged sludge can be measured.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a system and method for producing magnesium ammonium phosphate (MAP). [Background technology]

[0002] In recent years, the worldwide depletion of phosphorus resources has brought renewed attention to the recovery of phosphorus from sewage, wastewater, human waste, etc. For example, biological phosphorus removal methods are widely used in sewage treatment. Biological phosphorus removal methods utilize the action of microorganisms called polyphosphate accumulating bacteria (PAOs), which absorb phosphorus from sewage into the bodies of PAOs in a reaction tank and remove phosphorus from the sewage by extracting it from the water treatment system as excess sludge. However, PAOs release the phosphorus they absorbed in the subsequent digester tank. In addition, a large amount of ammonia is generated with the decomposition of proteins in the sludge, so that digested sludge contains a large amount of phosphorus and ammonia. For this reason, many treatment plants have a problem with scaling of equipment and piping caused by MAP crystals generated from phosphorus, ammonia, and magnesium.

[0003] A technology has been developed in which phosphorus contained in sludge is removed and recovered as MAP crystals by adding a magnesium-containing chemical to sludge containing phosphorus and ammonia and adjusting the pH to an appropriate level. For example, JP 2004-941 A describes a method for recovering phosphorus as MAP crystals from wastewater containing high concentrations of organic matter, phosphorus, and nitrogen, such as sludge dehydration eluate from sewage or septic tank sludge, digested sludge liquid, and chemical industry wastewater. JP 2013-718 A describes a method for efficiently separating or recovering phosphorus components in digested sludge discharged from an anaerobic digestion tank by combining a phosphorus separation process and a sludge reduction process. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2004-941 A [Patent Document 2] JP 2013-718 A Summary of the Invention [Problem to be solved by the invention]

[0005] Both of the methods shown in Patent Documents 1 and 2 are useful in that they can recover not only phosphorus present as dissolved ions in sludge, but also MAP and the like that are naturally generated in digester tanks, etc. Since MAP crystals are recovered directly from digested sludge, etc., there is also the great advantage that MAP scale troubles in the downstream dewatering equipment can be reduced.

[0006] However, in Patent Document 1, the Mg supply amount and the amount of MAP seed crystals are controlled by treating the digestion tank itself as a MAP crystallization reaction tank. 3 In the large-capacity digester, the stirring power required to rapidly disperse the input Mg source and uniformly fluidize the MAP seed crystals is large.

[0007] In Patent Document 2, MAP particles in sewage sludge are separated using a physical classification device such as a liquid cyclone, but fine MAP particles with a particle size of 1 μm or less are difficult to separate easily. Therefore, in Patent Document 2, in the sludge reduction process / pH adjustment process, an acid such as dilute sulfuric acid is added to dissolve the fine MAP particles mixed in the sludge, and the dehydrated filtrate containing the dissolved solution is again subjected to MAP crystallization in a MAP reaction tank. However, in Japan at present, the Fertilizer Control Act does not permit the addition of a chemical to sewage sludge to adjust it to an acidic state, followed by a MAP production process, and the use of the produced MAP particles as MAP fertilizer. Therefore, when using MAP particles for purposes such as fertilizer, it is currently difficult to adopt this method.

[0008] Considering the phosphorus recovery technologies described in Patent Documents 1 and 2 and the current Fertilizer Control Law, a method of providing a small crystallization reaction tank for recovering MAP is considered to be an effective method for efficiently recovering MAP from organic wastewater or sludge. However, due to various factors, the properties of the organic wastewater or sludge fed into the crystallization reaction tank fluctuate, making it difficult to follow the reaction conditions of the crystallization reaction according to the change in the properties of the fed sludge.

[0009] In order to optimize the crystallization reaction in the crystallization reaction tank, there is a method of periodically sampling the crystallization reaction solution in the crystallization reaction tank and measuring the phosphorus or magnesium concentration in the crystallization reaction solution. However, organic wastewater or sludge containing phosphorus is often relatively viscous and muddy, and it takes time and effort to measure the concentration. In addition, by the time the concentration measurement result is obtained, further property changes may have occurred in the organic wastewater or sludge. In addition, phosphorus recovery from digested sludge and sludge dewatering as a subsequent process are generally performed as 24-hour continuous treatments, but it is particularly difficult to perform the labor of manually analyzing sludge components, especially at night, and to follow up on changes in sludge properties. Therefore, with such a method, it is difficult to quickly change the operating conditions of the crystallization reaction tank to changes in the properties of the input sludge, and phosphorus production efficiency may not be sufficiently improved.

[0010] In view of the above problems, the present invention provides a magnesium ammonium phosphate production system and a magnesium ammonium phosphate production method that can rapidly adapt the operating conditions of a crystallization reaction tank to conditions suitable for changes in the properties of input sludge, thereby stably improving phosphorus production efficiency.

[0011] As a result of intensive research into solving the above problems, the present inventors have found that it is effective to provide a measuring device capable of measuring fluctuations in the phosphorus input load of input sludge.

[0012] In order to solve the above problems, in one aspect, the present disclosure provides a magnesium ammonium phosphate production system including a reaction tank for crystallizing MAP particles in a crystallization reaction solution in which seed crystals of magnesium ammonium phosphate are present by mixing input sludge containing phosphorus with the crystallization reaction solution, a chemical supplying device for supplying a chemical containing magnesium into the reaction tank, an extraction device for extracting the crystallization reaction solution containing the MAP particles from the reaction tank, a separation device for separating the MAP particles from the crystallization reaction solution, a washing device for storing and washing the MAP particles separated by the separation device, and a measuring device capable of measuring fluctuations in the phosphorus input load of the input sludge.

[0013] In one embodiment of the magnesium ammonium phosphate production system according to the present invention, the measurement device includes a phosphate concentration measurement device that automatically measures the phosphate concentration of the input sludge.

[0014] In another embodiment of the magnesium ammonium phosphate production system of the present invention, the measuring device includes an image assessment device that measures the amount of MAP particles stored in the cleaning device by image-assessing the height of the interface of the MAP particle layer stored in the cleaning device.

[0015] In yet another embodiment of the magnesium ammonium phosphate production system of the present invention, the measuring device includes a weight measuring device that measures the amount of MAP particles stored in the cleaning device by measuring the weight of the cleaning device.

[0016] In yet another embodiment of the magnesium ammonium phosphate production system of the present invention, the measuring device includes an ultrasonic measuring device that measures the amount of MAP particles stored in the cleaning device by using ultrasound to measure the height of the interface of the MAP particle layer stored in the cleaning device.

[0017] In yet another embodiment of the magnesium ammonium phosphate production system of the present invention, the measuring device includes an image assessment device that image-assesssss the height of the interface of the MAP particle layer stored in the cleaning device, and a weight measuring device that measures the weight of the cleaning device, thereby measuring the amount of MAP particles stored in the cleaning device.

[0018] In yet another embodiment of the system for producing magnesium ammonium phosphate according to the present invention, the measuring device includes a pH change measuring device that measures a change in pH over a predetermined time in the reaction tank.

[0019] In yet another embodiment, the magnesium ammonium phosphate production system of the present invention further comprises a chemical supply control unit that controls the amount of chemical supplied from the chemical supply device based on the measurement results of the measurement device so that the Mg / P ratio of the crystallization reaction solution in the reaction tank is within an appropriate range.

[0020] In yet another embodiment, the system for producing magnesium ammonium phosphate according to the present invention further includes an extraction control unit that controls at least one of the amount or frequency of the crystallization reaction solution extracted by the extraction device based on the measurement results of the measurement device.

[0021] In yet another embodiment, the magnesium ammonium phosphate production system of the present invention further includes a cleaning control unit that adjusts at least one of the flow rate and time of the cleaning solution supplied into the cleaning device based on the amount of impurity layer accumulated on the MAP particle layer stored in the cleaning device.

[0022] In yet another embodiment, the magnesium ammonium phosphate production system of the present invention further comprises a transfer device that transfers MAP particles in the cleaning device to outside the cleaning device, and a transfer control unit that controls at least one of the amount or frequency of transfer of MAP particles by the transfer device based on the measurement results of a measuring device that measures the amount of MAP particles stored in the cleaning device.

[0023] In yet another embodiment, the system for producing magnesium ammonium phosphate according to the present invention further includes a seed crystal concentration prediction unit which predicts a future seed crystal concentration in the crystallization reaction solution in the reaction tank based on the pH in the reaction tank and the measurement results of the amount of MAP particles stored in the cleaning device, and an Mg / P ratio prediction unit which predicts a supply amount of chemicals so that the future Mg / P ratio of the crystallization reaction solution in the reaction tank will be within an appropriate range based on the prediction result of the seed crystal concentration prediction unit.

[0024] In yet another embodiment, the magnesium ammonium phosphate production system of the present invention includes an information acquisition unit that acquires MAP particle production plan information including information on the production amount of MAP particles, and an operating condition determination unit that determines at least any one of the operating conditions, i.e., the Mg / P ratio in the crystallization reaction solution, the amount or frequency of withdrawal of the crystallization reaction solution from the reaction tank, or the input sludge flow rate, based on the MAP particle production plan information.

[0025] In another aspect, the present invention is a method for producing magnesium ammonium phosphate, which includes the steps of mixing input sludge containing phosphorus in a reaction tank containing a crystallization reaction liquid in which seed crystals of magnesium ammonium phosphate are present, thereby crystallizing MAP particles in the crystallization reaction liquid, extracting the crystallization reaction liquid containing the MAP particles from the reaction tank to separate the MAP particles from the crystallization reaction liquid, introducing the separated MAP particles into a washing device for washing treatment, and transporting and recovering the MAP particles after washing treatment, and further includes measuring fluctuations in the phosphorus input load of the input sludge, and controlling at least one of the amount of magnesium-containing chemical to be supplied into the reaction tank, the amount or frequency of withdrawal of the crystallization reaction liquid from the reaction tank, based on the measurement results. Effect of the Invention

[0026] According to the present invention, it is possible to provide a magnesium ammonium phosphate production system and a magnesium ammonium phosphate production method that can rapidly adapt the operating conditions of a crystallization reaction tank to conditions suitable for changes in the properties of the input sludge, thereby stably improving phosphorus production efficiency. [Brief description of the drawings]

[0027] [Figure 1] FIG. 1 is a schematic diagram showing an example of a system for producing magnesium ammonium phosphate according to an embodiment of the present invention. [Diagram 2] 1 is a block diagram showing an example of a phosphoric acid concentration measuring device according to an embodiment of the present invention. [Diagram 3] 1 is a block diagram showing an example of an image assessment device according to an embodiment of the present invention. [Figure 4] FIG. 4(a) is a block diagram showing an example of a weight measuring device according to an embodiment of the present invention, and FIG. 4(b) is a block diagram showing an example of an ultrasonic measuring device according to an embodiment of the present invention. [Diagram 5] 2 is a block diagram showing an example of a calculation unit included in the pH change measuring device according to the embodiment of the present invention. FIG. [Figure 6] FIG. 1 is a schematic diagram showing a modified example of a magnesium ammonium phosphate production system according to a modified example of an embodiment of the present invention. [Figure 7] FIG. 2 is an explanatory diagram illustrating an internal state of the cleaning device according to the embodiment of the present invention when left standing; [Figure 8] FIG. 2 is a schematic diagram showing an example of a method for cleaning a phosphorus compound according to an embodiment of the present invention. [Figure 9] FIG. 2 is a schematic diagram showing an example of a method for cleaning a phosphorus compound according to an embodiment of the present invention. [Figure 10] FIG. 2 is a schematic diagram showing an example of a method for cleaning a phosphorus compound according to an embodiment of the present invention. [Figure 11] FIG. 2 is a schematic diagram showing an example of a method for cleaning a phosphorus compound according to an embodiment of the present invention. [Figure 12] FIG. 2 is a schematic diagram showing the composition ratio of a crystallization reaction solution containing phosphorus compound particles that is withdrawn from the bottom of a reaction tank via an extraction device and supplied to a separation device in an embodiment of the present invention. [Figure 13] FIG. 1 is a schematic diagram showing another modified example of a system for producing magnesium ammonium phosphate according to an embodiment of the present invention. [Figure 14]1 is an explanatory diagram illustrating a configuration of a production management system according to an embodiment of the present invention; [Figure 15] 11 is a flow diagram showing an example of a processing flow using a learning unit included in the production management system according to the embodiment of the present invention. FIG. [Figure 16] FIG. 11 is a flow diagram showing an example of a flow for creating a predetermined model through machine learning by a learning unit included in a production management system according to an embodiment of the present invention, and controlling operating conditions. [Figure 17] 4 is a graph showing the change over time in the phosphorus input load of input sludge input into the reaction tank according to the present embodiment. [Figure 18] 1 is a graph showing the change over time in the Mg / P ratio of the crystallization reaction solution in the reaction tank according to the present embodiment. [Figure 19] 1 is a graph showing the change over time in the seed crystal concentration in a reaction tank according to this embodiment. [Figure 20] 13 is a graph showing an example of the results of creating a calibration curve using the relationship between the indicated values ​​of each device and the weight of dried MAP placed in the cleaning device when an image processing method, an ultrasonic method, and a load cell are used as measuring devices. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0028] The following describes embodiments of the present invention with reference to the drawings. In the following description of the drawings, the same or similar parts are denoted by the same or similar reference numerals. The following embodiments are illustrative of an apparatus and method for embodying the technical idea of ​​the present invention, and the technical idea of ​​the present invention does not limit the structure, arrangement, etc. of the components to those described below.

[0029] As shown in FIG. 1, a magnesium ammonium phosphate production system according to an embodiment of the present invention includes a reaction tank 1, a chemical supplying device 2 that supplies a chemical containing magnesium into the reaction tank 1, an extraction device 3 that extracts a crystallization reaction liquid containing MAP particles from the reaction tank 1, a separation device 4 that separates the MAP particles from the crystallization reaction liquid, a washing device 5 that stores and washes the MAP particles separated by the separation device 4, and a measuring device 6 that measures fluctuations in the phosphorus input load of the input sludge.

[0030] The reaction tank 1 is a crystallization reaction tank in which input sludge containing phosphorus is mixed in a crystallization reaction solution in which seed crystals of magnesium ammonium phosphate are present, thereby crystallizing MAP particles in the crystallization reaction solution. A chemical containing magnesium ions is supplied to the reaction tank 1 via a chemical supply device 2 to react with phosphorus ions in the input sludge to form MAP particles.

[0031] The seed crystals of magnesium ammonium phosphate are particles that are mixed into the crystallization reaction solution to grow phosphorus in the input sludge into MAP particles of a specified particle size. The seed crystals act as a catalyst for the MAP reaction, creating an environment in which the MAP reaction, in which phosphate ions, ammonium ions, magnesium ions, etc. are bound to the surface of the seed crystals, is easy to proceed with. These substrates do not create new fine MAP nuclei, but the MAP reaction proceeds as if they are coating the existing MAP seed crystals. Although not limited to the following, in one embodiment, the seed crystals refer to crystals with a particle size of about 30 μm or more. The particle size of the seed crystals varies somewhat depending on the viscosity of the sludge, the separation performance of the liquid cyclone that separates the MAP, and the like. In general, the higher the viscosity, the worse the separation performance, so the particle size of the seed crystals needs to be larger, and the larger the diameter of the liquid cyclone, the higher the separation performance, so the particle size of the seed crystals may be smaller. The particle size of the seed crystals is determined by, for example, measuring the 50% diameter (D) of the particle size distribution measured using a laser diffraction / scattering particle size distribution analyzer (LA-910, manufactured by Horiba, Ltd.) and an ultrasonic sieve (Shodex PS microsieve, manufactured by Showa Denko K.K.). 50 The particle size of the seed crystals can be appropriately set by those skilled in the art depending on the shape, volume, operating conditions, etc. of the reaction vessel 1, and it is needless to say that it is not limited to the above example.

[0032] The pH of the crystallization reaction solution in the reaction tank 1 is measured by a pH meter 11 provided in the reaction tank 1. The coarse MAP particles that have grown to a size suitable for recovery in the reaction tank 1, i.e., a particle size equal to or larger than that of the seed crystals, are constantly flowing throughout the reaction tank 1 due to the turbulent flow caused by the agitator provided in the reaction tank 1, maximizing the opportunity for contact with the substrate components adhering to the MAP particles.

[0033] The crystallization reaction liquid suspension containing MAP particles dispersed throughout the reaction tank 1 is drawn out of the reaction tank 1 by the drawing device 3 equipped with piping and a pump 31 and supplied to the separation device 4. A part of the crystallization reaction liquid drawn out of the reaction tank 1 by the drawing device 3 is separated into MAP particles and separated sludge by a separation device 4b using centrifugal separation such as a liquid cyclone. The MAP particles separated by the separation device 4b are returned to the reaction tank 1. The remaining crystallization reaction liquid drawn out of the reaction tank 1 by the drawing device 3 is separated into MAP particles and separated sludge by the separation device 4a. The MAP particles separated by the separation device 4a are sent to the washing device 5. The separated sludge separated by the separation devices 4a and 4b is sent to the sludge treatment process, and the MAP concentrated sludge containing a large amount of MAP particles is washed by the washing device 5.

[0034] The cleaning device 5 includes a cleaning tank 50 for storing and cleaning the MAP particles separated in the separation device 4a, and a window 51 arranged on the outer wall surface of the cleaning tank 50 and enabling the MAP particles in the cleaning tank 50 to be identified from the outside of the cleaning tank 50. The window 51 may be provided with a scale 52 for measuring the height of the interface I of the MAP particles stored in the cleaning device 5. The window 51 may be provided with a cleaning mechanism (not shown) for preventing contamination. A pipe 53 for supplying a cleaning liquid (cleaning water) into the cleaning tank 50 is connected to the lower part of the cleaning tank 50. By supplying cleaning water from the lower part of the cleaning tank 50 through the pipe 53, an upward flow is generated in the cleaning tank 50, and impurities such as sludge and screen residue attached to the MAP particles are pushed toward the upper part of the cleaning tank 50. The MAP particles collected in the cleaning device 5 have a specific gravity of about 1.7 g / cm 3Since the MAP particles have a particle diameter of about several hundred μm, they generally have a higher settling tendency than organic matter and impurities in sludge, and can remain in the cleaning tank 50 without being washed out even in an upward cleaning operation at a specified flow rate. After a series of cleaning steps is completed and a specified period of time has passed, the inside of the cleaning tank 50 is filled with a MAP particle layer in which light white transparent MAP particles have been densely settled and accumulated from the bottom to a certain height, and an impurity layer on the MAP particle layer, and the upper part is filled with almost colorless and transparent tap water up to the level of the upper cleaning drain pipe. From the window 51 attached to the side of the cleaning tank 50, it is possible to visually confirm the interface between the MAP particle layer, the impurity layer, and the tap water formed inside the cleaning tank 50, and it is also possible to grasp the height position of the interface by the scale 52.

[0035] The impurities pushed out toward the top of the cleaning tank 50 are discharged together with the cleaning wastewater to the outside of the cleaning device 5 through a drain pipe 54 connected to the upper part of the cleaning tank 50. The MAP particles supplied to the cleaning device 5 are subjected to a predetermined cleaning process, then transferred to the outside of the cleaning device 5 through a transfer device 8 connected to the lower part of the cleaning device 5, and are dried by a dryer (not shown) or the like, and then collected.

[0036] The reaction tank 1 is connected to a chemical supplying device 2 equipped with a pump and piping for supplying a chemical into the reaction tank 1, and a sludge supplying device 10 for supplying input sludge into the reaction tank 1. The sludge supplying device 10 can be equipped with, for example, a storage tank, a pump, piping, etc. for storing the input sludge to be supplied into the reaction tank 1 and supplying it into the reaction tank 1, but is not limited to this example. The chemical supplying device 2 can be equipped with a tank for storing a magnesium-containing chemical such as magnesium hydroxide, a pump and piping for supplying the chemical from the tank, and a dilution tank for diluting the chemical as necessary, but is not limited to this example.

[0037] The sludge supplying device 10 and the chemical supplying device 2 are connected to a control device 7. The control device 7 is composed of a general-purpose calculator (computer) or the like. The control device 7 may be installed at the site where the reaction tank 1 is installed, or may be installed on a server or the like on a cloud. The control device 7 includes a calculation unit 70, a chemical supply control unit 71, and an extraction control unit 72.

[0038] The calculation unit 70 calculates various parameters necessary for the operation control of the reaction tank 1, the extraction device 3, the separation device 4, the washing device 5, the measurement device 6, and the transfer device 8, and outputs a predetermined control signal to the reaction tank 1, the extraction device 3, the separation device 4, the washing device 5, the measurement device 6, and the transfer device 8 based on the calculation result. The chemical supply control unit 71 outputs a control signal for controlling the supply flow rate and supply time of the chemical supplied by the chemical supply device 2 to the reaction tank 1 based on the input of the setting condition of the ratio of magnesium to phosphorus (Mg / P ratio) in the reaction tank 1. The withdrawal control unit 72 outputs a control signal for controlling the process for the withdrawal device 3 to withdraw the crystallization reaction liquid from the reaction tank 1, specifically, at least one of the withdrawal amount or withdrawal frequency of the crystallization reaction liquid, so that the seed crystal concentration in the reaction tank 1 is maintained within a certain range.

[0039] In a crystallization reaction, the process of nucleation (crystallization) caused by the presence of seed crystals in a solution is called "secondary nucleation". In order to stably proceed with the crystallization reaction in the secondary nucleation, it is important to ensure a sufficient surface area of ​​the seed crystals in the reaction tank 1 (seed crystal concentration). If the seed crystal concentration in the reaction tank 1 is low, the surface area of ​​the seed crystals that proceed with the crystallization reaction is insufficient, and the probability that phosphorus, which is a dissolved ion contained in the input sludge, will be a MAP reaction that generates new fine MAP rather than on the surface of the seed crystals, increases, so that separation by a separation device 4a such as a liquid cyclone is not possible, and the removal performance of fine nuclei deteriorates. If the seed crystal concentration is too high, too many MAP particles are generated in the reaction tank 1, which may clog the piping, pump, etc. provided in the extraction device 3. Here, the "seed crystal concentration" refers to the concentration of the seed crystals, that is, the concentration (g-MAP / L) of MAP particles having a particle size that functions as a seed crystal in the crystallization reaction. Although not limited to the following, in this embodiment, it refers to the concentration of MAP particles with a particle size of 30 μm or more present in the crystallization reaction liquid, for example.

[0040] In crystallization reactions, the surface area of ​​the seed crystals is generally considered to be one of the main operating factors rather than the seed crystal concentration. In this system, the seed crystals are withdrawn from the reaction tank 1 and returned at a predetermined frequency by the withdrawal device 3 and the separation device 4b, so that the particle size of the seed crystals in the reaction tank 1 can be kept constant. In this system, the surface area of ​​the seed crystals can be controlled by appropriately managing the concentration of the seed crystals in the reaction tank 1. In addition, the crystallization rate of the seed crystals can be controlled arbitrarily. However, the particle size of the seed crystals may vary depending on the characteristics of the equipment, the characteristics of the area, changes in the properties of the input sludge, etc. In such cases, the surface area of ​​the seed crystals and the seed crystal concentration are calculated taking into account the variation in the particle size of the seed crystals.

[0041] Currently, the seed crystal concentration in the reaction tank 1 is analyzed only by periodic manual analysis. In addition, the number and frequency of withdrawals of the crystallization reaction liquid containing MAP particles in the reaction tank 1 are also controlled at a constant rate at predetermined times. However, the phosphate (PO4-P) concentration of the input sludge varies depending on the type and property fluctuations of the input sludge. The magnesium concentration of the chemical input into the reaction tank 1 may also be excessive or insufficient depending on the property fluctuations of the input sludge, regional characteristics, temperature, etc. Due to the involvement of various factors such as these, it is currently difficult to constantly maintain the crystallization reaction in the reaction tank 1 in an optimal state under uniform operating conditions.

[0042] According to this embodiment, a measuring device 6 capable of directly or indirectly measuring the fluctuation of the phosphorus input load of the input sludge is provided. Depending on the measurement results of this measuring device 6, the extraction process of MAP particles from the reaction tank 1 and the supply process of a chemical agent containing magnesium required for crystallization are controlled so that the seed crystal concentration in the reaction tank 1 is within an appropriate range. This makes it possible to quickly adjust the operating conditions in the reaction tank 1 to conditions suitable for the change in the properties of the input sludge, and thus makes it possible to stably improve the phosphorus production efficiency.

[0043] As the measuring device 6 for directly or indirectly measuring the fluctuation of the phosphorus input load of the input sludge, various measuring devices can be used. Some examples are described below.

[0044] (Phosphate concentration measuring device) As the measuring device 6 for directly measuring the fluctuation of the phosphorus input load of the input sludge, it is preferable to provide a phosphorus concentration measuring device 6a for automatically measuring the phosphorus concentration (PO4-P) of the input sludge. As shown in Fig. 2, the phosphorus concentration measuring device 6a includes, for example, a sampling unit 61a for automatically sampling the input sludge at regular intervals, a solid-liquid separation unit 62a for filtering or membrane separating the sampled input sludge, a dilution unit 63a for diluting the separated liquid after solid-liquid separation with dilution water such as water, a phosphorus concentration measuring unit 64a for measuring the PO4-P concentration in the diluted liquid, and an input load calculation unit 65a for calculating the phosphorus input load (kg-P / day) of the input sludge.

[0045] The input load calculation unit 65a is, for example, a calculation result of the PO4-P concentration by the phosphoric acid concentration measurement unit 64a and the input flow rate (m 3 The phosphorus input load of the input sludge (kg-P / day) is calculated by multiplying this by the phosphorus input load of the input sludge (kg-P / day). The input load calculation unit 65a calculates the phosphorus input load of the input sludge at regular intervals, so that fluctuations in the phosphorus input load of the input sludge can be grasped. Note that the input load calculation unit 65a does not necessarily have to be disposed within the phosphate concentration measuring device 6a. The calculation processing of the input load calculation unit 65a may be performed by a calculation unit 70 provided in the control device 7, or may be performed by a server on the cloud.

[0046] According to this embodiment, by providing the phosphate concentration measuring device 6a, the phosphorus input load of the input sludge can be automatically measured at any time interval, so that fluctuations in the phosphorus input load can be measured without bothering the workers, even on holidays or at night when there are few workers.

[0047] When the phosphate concentration measuring device 6a is used as the measuring device 6, the amount of MAP produced in the reaction tank 1 can also be estimated from the phosphorus input load (kg-P / day) of the input sludge. Based on the estimation result of the amount of MAP produced, the chemical supply control unit 71 in Fig. 1 automatically controls the supply amount of chemicals supplied from the chemical supply device 2 so that the Mg / P ratio of the crystallization reaction solution in the reaction tank 1 falls within an appropriate range, thereby making it possible to maintain the seed crystal concentration in the reaction tank 1 within an appropriate range. A specific example of chemical supply control by the chemical supply control unit 71 will be described later.

[0048] (Image assessment device) As the measuring device 6 that indirectly measures the fluctuation in the phosphorus input load of the input sludge, an image assessment device 6b that measures the amount of MAP particles stored in the cleaning device 5 by image-assessing the height of the interface of the MAP particle layer that mainly contains MAP particles stored in the cleaning device 5 is preferably used. The image assessment device 6b performs image assessment using an image that captures the height of the interface of the MAP particle layer stored in the cleaning device 5 through a window portion 51 provided in the cleaning device 5, which allows the MAP particles stored in the cleaning device 5 to be identified from outside the cleaning device 5.

[0049] The crystallization reaction liquid withdrawn from the reaction tank 1 is separated into MAP particles and separated sludge by centrifugation or the like in the separator 4a. The MAP particles are introduced into the washing device 5 for washing to wash off impurities such as sludge and residue on the surface. Here, when the amount and frequency of the crystallization reaction liquid withdrawn from the reaction tank 1 are constant, if the phosphorus input load of the input sludge input into the reaction tank 1 fluctuates, the amount of MAP particles produced in the reaction tank 1 fluctuates. As a result, the amount of MAP particles in the crystallization reaction liquid withdrawn from the reaction tank 1 also fluctuates, and the amount of MAP particles stored in the washing device 5 also changes. In general, when PO4-P derived from raw sludge is crystallized to cause a MAP reaction, the "Mg / P ratio", which is the molar ratio of Mg added to P that can become MAP, is set to 1.0 to 1.4, and Mg is often set to a slightly excessive amount. In this case, when the PO4-P in the raw sludge increases, the amount of MAP generated by reacting with a slightly excess amount of Mg increases accordingly, and when the PO4-P decreases, the amount of MAP generated decreases. That is, the increase or decrease in the amount of MAP particles stored in the cleaning device 5 takes a value that reflects the PO4-P derived from the input sludge. Therefore, by measuring the transition of the amount of MAP particles stored in the cleaning device 5 by image judgment using the image judgment device 6b, the fluctuation of the phosphorus input load of the input sludge can be indirectly measured. As a result, the fluctuation of the phosphorus input load of the input sludge can be measured in a short time without manual analysis, compared to the case of analyzing the phosphate concentration of the input sludge.

[0050] As shown in Fig. 3, the image assessment device 6b includes an image capture unit 61b that captures still images or continuous images (moving images), and a processing unit 60b that performs predetermined arithmetic processing using image data from the image capture unit 61b. The processing unit 60b is configured with a general-purpose computer or the like. The processing unit 60b may include a graphic processor (GPU). A storage device (not shown) for storing the arithmetic results of the processing unit 60b and various information required for the arithmetic processing of the processing unit 60b may be located inside or outside the processing unit 60b.

[0051] The arithmetic processing by the processing unit 60b may be performed in the control unit 7 or a server on a cloud (not shown). Although not shown, the processing unit 60b may include an input / output control device (not shown) and may receive input / output of various information such as image data of the height of the interface of the MAP particle layer in the cleaning device 5 from outside the processing unit 60b or a learned model for image judgment using machine learning via the input / output control device. The processing unit 60b may output the processing result of the processing unit 60b to the outside of this system via the input / output control device. The processing unit 60b may further include a display device (not shown) for displaying the arithmetic processing result of the processing unit 60b to the operator.

[0052] The photographing unit 61b is not particularly limited as long as it is a device capable of photographing still or continuous images of the window 51 and the scale 52 of the cleaning device 5. For example, a camera equipped with an imaging element such as a CCD image sensor or a CMOS image sensor is suitable as the photographing unit 61b. The photographing unit 61b may be installed anywhere as long as the interface of the MAP particles in the window 51 can be photographed from outside the cleaning device 5, and the number of the photographing units 61b may be one or more. Typically, the photographing unit 61b is disposed so as to face the window 51 as shown in FIG. 1.

[0053] The photographing unit 61b may further include a light (not shown) for irradiating light around the window portion 51, which is the photographing area. The photographing unit 61b may include a cover (not shown) for maintaining a constant photographing environment. The shutter speed of the photographing unit 61b can also be adjusted as appropriate. The image data of the photographing unit 61b may be color or monochrome. When binarization processing is performed on the image data, monochrome is more preferable, but color may also be used. The lens provided in the photographing unit 61b can be selected arbitrarily. To prevent the lens from becoming dirty, a transparent plate or filter may be placed between the camera and the lens.

[0054] The processing unit 60b includes an image determination unit 62b that acquires image data including the window portion 51 of the cleaning device 5 photographed by the photographing unit 61b and determines by image determination the height of the interface of the MAP particle layer stored in the cleaning device 5 identified from the image data, and a storage amount calculation unit 63b that calculates the storage amount of MAP particles in the cleaning device 5 based on the image determination result of the image determination unit 62b. The processing unit 60b may further include a learning unit 64b that creates a trained model that outputs a predetermined explanatory variable in response to an input of a predetermined objective variable using a machine learning algorithm.

[0055] The image determination unit 62b is not limited to a specific operation procedure as long as it can detect the height of the interface of the MAP particle layer in the cleaning device 5. For example, the image determination unit 62b may perform various processing processes including trimming, black and white conversion, binarization, multi-value conversion, standardization, compression processing, etc. of the acquired image data in order to improve the identification of the height of the interface of the MAP particle layer in the cleaning device 5. The image determination unit 62b may determine the change in the height of the interface of the MAP particle layer in the cleaning device 5 in more detail by combining multiple image data captured by multiple shooting units 61b or combining multiple image data captured at different times by the shooting unit 61b. The image determination unit 62b may divide the acquired image data into several objects and classify the divided objects by region to perform image judgment. For example, it is also preferable that the image determination unit 62b classifies the image data identified from the window unit 51 into a MAP particle layer mainly containing MAP particles, an impurity layer composed of impurities such as screen residue, and a cleaning liquid layer on the impurity layer by region, and image-judges the height of the interface of the MAP particle layer in the cleaning device 5 based on the classification result.

[0056] The image determination unit 62b preferably performs image determination processing using a machine learning algorithm. For example, the image determination unit 62b may be configured to perform calculation processing using a machine learning algorithm using a model created using learning data that at least associates image data including information on the height of the interface of the MAP particle layer with information on the measurement result of the height of the interface of the MAP particle layer, and calculate the height of the interface of the MAP particle layer in the cleaning device 5. As the machine learning algorithm, any algorithm such as a support vector regression method (SVR method), a partial least squares method (PLS method), a random forest method, a decision tree method, etc. may be used. The learning data may further include information such as the seed crystal concentration in the reaction tank 1, the Mg / P ratio of the agent supplied to the reaction tank 1, the phosphorus input load of the input sludge, the extraction amount and the extraction frequency by the extraction device 3, etc.

[0057] The storage amount calculation unit 63b calculates the amount of MAP particles stored in the cleaning device 5 based on the image determination result of the height of the interface of the MAP particle layer in the cleaning device 5 by the image determination unit 62b. The storage amount calculation unit 63b calculates the volume of the MAP particles in the cleaning device 5, for example, from the measurement result of the height of the interface of the MAP particle layer in the cleaning device 5 and the capacity of the cleaning device 5. The storage amount calculation unit 63b further calculates the storage amount (weight) of MAP particles in the cleaning device 5 from the calculation result of the volume of the MAP particles using a known calibration curve (volume of MAP particles-weight of MAP particles) prepared in advance.

[0058] The storage amount calculation unit 63b may estimate the storage amount of MAP particles in the cleaning device 5 by machine learning using a predetermined model for receiving at least the input of information on the image determination result of the height of the interface of the MAP particle layer as an explanatory variable and outputting at least the estimated result of the storage amount of MAP particles in the cleaning device 5 as a target variable. As the machine learning algorithm, as described above, any algorithm such as a support vector regression method (SVR method), a partial least squares method (PLS method), a random forest method, a decision tree method, etc. can be used. Among them, deep learning can be preferably used, but is not limited to this. The machine learning may be supervised learning or unsupervised learning. The trained model constructed by machine learning may be updated periodically or at any timing. There may be one or more trained models. The training data may include the operation data of the present system in addition to the image data.

[0059] Alternatively, the image determination device 6b may estimate the seed crystal concentration in the reaction tank 1 by image-determining the height of the interface of the MAP particle layer in the cleaning device 5, and calculating the difference between the image determination result of the height of the interface of the MAP particle layer and the set value of the height of the interface of the MAP particle layer. Based on this estimation result, the withdrawal control unit 72 in FIG. 1 adjusts the withdrawal amount or withdrawal frequency of the crystallization reaction liquid containing MAP particles from the reaction tank 1 so that it is appropriate, thereby making it possible to automatically manage the seed crystal concentration in the reaction tank 1. Furthermore, based on the measurement result of the difference from the set value of the interface of this MAP particle layer, it is also possible to estimate the amount of MAP generated in a certain period of time in the reaction tank 1. Based on the estimation result of the amount of MAP generated in the reaction tank 1, the phosphorus input load of the input sludge to be input into the reaction tank 1 is estimated, and if the difference between the set value of the phosphorus input load and the estimated value exceeds the set value, the set value of the phosphorus input load may be corrected so that the phosphorus input load in the input sludge is within an appropriate range.

[0060] The image determination device 6b uses image discrimination using a machine learning algorithm to identify the color inside the cleaning device 5, making it easy to distinguish between an area made up of a MAP particle layer containing many MAP particles, which are usually white and transparent, and an impurity layer containing impurities derived from plants such as sesame, which are usually black. This makes it possible to more accurately determine the height of the interface of the MAP particle layer.

[0061] (Weight measuring device) As another measuring device 6 for indirectly measuring the phosphorus input load of the input sludge, a weight measuring device 6c is suitable which measures the amount of MAP particles stored in the cleaning device 5 by measuring the weight of the cleaning device 5.

[0062] As shown in Fig. 4(a), the weight measuring device 6c includes a weight measuring unit 61c that measures the weight of the cleaning device 5, and a storage amount calculation unit 63c that calculates the storage amount of MAP particles stored in the cleaning device 5 from the weight measurement result of the weight measuring unit 61c. The storage amount calculation unit 63c does not necessarily have to be disposed in the weight measuring device 6c. The calculation processing of the storage amount calculation unit 63c may be performed by a calculation unit 70 included in the control device 7, or may be performed by a server on the cloud.

[0063] The weight measuring unit 61c is not limited as long as it is a device capable of measuring the total weight inside the cleaning device 5, but it is preferable to use a load cell that can more easily detect a change in load over time. The storage volume calculating unit 63c can measure the change in weight of the cleaning device 5 before and after the extraction process of the crystallization reaction liquid from the reaction tank 1 by detecting a change in the electric signal output from the load cell.

[0064] The specific gravity of water is approximately 1.0 g / cm 3 The specific gravity of MAP is about 1.7 g / cm 3Therefore, the difference between the weight of the cleaning device 5 when it is filled with tap water and the weight of the cleaning device 5 when it is filled with MAP particles in the cleaning device 5 changes depending on the weight of the MAP particles in the cleaning device 5. Therefore, the storage amount calculation unit 63c can calculate the weight of the MAP particles stored in the cleaning device 5 from the relationship between the difference between the weight of the cleaning device 5 when it is filled with tap water and the weight of the cleaning device 5 when it is filled with MAP particles in the cleaning device 5, and the specific gravity of MAP. The storage amount calculation unit 63c may be configured to automatically calculate the weight of the MAP particles in the cleaning device 5 by using machine learning like the storage amount calculation unit 63b.

[0065] The weight measuring device 6c directly measures the weight of the cleaning device 5, and it is considered that the measurement error caused by impurities such as plant-derived contaminants having a low specific gravity is relatively small. In this way, the weight measuring device 6c automatically measures the change in the total weight before and after the crystallization reaction solution containing MAP particles is withdrawn from the reaction tank 1, thereby making it possible to grasp the increased weight of the MAP particles stored in the cleaning device 5. Taking into account the increased weight of the MAP particles, the withdrawal control unit 72 in Fig. 1 controls either the amount or frequency of withdrawal of the crystallization reaction solution containing MAP particles from the reaction tank 1, thereby controlling the seed crystal concentration in the reaction tank 1 to an appropriate range.

[0066] (Ultrasonic measuring device) As yet another measuring device 6 for indirectly measuring the phosphorus input load of the input sludge, an ultrasonic measuring device 6d is suitable which measures the amount of MAP particles stored in the cleaning device 5 by using ultrasonic waves to measure the height of the interface of the MAP particle layer stored in the cleaning device 5.

[0067] 4(b), the ultrasonic measurement device 6d includes an ultrasonic diagnostic unit 61d that measures the height of the interface of the MAP particle layer in the cleaning device 5, and a storage amount calculation unit 63d that calculates the storage amount of MAP particles stored in the cleaning device 5 from the ultrasonic diagnosis result of the ultrasonic diagnostic unit 61d. The storage amount calculation unit 63d does not necessarily have to be disposed in the ultrasonic measurement device 6d. The calculation processing of the storage amount calculation unit 63d may be performed by a calculation unit 70 included in the control device 7, or may be performed by a server on the cloud.

[0068] The specific configuration of the ultrasonic diagnostic unit 61d is not limited. For example, the ultrasonic diagnostic unit 61d may be an ultrasonic interface measuring device that measures the interface of a MAP particle layer containing MAP particles in the cleaning device 5 by utilizing the propagation of an ultrasonic signal through a liquid, or an ultrasonic sludge interface meter. In the example of Fig. 1, the tip of the ultrasonic diagnostic unit 61d is provided on the top of the cleaning device 5 so as to face the water surface of the cleaning device 5, but the present invention is not limited to this example.

[0069] The storage amount calculation unit 63d calculates the amount of MAP particles stored in the cleaning device 5 based on the ultrasonic diagnosis result of the interface height of the MAP particle layer in the cleaning device 5 by the ultrasonic diagnosis unit 61d. The storage amount calculation unit 63b calculates the volume of the MAP particles in the cleaning device 5 from, for example, the ultrasonic diagnosis of the interface height in the cleaning device 5 and the capacity of the cleaning device 5. The storage amount calculation unit 63d calculates the storage amount (weight) of the MAP particles in the cleaning device 5 from the calculation result of the volume of the MAP particles using a known calibration curve (volume of MAP particles-weight of MAP particles) prepared in advance. The storage amount calculation unit 63d may be configured to automatically calculate the weight of the MAP particles in the cleaning device 5 by using machine learning like the storage amount calculation unit 63b.

[0070] According to the ultrasonic measuring device 6d, for example, the seed crystal concentration in the reaction tank 1 can be estimated by calculating the difference between the ultrasonic measurement result of the interface height of the MAP particle layer in the cleaning device 5 and the set value of the interface height. Based on this estimation result, the withdrawal control unit 72 in FIG. 1 adjusts the withdrawal amount or withdrawal frequency of the crystallization reaction liquid containing MAP particles from the reaction tank 1, thereby enabling automatic control of the management of the seed crystal concentration in the reaction tank 1. Furthermore, based on the measurement result of the difference from the set value of the interface of the MAP particles, the amount of MAP produced in a certain period of time in the reaction tank 1 is estimated, and based on this amount of MAP production, the phosphorus input load of the input sludge to be input into the reaction tank 1 is estimated. This may correct the set value of the phosphate ion concentration in the input sludge.

[0071] By combining two or more of the image assessment device 6b, weight measurement device 6c, and ultrasonic measurement device 6d as the measurement device 6, the amount of MAP particles stored in the cleaning device 5 can be measured with higher accuracy.

[0072] For example, the measuring device 6 may include an image determining device 6b that performs image determination of the height of the interface of the MAP particle layer stored in the cleaning device 5 through a window portion 51 provided in the cleaning device 5 that allows the MAP particles stored in the cleaning device 5 to be identified from the outside of the cleaning device 5, and a weight measuring device 6c that measures the weight of the cleaning device 5, and the amount of stored MAP particles may be measured by the image determining device 6b and the weight measuring device 6c. The measuring device 6 may include at least one of the image determining device 6b and the ultrasonic measuring device 6d, and the weight measuring device 6c.

[0073] The image assessment device 6b and ultrasonic measurement device 6d, which are capable of measuring the height of the interface of the MAP particle layer in the cleaning device 5, visually or ultrasonically measure the height of the interface between the MAP particle layer containing the solid MAP particles and the supernatant water. Then, the weight of the MAP particles in the capacity of the cleaning device 5 located below the height of the interface of the MAP particle layer is calculated from the specific gravity of the MAP particles, 1.7, the porosity of the layer containing the MAP particles in the cleaning device 5, the proportion of impurities mixed in, the amount of water in the cleaning device 5, the empty weight of the main body of the cleaning device 5, etc.

[0074] The image determination device 6b can also distinguish the difference between the color of the MAP layer, which is usually white and transparent, and the impurities derived from plants, such as sesame, which are usually black, by image analysis, so that the purity of the MAP particles contained in the MAP particle layer can be grasped by image analysis. The weight measuring device 6c directly measures the weight, so the influence of errors due to impurities derived from plants, which are impurities, is small, but when the sludge contains a lot of fine sand components (specific gravity 2.0 to 2.2) such as SiO2, there is a possibility that the sand components may be mistaken for MAP particles. Therefore, by combining two or more of the above-mentioned image determination device 6b, weight measuring device 6c, and ultrasonic measuring device 6d as the measuring device 6, for example, the image determination device 6b can accurately grasp the volumes of the MAP particle layer containing MAP particles that are settled and deposited in the cleaning device 5 and the impurity layer containing impurities, and the weight measuring device 6c can accurately grasp the total weight of solids in the cleaning device 5, so that the weight of the MAP particles can be more accurately grasped by calculation using the respective volumes of the impurity layer and the MAP particle layer and the total weight of solids.

[0075] (pH change measuring device) As yet another measuring device 6 for indirectly measuring the phosphorus loading amount of the input sludge, a pH change measuring device 6e for measuring the change in pH over a predetermined time in the reaction tank 1 may be provided. The pH change measuring device 6e is connected to a pH meter 11 for measuring the pH in the reaction tank 1. The pH meter 11 is installed at a sufficient distance from the position where the chemical is added by the chemical supply device 2. In order to maintain the measurement accuracy of the pH meter 11, an automatic cleaning mechanism for cleaning the pH meter 11 may be provided. The pH change measuring device 6e is preferably further measured by a pH meter 12 for measuring the pH of the input sludge.

[0076] The pH of the input sludge at pH measurement time t1 may show the following patterns when compared with the pH at measurement time t0, which is prior to measurement time t1. (A-1) The pH of input sludge tends to decrease (A-2) The pH of the input sludge tends to be maintained. (A-3) The pH of input sludge tends to increase

[0077] The pH in the reaction vessel 1 at pH measurement time t1 may have the following patterns when compared with the pH at measurement time t0 prior to measurement time t1. (B-1) The pH in reactor 1 tends to decrease (B-2) The pH in the reaction tank 1 tends to be maintained (B-3) The pH in reactor 1 tends to rise

[0078] When the actual amount of MAP particles stored in the washing device 5 (hereinafter also referred to as the "MAP recovery amount" or "MAP production amount") separated from the crystallization reaction liquid extracted in one extraction process by the extraction device 3 is compared with a predetermined theoretical MAP recovery value (hereinafter also referred to as the "assumed theoretical MAP recovery value" or simply the "theoretical value"), the following patterns are possible. (C-1) The amount of MAP recovered per extraction process is lower than the theoretical value. (C-2) The amount of MAP recovered from each extraction process is equivalent to the theoretical value. (C-3) The amount of MAP recovered from each extraction process exceeds the theoretical value.

[0079] In this embodiment, it is preferable to always maintain the "Mg / P ratio" and "MAP amount" in the reaction tank 1 at appropriate predetermined values ​​as operation control items in a manner that automatically follows changes in the properties of the input sludge. Here, the three values ​​of (1) "pH of input sludge", (2) "pH in the reaction tank 1", and (3) "difference between MAP recovery amount and theoretical value" in the cleaning device 5 contain factors that suggest changes in the sludge properties of the input sludge. By recording the changes over time in these three values ​​and having AI or the like learn them, the "Mg / P ratio" and the "MAP concentration in the reaction tank 1" can be utilized for more optimal control.

[0080] For example, (1) "pH of input sludge" is a value that indicates to some extent the amount of alkaline component in the input sludge required for MAP production. Here, the solubility product of the MAP reaction is expressed as [OH - If we rewrite the equation to include ], it becomes as follows. [Mg 2+ ][HPO4 2- ][NH4+ ][OH - ]=K sp (Solubility product)

[0081] The mechanism of MAP formation can be expressed in this way because the ionization state of each constituent ion changes with pH, ​​and the proportion of the constituent components changes. This solubility product K sp Since the MAP reaction proceeds so that it always approaches this value, for example, the left side of [HPO4 2- ] increases or the pH of the input sludge increases, this means that the MAP reaction progresses more than before and the amount of MAP produced increases.

[0082] Based on these assumptions, (2) "pH in Reactor 1" is the pH of [OH] in the reactor where the MAP reaction is actively taking place. - The value of "(2) - (1)" is a factor for determining whether the change in (2) is mainly due to the change in input sludge. Magnesium hydroxide, which is a Mg source supplied to the reaction tank 1, is 2+ :OH - Since this agent is supplied in a ratio of 1:2, it is supplied from the input sludge [HPO4 2- ] decreases, and [Mg 2+ ] and [OH - ] increases and remains unconsumed for MAP production, resulting in an increase in pH.

[0083] In addition, (3) “Difference between MAP recovery amount and theoretical value” also shows that [HPO4 2- In other words, if the trend of "MAP recovery amount > theoretical value" continues, the increase or decrease in [HPO4 2- ] is on an increasing trend, and in the opposite case it is on a decreasing trend. This indicates that with the settings unchanged, the actual Mg / P ratio will be lower than the target value, and therefore indicates that it is necessary to increase the supply of magnesium hydroxide to return the Mg / P ratio to the target value.

[0084] By inputting the time behavior of the values ​​of (1) to (3) and calculating the results in the control device 7, and reflecting the results in the magnesium hydroxide supply amount and the sludge extraction amount, it becomes possible to maintain the Mg / P ratio and the MAP amount in the reaction tank 1 at the target set values. Also, in some cases, a simplified control method using only the values ​​of (2) and (3) without using (1), or a machine learning control method in which (2) and (3) are input into a learning model as explanatory variables and the magnesium supply amount and the sludge extraction amount are output as objective variables, may be effective to a certain extent.

[0085] In the present invention, it is preferable to prepare a matrix table of (1) to (3) and pattern the results according to the increase or decrease in (1) to (3). The amount of change in the Mg supply amount and the amount of sludge extracted are appropriately set based on past data and optimization using AI learning, thereby making it possible to maintain the Mg / P ratio and the MAP amount in the reaction tank 1 as close as possible to the target set values.

[0086] That is, according to this embodiment, the chemical supply control unit 71 automatically controls the chemical supply by using a previously obtained calculation result or machine learning so that the Mg / P ratio of the crystallization reaction solution in the reaction tank is within an appropriate range based on three factors: (1) "pH of input sludge," (2) "pH in the reaction tank 1," and (3) "difference between the MAP recovery amount and the theoretical value," thereby making it possible to keep the seed crystal concentration in the reaction tank 1 constant. This makes it possible to optimize the crystallization reaction in the reaction tank 1, and to always stably recover MAP particles from the cleaning device 5.

[0087] The chemical supply control unit 71 may be configured to predict the future seed crystal concentration and Mg / P ratio in the reaction tank 1 by machine learning using learning data including at least information on the pH in the reaction tank 1, the seed crystal concentration in the reaction tank 1, the Mg / P ratio of the crystallization reaction solution in the reaction tank 1, and the phosphorus input load of the input sludge. Based on the prediction result, the chemical supply control unit 71 can also output a predetermined control signal to the chemical supply device 2 at a predetermined timing.

[0088] For example, the pH change measuring device 6e may include a calculation unit 60e for predicting the seed crystal concentration and Mg / P ratio in the future reaction tank 1. As shown in Fig. 5, the calculation unit 60e may include a seed crystal concentration prediction unit 61e for predicting the seed crystal concentration in the crystallization reaction solution in the future reaction tank 1 based on the pH in the reaction tank 1 and the measurement result of the MAP storage amount stored in the cleaning device 5, and an Mg / P ratio prediction unit 62e for predicting the future Mg / P ratio based on the prediction result of the seed crystal concentration prediction unit 61e and predicting the supply amount (supply flow rate) of the chemical so that the future Mg / P ratio of the crystallization reaction solution in the reaction tank 1 is within an appropriate range.

[0089] The seed crystal concentration prediction unit 61e can predict the seed crystal concentration based on a relational expression that is determined in advance based on the relationship between the pH measurement value in the reaction tank 1, the difference between the measurement value and the set value of the MAP recovery amount in the cleaning device 5, and the relationship between the set value of the seed crystal concentration in the reaction tank 1. Alternatively, the seed crystal concentration prediction unit 61e may predict the seed crystal concentration using a machine learning algorithm. For example, the seed crystal concentration prediction unit 61e may be configured to output a predicted value of the future seed crystal concentration by inputting the pH measured by the pH meter 11 and the MAP storage amount in the cleaning device 5 into a model constructed by machine learning using training data including at least information on past pH measurement values, past measurement values ​​of the MAP recovery amount in the cleaning device 5, and past measurement values ​​of the seed crystal concentration in the reaction tank 1.

[0090] The Mg / P ratio prediction unit 62e predicts the future Mg / P ratio and the supply amount of the chemical based on the seed crystal concentration prediction result of the seed crystal concentration prediction unit 61e and the set value or measured value of the phosphorus input load of the input sludge. The Mg / P ratio prediction unit 62e may predict the Mg / P ratio from a predetermined relational expression, as with the seed crystal concentration prediction unit 61e, or may predict the Mg / P ratio using a machine learning algorithm. The chemical supply control unit 71 outputs a control signal to the chemical supply device 2 for controlling the supply flow rate of the chemical to be supplied into the reaction tank 1 based on the prediction result of the Mg / P ratio. The pH change measurement device 6e may further include a correction unit 63e for correcting and resetting the set value of the Mg / P ratio or the set value of the MAP storage amount in the cleaning device 5 when the set value of the Mg / P ratio or the set value of the MAP storage amount in the cleaning device 5 is significantly different from the measured value or the expected value.

[0091] The chemical supply control unit 71 is preferably configured to control at least one of the supply amount, supply time, and supply frequency of the chemical supplied from the chemical supply device 2 so that the Mg / P ratio in the reaction tank 1 is maintained within a suitable range based on the pH measurement result of the pH meter 11 that measures the pH in the reaction tank 1, the pH measurement result of the sludge before it is introduced into the reaction tank 1, and the amount of MAP particles stored in the cleaning device 5.

[0092] When digested sludge generated by anaerobic digestion of organic matter is used as the input sludge, the pH of the digested sludge itself may fluctuate greatly. One possible reason is that the pH fluctuation of the digested sludge is not linked to the phosphorus input load of the input sludge. By further taking into account the pH measurement information of the input sludge before it is input to the reaction tank 1 as input information, the chemical supply control unit 71 can further improve the accuracy of automatic control by linking the pH measurement results of the input sludge and the pH measurement results in the reaction tank 1 to determine whether the change in pH in the reaction tank 1 is the result of being dragged by the change in pH of the input sludge, whether the change in pH of the input sludge is the result of an increase or decrease in the phosphoric acid of the input sludge without a change in pH of the input sludge, or whether the change in pH in the reaction tank 1 is due to an excess or deficiency of the Mg / P ratio when the chemical is added to the reaction tank 1.

[0093] (Adjustment of extraction amount and / or extraction frequency) In order to stably carry out the crystallization reaction in the reaction tank 1, it is preferable to always stably maintain a constant concentration of seed crystals in the crystallization reaction solution in the reaction tank 1. In order to control the seed crystal concentration in the crystallization reaction solution in the reaction tank 1 to be within a constant concentration range even if the phosphorus input load of the input sludge fluctuates, it is preferable to pay attention to the following variables. (a) Actual amount of MAP particles produced in reactor 1 at a given time (=Actual amount of MAP particles stored in the cleaning device 5) (b) Seed crystal concentration in reactor 1 (c) Seed crystal production rate in reaction vessel 1 (amount of seed crystals produced per hour) (d) Phosphorus input load of input sludge Taking these factors (a) to (d) into consideration, it is preferable to adjust the amount or frequency of the MAP particles extracted from the reaction tank 1 via the extractor 3.

[0094] Specifically, the amount (or time) of the crystallization reaction liquid to be withdrawn from the reaction tank 1 or the withdrawal interval is adjusted based on the following procedure. For example, when the phosphorus input load of the input sludge can be measured, the amount of MAP particles produced in the reaction tank 1 is estimated from the phosphorus input load of the input sludge and the supply flow rate of the input sludge to the reaction tank 1. The interval is adjusted so that the crystallization reaction liquid containing the seed crystals is withdrawn from the reaction tank 1 when the amount of seed crystals produced in the reaction tank 1 reaches a set value.

[0095] From the measurement results of the image determination device 6b, the weight measurement device 6c, and the ultrasonic measurement device 6d, the actual amount of MAP particles produced in the reaction tank 1 and the time required for production (or the amount of sludge input) can be calculated. Assuming that the same amount of seed crystals as that of the previous seed crystal extraction will be extracted next time, the recovery interval of the crystallization reaction liquid by the extraction device 3 is adjusted based on the amount of MAP particles stored in the cleaning device 5 measured by the image determination device 6b, the weight measurement device 6c, and the ultrasonic measurement device 6d. For example, if the previous extraction frequency setting was every 4 hours, the extraction amount was 100 kg, and the current amount of MAP particles stored in the cleaning device 5 is 120 kg, the extraction control unit 72 can control the extraction device 3 so that the next extraction interval by the extraction device 3 is "extraction frequency setting (4 hours) x (100 / 120)".

[0096] The withdrawal control unit 72 can control the amount of reaction crystallization liquid withdrawn from the reaction tank 1, for example, as follows. For example, if the previous withdrawal frequency was every 4 hours and the set value of the withdrawal amount was 100 kg, and the current amount of MAP particles stored in the washing device 5 is 96 kg, the withdrawal control unit 72 can control the withdrawal device 3 so that the next amount withdrawn by the withdrawal device 3 is "extraction amount (100 kg) x (100 / 96)".

[0097] It is also preferable that the withdrawal control unit 72 calculates the seed crystal concentration in the reaction tank 1 based on the phosphorus input load of the input sludge measured by the phosphorus concentration measuring device 6a, and determines optimal values ​​of the amount and frequency of withdrawal of the crystallization reaction liquid from the reaction tank 1 so that the seed crystal concentration in the reaction tank 1 is within a predetermined range based on the calculation result of the seed crystal concentration, and controls the withdrawal device 3 based on the optimal values. Alternatively, it is also preferable that the withdrawal control unit 72 determines optimal values ​​of the amount and frequency of withdrawal of the crystallization reaction liquid from the reaction tank 1 based on fluctuations in the seed crystal concentration in the reaction tank 1 estimated from the image determination device 6b, the weight measuring device 6c, and the ultrasonic measuring device 6d, and controls the withdrawal device 3 based on the optimal values.

[0098] According to this embodiment, the withdrawal control unit 72 includes the withdrawal control unit 72 that controls at least one of the withdrawal amount or withdrawal frequency of the crystallization reaction liquid by the withdrawal device 3 based on the measurement results of the measurement device 6. This makes it possible to adjust the seed crystal concentration in the reaction tank 1 to an optimal range even when the load of the input sludge fluctuates, and therefore makes it possible to quickly adapt to conditions suitable for changes in the properties of the input sludge, thereby making it possible to stably improve the phosphorus production efficiency.

[0099] (Cleaning liquid supply control for cleaning device 5) As shown in FIG. 6, the magnesium ammonium phosphate production system according to the embodiment of the present invention includes a cleaning control unit 73 that controls the supply of cleaning water to a cleaning device 5 that cleans the MAP particles recovered from a reaction tank 1 that performs a crystallization process.

[0100] The MAP particles separated from the separator 4a have impurities such as sludge attached to their surfaces. Most of the impurities fed into the cleaner 5 together with the MAP particles have a specific gravity slightly lower than that of the seed crystals. As shown in Fig. 7, the cleaner 5 is left stationary in a state where a certain amount of the MAP particle layer is always maintained before and after the MAP particles are extracted from the reaction tank 1, and during the stationary state, an impurity layer with a specific gravity lower than that of the MAP particles is deposited on top of the MAP particle layer.

[0101] The inside of the cleaning device 5 is adjusted so that an appropriate amount of MAP particles is always held. If the amount of MAP particles and the amount of impurities in the cleaning device 5 increase due to a sudden change in the supply amount of MAP particles, it may become difficult to hold an appropriate amount of MAP particles in the cleaning device 5. If the amount of MAP particles held in the cleaning device 5 exceeds the appropriate amount and becomes excessive, consolidation of the MAP particles may occur in the cleaning device 5. If consolidation of the MAP particles occurs in the cleaning device 5, the cleaning liquid may not be sufficiently distributed throughout the cleaning device 5, and the MAP particles may not be properly cleaned. On the other hand, if the amount of MAP particles supplied to the cleaning device 5 becomes too small due to a sudden change in the supply amount of MAP particles, it may become difficult to supply only the MAP particles that have been sufficiently cleaned by the transfer device 8 to the drying equipment outside the cleaning device 5, and the purity of the MAP particles may decrease.

[0102] As shown in Fig. 7, the cleaning device 5 is provided with detectors 58b, 58c, and 58d for detecting the amount of the impurity layer formed on the upper layer of the MAP particle layer stored in the cleaning device 5, and the cleaning control unit 73 preferably controls at least one of the flow rate and time of the cleaning liquid supplied into the cleaning device 5 based on the detection result of the amount of the impurity layer detected by at least one of the detectors 58b, 58c, and 58d. In this embodiment, by providing the cleaning control unit 73, even if the supply amount of MAP particles from the reaction tank 1 fluctuates, it is possible to always hold an appropriate amount of MAP particles in the cleaning device 5 and sufficiently clean the MAP particles stored in the cleaning device 5, so that the purity of the MAP particles transferred out of the cleaning device 5 can be improved.

[0103] As the detection devices 58b, 58c, and 58d in FIG. 7, any detection device can be used as long as it is capable of detecting the height of the interface of the impurity layer in the cleaning device 5, or the volume or weight of the impurity layer. For example, as the detection device 58b, an image determination device 6b that is provided with a photographing unit that photographs the MAP particle layer and the impurity layer inside the cleaning device 5 from the outside of the cleaning device 5 through the window portion 51 provided in the cleaning device 5, and measures the amount of the impurity layer stored in the cleaning device 5 by image-determining the layer thickness or interface height of the impurity layer stored in the cleaning device 5 from the image photographed by the photographing unit can be suitably used. Note that, as the detection device 58b, it is more preferable to use the image determination device 6b in FIG. 1 in terms of reducing the number of devices, but this example is not limited to this example.

[0104] As the detection device 58c, a weight measuring device 6c that measures the amount of the impurity layer stored in the cleaning device 5 by measuring the weight of the cleaning device 5 can be preferably used. As the detection device 58c, it is more preferable to use the measuring device 6c of Fig. 1 in terms of reducing the number of devices. As the detection device 58d, an ultrasonic measuring device 6d that measures the amount of the impurity layer stored in the cleaning device 5 by measuring the layer thickness or interface height of the impurity layer stored in the cleaning device 5 using ultrasonic waves can be preferably used. As the detection device 58d, it is more preferable to use the measuring device 6d of Fig. 1 in terms of reducing the number of devices.

[0105] Impurities that may enter the cleaning device 5 include impurities with relatively dark colors that are easily distinguishable from the outside, such as sesame seeds and wood chips. On the other hand, MAP particles that have been thoroughly cleaned in the cleaning device 5 are white in color. According to this embodiment, by using the detection device 58b that can image-judge the state of the impurity layer from images of the MAP particle layer and the impurity layer inside the cleaning device 5, the impurity layer and the MAP particle layer can be detected more precisely and efficiently.

[0106] When the amount of impurities accumulated in the cleaning device 5 becomes large, the cleaning control unit 73 utilizes the specific gravity difference between the MAP layer and the impurity layer to change the supply flow rate of the cleaning liquid and control at least one of the flow rate or time of the cleaning liquid supplied into the cleaning device 5 so as to preferentially discharge impurities with low specific gravity.

[0107] Specifically, as shown in FIG. 7, the cleaning device 5 preferably includes at least two pipes, a first water supply pipe 55 and a second water supply pipe 56 having a larger pipe diameter than the first water supply pipe 55 and capable of flowing a cleaning liquid at a higher flow rate than the first water supply pipe 55. The first water supply pipe 55 and the second water supply pipe 56 are connected to a pipe 53 that supplies the cleaning liquid to the cleaning device 5. When cleaning the MAP particles in the cleaning device 5, for example, the cleaning liquid is supplied from the first water supply pipe 55, and the cleaning process is performed at a first water flow rate and a first water flow time, and then the cleaning liquid is left to stand. Next, the detection devices 58b, 58c, and 58d detect the height of the interface of the impurity layer in the cleaning device 5 and the amount of the MAP particle layer. As a result, when the amount of the impurity layer in the cleaning device 5 exceeds a predetermined value, cleaning is performed for a predetermined time at a flow rate (higher water flow rate) than the first water flow rate during cleaning. This separates impurities from the MAP particle layer, and the separated impurities are swept further upward, and further swept out of the cleaning device 5 through the drain pipe 54. When discharging the impurities, for example, the second water supply pipe 56 is used, or when the flow rate is insufficient through the second water supply pipe 56 alone, the first water supply pipe 55 and the second water supply pipe 56 are used to supply the cleaning liquid into the cleaning device 5. After the impurities are discharged out of the cleaning device 5, the supply of the cleaning liquid from the second water supply pipe 56 is stopped, and the cleaning liquid is run through the first water supply pipe 55 to thoroughly wash the MAP particles. The MAP particles after the washing process are transferred to the outside of the cleaning device 5 through the transfer device 8 and collected.

[0108] The washing device 5 typically uses an air lift pump as a transfer device 8 that transfers the MAP particles in the washing device 5 to the outside of the washing device 5. For this reason, it is preferable to keep the interface height of the MAP particle layer in the washing device 5 constant. However, if there is an increase or decrease in the amount of MAP particles produced in the reaction tank 1, the interface height of the MAP particle layer in the washing device 5 may fall outside a certain range, making it impossible to efficiently transfer the MAP particles using the air lift pump.

[0109] According to this embodiment, the cleaning control unit 73 changes, for example, the flow rate and time of the cleaning liquid in the cleaning device 5 in stages, and when impurities are pushed upward, a booster cleaning is performed in which a cleaning liquid at a larger flow rate than that during normal cleaning processing is supplied from the lower part of the cleaning device 5. In this way, the cleaning control unit 73 controls the flow rate and time of the cleaning water in multiple stages, making it possible to simultaneously discharge impurities from the cleaning device 5 and clean the MAP particles.

[0110] 8(a) is a schematic diagram showing the internal state of the cleaning device 5 in a stationary state (cleaning standby state) before the cleaning process. A certain amount of MAP particles is always retained in the cleaning device 5 (hereinafter, the MAP retained in advance in the cleaning device 5 is referred to as "retained MAP"). An impurity layer made of impurities having a lower specific gravity than MAP is deposited on the retained MAP layer made of the retained MAP.

[0111] As shown in Fig. 8(b), when new MAP particles (new MAP) containing impurities are supplied from the separator 4a into the cleaning device 5, the new MAP particles are deposited on the impurity layer (supply process). After the new MAP particles are supplied, a small amount of MAP remains in the separator 4a, so in Fig. 8(c), the separator 4a is washed with water. The small amount of MAP washed away with water and impurities such as sludge are supplied into the cleaning device 5.

[0112] 9(a), cleaning water is supplied from the lower part of the cleaning device 5 at a first water flow rate capable of cleaning the MAP particles in the cleaning device 5, and the retained MAP and impurities previously stored in the cleaning device 5 are mixed with the new MAP newly supplied to the cleaning device 5 for cleaning (tank cleaning process). The cleaning control unit 73 opens the water supply valve of the first water supply pipe 55, and supplies cleaning water from the lower part of the cleaning device 5 at a first water flow rate and for a first water flow time.

[0113] The first water flow rate is, for example, preferably 0.2 to 0.5 m / min, more preferably 0.25 to 0.4 m / min, and even more preferably 0.30 to 0.35 m / min. The supply flow rate of the washing water into the washing device 5 at this time is, for example, preferably 75 to 195 L / min, more preferably 90 to 155 L / min, and even more preferably 110 to 130 L / min. The first water flow time is not particularly limited as long as it is possible to ensure that the new MAP supplied into the washing device 5 is not discharged outside the washing device 5 and that the impurities can be pushed upward to a height required for specific gravity separation from the new MAP particles. Typically, the first water flow time can be about 1 to 10 minutes, preferably about 1 to 5 minutes, more preferably 2 to 4 minutes, and even more preferably about 3 minutes. The supply amount of new MAP into the washing device 5 and the composition of impurities vary depending on the region and treatment plant in which the washing device 5 is installed. Therefore, it goes without saying that the first water flow rate, the supply flow rate, and the first water flow time are not limited to the above specific examples.

[0114] After the tank cleaning step shown in Fig. 9(a), it is preferable to further provide a settling step in which the supply of cleaning water is stopped and the mixture is settling as shown in Fig. 9(b). By setting the mixture after the tank cleaning step, it is possible to promote gravity separation of the retained MAP, new MAP, and impurities stirred in the cleaning device 5. In addition, the settling step allows the impurities to settle and consolidate as quickly as possible, which makes it possible to efficiently discharge the impurities in the impurity discharge step described below. After the settling step, as shown in Fig. 9(b), the MAP consisting of the retained MAP and new MAP settles to the bottom of the cleaning device 5, and impurities accumulate on the MAP.

[0115] The settling step in Fig. 9(b) may be omitted. For example, in the tank cleaning step in Fig. 9(a), after mixing and cleaning of the new MAP, the retained MAP, and impurities at a first water flow rate, water flow may be continued while gradually decreasing the water flow rate to an extent that allows gravity separation of the MAP and the impurities, and after water flow is completed, separation of the MAP and the impurities as shown in Fig. 9(b) may be performed. Alternatively, in the tank cleaning step, mixing and cleaning of the new MAP, the retained MAP, and impurities may be performed, and a water flow rate may be set such that gravity separation of the MAP and the impurities is possible.

[0116] After the tank cleaning step, as shown in FIG. 9(c), cleaning water is supplied from the lower part of the cleaning device 5 at a second water flow rate higher than the first water flow rate. This causes impurities in the cleaning device 5 to be flushed to the upper part of the cleaning device 5, and the overflow water containing impurities is made to overflow from the upper part of the cleaning device 5 through the drain pipe 54 and discharged to the outside of the cleaning device 5 (impurity discharge step). The impurity discharge step is also intended to suppress the consolidation of MAP held at the bottom of the cleaning device 5, including the discharge of impurities in the cleaning device 5. Therefore, it is necessary to make the second water flow rate higher than the first water flow rate. The second water flow rate is preferably 1.25 to 3.0 times the first water flow rate, more preferably 1.5 to 2 times, and even more preferably 1.6 to 1.8 times.

[0117] The second water flow rate is, for example, preferably 0.25 to 1.5 m / min, more preferably 0.3 to 0.8 m / min, and even more preferably 0.50 to 0.55 m / min. The supply flow rate is, for example, preferably 95 to 400 L / min, more preferably 115 to 310 L / min, and even more preferably 190 to 215 L / min.

[0118] If the second water flow time is set too long, the MAP in the cleaning device 5 may also be swept to the upper part of the cleaning device 5 and leak from the upper part of the cleaning device 5 to the outside of the cleaning device 5. It is preferable to adjust the second water flow speed appropriately according to the amount of impurities accumulated on the MAP. The second water flow time can typically be 1 to 15 minutes, preferably 1 to 10 minutes, and more preferably 1 to 7 minutes. As with the first water flow speed, supply flow rate, and first water flow time, the second water flow speed, supply flow rate, and second water flow time are of course not limited to the above specific examples.

[0119] In the impurity discharge step, in order to ensure a sufficient second water flow rate and supply flow rate, it is preferable to supply cleaning water into the cleaning device 5 through two or more water supply pipes (first water supply pipe 55, second water supply pipe 56). By controlling the water flow rate of the cleaning liquid using two or more water supply pipes, cleaning water at a desired water flow rate can be supplied into the cleaning device 5 more simply, quickly and stably than by adjusting the flow rate of the water supply pipes. For example, the cleaning control unit 73 opens the water supply valves of the first water supply pipe 55 and the second water supply pipe 56 to supply the cleaning liquid into the cleaning device 5 through the piping 53 (see FIG. 7).

[0120] After the impurity washing step, it is preferable to supply washing water into the washing device 5 at a third water flow rate that is slower than the second water flow rate, as shown in Fig. 10(a), and further wash the MAP particles in the washing device 5 while discharging the impurities from the upper part of the washing device 5 (additional washing step). This makes it possible to further remove impurities such as sludge adhering to the MAP particles, thereby improving the purity of the MAP particles recovered from the washing device 5.

[0121] In the additional washing step, washing may be performed at a water flow rate that does not cause the MAP stored in the washing device 5 to leak from the washing device 5. Therefore, the third water flow rate must be set lower than that in the second water flow step. The third water flow rate may be the same as the first water flow rate. In order to reliably wash away sludge adhering to the surface of the MAP, it is desirable to set the third water flow time longer than the first water flow time.

[0122] The third water flow rate is, for example, preferably 0.2 to 0.5 m / min, more preferably 0.25 to 0.4 m / min, and even more preferably 0.30 to 0.35 m / min. The supply flow rate is, for example, preferably 75 to 195 L / min, more preferably 95 to 155 L / min, and even more preferably 110 to 130 L / min. The third water flow time can typically be about 1 to 10 minutes, preferably about 1 to 5 minutes, and even more preferably 5 to 7 minutes. As with the first water flow rate, supply flow rate, and first water flow time, the third water flow rate, supply flow rate, and third water flow time are of course not limited to the above.

[0123] It is preferable to perform the washing process continuously without providing a standing step between the impurity washing step in Fig. 9(c) and the additional washing step in Fig. 10(a). By performing the impurity washing step and the additional washing step continuously, the washing process can be performed continuously in the additional washing step while the MAP is in a fluidized state in the impurity washing step, so that the efficiency of the washing time and the amount of washing water can be improved.

[0124] After the additional washing step in Fig. 10(a), the mixture is left to stand as shown in Fig. 10(b). This allows the interface of the MAP particle layer in contact with the impurity layer to be smoothed. After the additional washing step, when the interface of the MAP particle layer in the washing device 5 is smoothed, the MAP in the washing device 5 is transferred to the outside of the washing device 5 using a transfer device 8 such as an air lift pump (transfer step) in Fig. 10(c). In the transfer step, for example, the washed MAP particles are extracted by introducing instrument air into the air lift pump of the transfer device 8, and are transferred to a water draining device or the like to remove moisture adhering to the MAP particles.

[0125] After the transfer is completed, as shown in Fig. 11(a), in order to smooth out the interface of the retained MAP layer that was disturbed during the transfer process, cleaning water is supplied into the cleaning device 5 using the first water supply pipe 55. Thereafter, the supply of cleaning water is stopped, and the inside of the cleaning device 5 is placed in a cleaning standby state as shown in Fig. 11(b).

[0126] According to this embodiment, in each of the tank washing step, the impurity discharge step, and the additional washing step, which are steps for substantially washing the MAP particles, the washing process can be sufficiently performed by only supplying the washing liquid to the washing device 5. Therefore, aeration is not required in the tank washing step, the impurity discharge step, and the additional washing step. This makes it possible to omit the power for aeration, thereby reducing the mechanical power and thermal energy required for phosphorus recovery.

[0127] 8(a) to 11(b), it is preferable to measure the amount of impurities accumulated on the upper layer of the phosphorus compound layer stored in advance in the cleaning device 5, and control the flow rate of cleaning water supplied into the cleaning device 5 based on the measurement result of the amount of impurities. The amount of impurities can be measured by, for example, measuring the interface height of the impurity layer containing impurities formed in the cleaning device 5 in a cleaning standby state as shown in FIG. 8(a) or FIG. 11(b).

[0128] The height of the interface of the impurity layer in the cleaning device 5 may be measured by an operator identifying the internal condition of the cleaning device 5 from outside the cleaning device 5 through a window 51 shown in Fig. 1. The operator can measure the height of the interface of the impurity layer from the window 51 at regular intervals using a scale 52 arranged adjacent to the window 51.

[0129] The cleaning control unit 73 sets the water flow conditions for the cleaning water to be supplied into the cleaning device 5 based on the amount of impurities stored in the cleaning device 5, in this case, the measurement result of the height of the interface of the impurity layer. For example, when the height of the interface of the impurity layer, which is the measurement result of the amount of impurities stored, exceeds a reference value, the cleaning control unit 73 performs at least a tank cleaning process as shown in Fig. 9(a) and an impurity discharge process as shown in Fig. 9(c) on the phosphorus compound particles newly supplied into the cleaning device 5.

[0130] That is, when the height of the interface of the impurity layer exceeds a reference value, the cleaning control unit 73 controls the cleaning conditions so as to carry out a tank cleaning process in which cleaning water is supplied from the lower part of the cleaning device 5 at a first water flow rate capable of cleaning the phosphorus compound particles in the cleaning device 5, and the phosphorus compound particles and impurities previously stored in the cleaning device 5 are mixed with the phosphorus compound particles newly supplied to the cleaning device 5 and cleaned, and after the tank cleaning process, cleaning water is supplied from the lower part of the cleaning device 5 at a second water flow rate greater than the first water flow rate, thereby washing the impurities in the cleaning device 5 to the upper part of the cleaning device 5, and the impurities are discharged from the upper part of the cleaning device 5 to the outside of the cleaning device 5.

[0131] On the other hand, when the height of the interface of the impurity layer, which is the measurement result of the amount of impurities stored in the cleaning device 5, is equal to or less than the reference value, it is sufficient to perform the tank cleaning process as shown in FIG. 9(a). That is, when the height of the interface of the impurity layer is equal to or less than the reference value, the cleaning control unit 73 controls the control conditions so that the cleaning water is supplied from the lower part of the cleaning device 5 at a first water flow rate capable of cleaning the phosphorus compound particles in the cleaning device 5, and the phosphorus compound particles and impurities stored in advance in the cleaning device 5 are mixed with the phosphorus compound particles newly supplied to the cleaning device 5, and a tank cleaning process is performed to clean them. When the tank cleaning process is performed as the cleaning process, the first water flow time is longer than the standard time, and can be, for example, about 12 to 15 minutes so that the sludge attached to the phosphorus compound particles can be sufficiently washed within the tank cleaning time. After the tank cleaning process, an additional cleaning process shown in FIG. 10(a) may be further performed.

[0132] Alternatively, when the height of the interface of the impurity layer is equal to or less than the reference value, only the additional cleaning step may be performed instead of the tank cleaning step. That is, when the height of the interface of the impurity layer is equal to or less than the reference value, after the steps shown in Fig. 8(a) to Fig. 8(b), the steps shown in Fig. 8(c), Fig. 9(a), Fig. 9(b), and Fig. 9(c) are omitted and the additional cleaning step shown in Fig. 10(a) is performed. It is known that the composition of impurities supplied to the cleaning device 5 varies depending on the region, treatment plant, and season. Therefore, the cleaning control unit 73 or the operator can select cleaning treatment conditions depending on the region, treatment plant, and season where the cleaning device 5 is located, thereby enabling more efficient cleaning treatment.

[0133] 12 is a schematic diagram showing the configuration of the crystallization reaction liquid containing phosphorus compound particles, which is withdrawn from the bottom of the reaction tank 1 via the withdrawal device 3 and supplied to the separation device 4a. In the separation device 4a, separated sludge is separated from the crystallization reaction liquid, the separated sludge is supplied to sludge treatment, and the remaining crystallization reaction liquid w0 containing phosphorus compound particles is charged into the washing device 5. In the washing device 5, the phosphorus compound particles can be classified into phosphorus compound particles (hereinafter referred to as "MAP leak") that flow out (leak) to the outside of the washing device 5 during the above-mentioned washing process, impurities w2 that are stored in the washing device 5, and phosphorus compound particles (hereinafter referred to as "pure MAP w1") that are separated from the sludge by the washing process in the washing device 5 and transferred to a drying facility in the downstream stage.

[0134] Taking into consideration the material balance in the reaction vessel 1, the separation device 4a, and the washing device 5, the weight W [kg-net MAP] of the pure MAP extracted from the reaction vessel 1 can be calculated by the following relational expression (1). W = Q × T × C × R / 100 (1) Where, W: Weight of pure MAP withdrawn from reactor 1 [kg-pure MAP] Q: Flow rate of crystallization reaction liquid containing MAP withdrawn from reactor 1 [m 3 -Crystallization reaction solution / min] T: Time taken for the crystallization reaction solution containing MAP to be withdrawn from reactor 1 [min] C: Net MAP concentration in reactor 1 [kg-net MAP / m 3 -Crystallization reaction solution] R: MAP recovery rate of separation device 4a [%]

[0135] Here, the volume of pure MAPw1 fed into the cleaning device 5 is v1, the volume of impurities w2 is v2, and the total volume of solid matter represented by the sum of the volume v1 of pure MAPw1 and the volume v2 of impurities w2 stored in the cleaning device 5 is v0. The density of pure MAP is ρ1, and the density of impurities is ρ2. When the total volume v0 is 100% pure MAP, the weight wM is: wM = v0 × ρ1 (2) It becomes.

[0136] If the weight of the actual total volume v0 is w0, the weight difference w3 caused by impurities entering the cleaning device 5 is w3=wM-w0 (3) Since the density difference ρ3 between the pure MAP and the impurity is ρ1-ρ2, the volume v2 of the impurity w2 is v2=w3 / ρ3 (4) It becomes.

[0137] Therefore, the weight w1 of the pure MAP fed into the cleaning device 5 is w1 = (v0-v2) × ρ1 ={v0-(w3 / ρ3)}×ρ1 =[v0-{(wM-w0) / (ρ1-ρ2)}]×ρ1 =[v0-{(v0×ρ1-w0) / (ρ1-ρ2)}]×ρ1···(5) It becomes.

[0138] The weight of impurity w2 is w2 = w0 - w1 (6) It becomes.

[0139] Furthermore, the concentration C of the pure MAP produced in the reactor 1 is given by the relations (1) and (5): C=[V0-{(v0×ρ1-w0) / (ρ1-ρ2)}]×ρ1÷(Q×T×R / 100) ···(7) It becomes.

[0140] The total volume v0 in the cleaning device 5 can be automatically measured by measuring the height of the interface of the impurity layer or the phosphorus compound particle layer with the image judgment device 6b or the ultrasonic measurement device 6d, and the total weight wo in the cleaning device 5 can be automatically measured with the weight measurement device 6c. The density ρ1 of the pure MAP, the density ρ2 of the impurities, and the recovery rate R may be actually measured values.

[0141] According to this embodiment, the amount of impurities stored in the cleaning device 5 is calculated from such a material balance relationship, and the above-mentioned cleaning process is controlled based on the calculation result of the amount of impurities stored, thereby making it possible to determine suitable cleaning conditions according to the amount of impurities stored in the cleaning device 5, regardless of the subjective judgment of the operator.

[0142] The concentration C of pure MAP generated in the reaction tank 1 can also be estimated from the measurement results of the cleaning device 5 using the relational expression (7). Therefore, the crystallization reaction in the reaction tank 1 can always be maintained in a stable and good state by appropriately controlling the treatment conditions in the reaction tank 1, such as the amount of chemical agent supplied from the chemical supply device 2, the amount of sludge supplied from the sludge supply device 10, and the frequency and amount of withdrawal of the crystallization reaction liquid containing the crystallized product of phosphorus compound particles from the withdrawal device 3, so that the concentration C of pure MAP falls within a suitable range according to the estimation result.

[0143] (Transportation control of transport equipment) As shown in Figure 13, the magnesium ammonium phosphate production system of the embodiment of the present invention further includes a transfer device 8 that transfers MAP particles in the cleaning device 5 to the outside of the cleaning device 5, and a transfer control unit 74 that controls at least one of the amount or frequency of transfer of MAP particles by the transfer device 8 based on the measurement results of a measuring device 6 that measures the amount of MAP particles stored in the cleaning device 5.

[0144] The transfer device 8 is not particularly limited as long as it is a means capable of transferring the washed MAP particles in the cleaning device 5 to the outside of the cleaning device 5, but an air lift pump is preferred in that it allows the device to be made smaller and simpler. When an air lift pump is used, it is preferred to keep the interface height of the MAP particle layer in the cleaning device 5 constant, as described above.

[0145] In this system, the extraction control unit 72 controls at least one of the extraction amount or the extraction frequency of the crystallization reaction liquid containing MAP particles in the reaction tank 1, so that the seed crystal concentration in the reaction tank 1 can be controlled within a certain range. Therefore, the height of the interface of the MAP particle layer in the cleaning device 5 can also be kept within a certain range. However, if a slight fluctuation in the seed crystal concentration in the reaction tank 1 causes a fluctuation in the amount of MAP particles stored in the cleaning device 5, the MAP particles stored in the cleaning device 5 may not be properly maintained. For example, if the amount of MAP particles stored becomes too large and the height of the interface exceeds a predetermined range, impurities in the cleaning device may be transferred from the extraction device 3 together with the MAP particles, and the purity of the MAP particles may decrease. If the amount of MAP particles stored becomes too small, the extraction device 3 may not be able to transfer the MAP particles sufficiently.

[0146] 13, when the amount of MAP particles stored in the cleaning device 5 exceeds a set value based on the measurement results of the measuring devices (image assessment device 6b, weight measuring device 6c, ultrasonic measuring device 6d) that measure the amount of MAP particles stored in the cleaning device 5, the transfer control unit 74 controls the extraction device 3 to adjust the amount of MAP particles transferred at one time and the transfer frequency (number of transfers) from the cleaning device 5. For example, when the extraction device 3 is an air lift pump, the amount of MAP transferred and the transfer frequency can be controlled by adjusting the amount of air.

[0147] For example, the weight w1 or volume v1 of the pure MAP stored in the cleaning device 5 is calculated based on the above-mentioned material balance relational expressions (1) to (7), and the transfer control unit 74 adjusts at least one of the transfer amount or transfer frequency of the pure MAP from the cleaning device 5 based on this calculation result. For example, when the calculation result of the weight w1 of the pure MAP is greater than a set value, the transfer control unit 74 performs control to increase the amount of pure MAP transferred per time from the cleaning device 5 by the transfer device 8 or increase the transfer frequency (number of times), for example, according to the proportion of the increase. When the calculation result of the weight w1 of the pure MAP is equal to or less than the set value, the transfer control unit 74 performs control to decrease the amount of pure MAP transferred per time from the cleaning device 5 by the transfer device 8 or decrease the transfer frequency. This makes it possible to provide a phosphorus compound production method and a phosphorus compound production system capable of efficiently transferring the phosphorus compound from the cleaning device 5.

[0148] (Control based on MAP production plan) As shown in FIG. 14, a magnesium ammonium phosphate production system according to an embodiment of the present invention includes one or more phosphorus production facilities 1000a, 1000b, and a production management system 1001 that manages the operating conditions of the phosphorus production facilities 1000a, 1000b based on a MAP particle production plan including information on the production volume of MAP particles.

[0149] The production management system 1001 may be connected to each phosphorus production facility 1000a, 1000b via a network 40 as shown in FIG. 14, and the operating conditions of each phosphorus production facility 1000a, 1000b may be shared with each other via the network 40. The production management system 1001 may be mounted in a general-purpose computer or the like in each phosphorus production facility 1000a, 1000b. The production management system 1001 may be connected to a server 1002 via the network 40, and may be configured to be remotely controlled or automatically controlled by the server 1002 or the like connected via the network 40. The server 1002 may perform a predetermined calculation process executed by the production management system 1001.

[0150] The information included in the MAP particle production plan includes, but is not limited to, for example, seasonality of MAP particle demand, regional information on the storage locations of one or more storage facilities 100A, 100B, 100C that store MAP particles, the amount of MAP particles to be transported to the storage facilities 100A, 100B, 100C, the deadline for transport, the transport time and transport distance when transporting MAP particles from the phosphorus production facilities 1000a, 1000b to the storage facilities 100A, 100B, 100C, the load weight and number of transport means such as transport vehicles, etc. 101A, 101B, 101C, etc. Furthermore, the MAP particle production plan may include information on prices such as the price of MAP particles, the price of chemicals required for the production of MAP particles, the price of electricity for operating the phosphorus production facilities 1000a, 1000b, and the price of phosphorus resources other than MAP particles.

[0151] The production management system 1001 includes an information acquisition unit 1001A that acquires MAP particle production plan information including information on the production volume of MAP particles; an analysis unit 1001B that acquires the MAP particle production plan information and the operating conditions of the phosphorus production facilities 1000a, 1000b, and analyzes whether to control the Mg / P ratio in the crystallization reaction solution in the reaction tank 1 provided in each phosphorus production facility 1000a, 1000b, the amount or frequency of withdrawal of the crystallization reaction solution containing MAP from the reaction tank 1, or the input sludge flow rate; an operating condition determination unit 1001C that determines, based on the MAP particle production plan information, so that at least one of the operating conditions, the Mg / P ratio in the crystallization reaction solution, the amount or frequency of withdrawal of the crystallization reaction solution from the reaction tank 1, or the input sludge flow rate, is an appropriate condition; and a learning unit 1001D.

[0152] The operating condition determination unit 1001C determines the operating conditions of the phosphorus production facilities 1000a and 1000b based on the analysis results of the analysis unit 1001B, and transmits the determination results to the phosphorus production facilities 1000a and 1000b. The learning unit 1001D uses learning data including information on past MAP particle production plans, property information on the input sludge of the phosphorus production facilities 1000a and 1000b, and operation control information associated with the information on the MAP particle production plans and the property information on the input sludge to create a model 1003 that receives input of a predetermined explanatory variable (here, input information including the MAP particle production plan and the property information on the input sludge) and outputs a predetermined objective variable (here, output information including the optimized operating conditions) using the model created. The learning unit 1001D can further predict optimal operating information for the phosphorus production facilities 1000a and 1000b.

[0153] 15, the information acquiring unit 1001A acquires information on the MAP particle production plan including information on the production amount of MAP particles. For example, in step S1, the information acquiring unit 1001A acquires information that the MAP particles are already sufficiently stored in the storage facilities 100A, 100B, and 100C and that there will be insufficient storage space, and information that the price of MAP particles is likely to fall in the near future and that it is necessary to reduce the production amount of MAP particles in the phosphorus production facilities 1000a and 1000b.

[0154] In step S2, the analysis unit 1001B selects the operating conditions to be controlled in the phosphorus production facilities 1000a and 1000b based on the acquired information on the MAP particle production plan. For example, the analysis unit 1001B selects the operating conditions to supply the input sludge flow rate of the phosphorus production facility 1000a less than the set value based on the information on the decrease in the MAP particle production amount included in the MAP particle production plan.

[0155] In step S3, the operating condition determination unit 1001C determines a set value of the future input flow rate of input sludge so as to reduce the input flow rate of input sludge of the phosphorus production facility 1000a in operation based on the analysis result of the analysis unit 1001B, and optimizes the operating conditions, for example, by increasing the amount or frequency of withdrawal of the crystallization reaction liquid containing MAP from the reaction tank 1, or by reducing the amount of chemicals supplied, according to the determination result. In step S4, the operating condition determination unit 1001C transmits the determined new operating conditions to the phosphorus production facility 1000a via the network 40, and ends the work. The operator can change various set values ​​of the operating conditions of the phosphorus production facility 1000a based on the new operating conditions.

[0156] In the reaction tank 1 (see FIG. 1, for example) provided in the phosphorus production facilities 1000a and 1000b, when the supply of input sludge is stopped once, it takes time to start up the reaction tank 1 when the supply is resumed, and crystallization of MAP particles may not be performed smoothly. According to this embodiment, the production management system 1001 can adjust the operating conditions of the phosphorus production facilities 1000a and 1000b based on the MAP particle production plan, so that, for example, the amount of MAP produced can be increased during a period when a large amount of fertilizer or the like needs to be produced. In addition, when the storage amount in the storage facilities 100A, 100B, and 100C is already sufficient, the production of phosphorus can be temporarily reduced. In this way, the operating conditions of the phosphorus production facilities 1000a and 1000b can be quickly adapted to conditions suitable for the change in the properties of the input sludge, and thus the phosphorus production efficiency can be stably improved.

[0157] An example of a processing flow using the learning unit 1001D is shown in FIG. 16. When the information acquisition unit 1001A of FIG. 14 acquires a MAP particle production plan in step S21, the learning unit 1001D creates a model for calculating optimal operating conditions by machine learning according to the input of the MAP particle production plan in step S22. The learning unit 1001D creates a model that outputs a predetermined objective variable in response to the input of a predetermined explanatory variable using learning data including information on past MAP particle production plans, property information on the input sludge of the phosphorus production facilities 1000a and 1000b, and operation control information associated with the information on the MAP particle production plan and the property information on the input sludge. In step S23, the operating condition determination unit 1001C of FIG. 14 inputs the MAP particle production plan acquired by the information acquisition unit 1001A into the model, and determines the operating conditions of the phosphorus production facilities 1000a and 1000b by calculating the optimal operating conditions of the phosphorus production facilities 1000a and 1000b. In step S24, the operating condition determination unit 1001C transmits operating information to the phosphorus production facilities 1000a, 1000b. In step S25, MAP particles are produced in the phosphorus production facilities 1000a, 1000b based on the new operating conditions. In step S26, the operating condition determination unit 1001C determines whether the amount of phosphorus produced by the phosphorus production facilities 1000a, 1000b is in line with the plan based on information on the amount of MAP particles recovered from the phosphorus production facilities 1000a, 1000b. If it is not in line, the process returns to step S23, and if it is in line, the process ends.

[0158] According to this embodiment, the operating conditions of the phosphorus production facilities 1000a, 1000b that are optimal according to the MAP particle production plan are calculated using machine learning, making it possible to determine more accurate optimized operating conditions that take into account past operating conditions and production plan information, thereby enabling phosphorus production to be carried out stably over a long period of time.

[0159] (Other embodiments) Although the present invention has been described by the above embodiments, the descriptions and drawings forming a part of this disclosure should not be understood as limiting the present invention. From this disclosure, those skilled in the art can realize various alternative embodiments and operating techniques.

[0160] The fluctuation in the phosphorus input load of the input sludge can basically be obtained from the difference between the measurement result by the measuring device 6 at the Xth extraction process and the measurement result by the measuring device 6 at the X+1th extraction process. The MAP particle concentration in the crystallization reaction liquid extracted from the reaction tank 1 is not particularly problematic if the distribution of MAP in the reaction tank 1 is uniform. 3 Since the error is relatively large, if a zone where the fluidity is deteriorated even slightly occurs in the reaction tank 1, a bias in the MAP concentration (concentration gradient) may occur. The magnitude of the error may also affect the viscosity of the sludge in the reaction tank 1 and the MAP concentration itself. In order to reduce such errors, the fluctuation in the phosphorus input load of the input sludge may be measured based on the average value obtained from the most recent several measured values. For example, the average value may be obtained from multiple measurement results during the X-1th and X-2th withdrawal treatments prior to the Xth, and the amount of the crystallization reaction liquid withdrawn from the reaction tank 1 may be increased or decreased by comparing the average value with a set value. EXAMPLES

[0161] Examples of the present invention will be described below together with comparative examples. These examples are provided for a better understanding of the present invention and its advantages, and are not intended to limit the invention.

[0162] Phosphorus production treatment was carried out using a 24-hour continuous inflow / continuous outflow phosphorus recovery pilot system at a sewage treatment plant that employs anaerobic digestion of sludge. Sewage anaerobic digestion sludge with a retention period of approximately 35 days was used as the sludge fed into the system. Continuous operation for recovering phosphorus from sludge was carried out for approximately 200 days through Examples 1 and 2 described below. Measurement results of the properties of the fed sludge and the sludge in the reaction tank during the test period (average values ​​within the period) are shown in Table 1.

[0163] [Table 1]

[0164] The input flow rate of the input sludge was always constant, and magnesium hydroxide was supplied as a chemical agent as a Mg source for PO4-P in the supplied sludge. The supply amount of magnesium hydroxide was set to an initial value of 1.15 (initial PO4-P: 350 mg / L) as the Mg / P ratio (molar ratio), and was automatically controlled to be within the range of 1.10 to 1.20 throughout the operation period. The seed crystal concentration in the reaction tank (total weight of particles with a MAP particle size of 30 μm or more in 1 L of sludge) was set to an initial value of 50 g / L. The seed crystal concentration in the reaction tank was automatically controlled to be within the range of 45 g / L to 55 g / L throughout the operation period. Sludge was withdrawn from the reaction tank to the MAP washing device every 6 hours, and a method was adopted in which the sludge withdrawal time was automatically controlled according to the increase or decrease in the seed crystal concentration in the reaction tank, and the initial value of the sludge withdrawal time was set to 117 seconds based on the sludge withdrawal pump capacity, etc.

[0165] Example 1 First, in Example 1, in order to control the amount of Mg in the chemicals supplied to the reaction tank during the first 100 days and to control the Mg / P ratio to be approximately constant, a method was adopted in which the interface height of the MAP particle layer after the cleaning process of the most recent cleaning device was measured, the total weight of the MAP particles stored in the cleaning device was calculated from the interface height measurement results, and an automatic adjustment was made so that the chemicals with an appropriate amount of Mg were supplied to the reaction tank 1 according to the change in the total weight. An image judgment method was adopted as a control device that automatically measures the interface height of the MAP layer in the cleaning device, calculates the amount of MAP recovered in the cleaning device from the interface height, and issues control instructions for the magnesium hydroxide supply amount and sludge extraction time, allowing the inside of the cleaning device to be visually captured as an image through a transparent window on the side of the cleaning device.

[0166] To briefly explain an example of automatic adjustment of the Mg / P ratio, if the PO4-P concentration of the input sludge is 350 mg / L, the molar concentration of P is 350 / 31 = 11.3 mmol, so the amount of Mg required to synthesize all of this PO4-P into MAP is 24.3 x 11.3 = 275 mg / L. When the Mg / P ratio is 1.15, the amount of magnesium hydroxide supplied was automatically adjusted so that 275 x 1.15 = 316 mg / L of Mg was supplied.

[0167] If the MAP storage amount calculated from the height of the MAP particle layer interface after the cleaning process in the cleaning device 6 hours ago is assumed to be a relative value of 100, the MAP storage amount calculated from the height of the MAP particle layer interface after the cleaning process due to the withdrawal of the crystallization reaction liquid from the reaction tank this time is 102, an increase of 2%. This means that the amount of MAP generated in the reaction tank has increased due to an increase in the PO4-P concentration of the input sludge. Therefore, in such a case, the Mg / P ratio was automatically adjusted to 1.15 by increasing the amount of Mg supplied next time, taking into account the increase in the MAP storage amount, the volume of the reaction tank, the amount of withdrawn sludge, and the amount of MAP newly generated in the reaction tank over the past 6 hours.

[0168] To briefly explain an example of automatic adjustment of the seed crystal concentration in the reaction tank, the total weight of MAP particles calculated from the interface height of the MAP particle layer in the cleaning device directly reflects the amount of MAP particles in the reaction tank. Therefore, in order to maintain the amount of MAP in the reaction tank at 50g / L, a method was adopted in which the increase / decrease in the sludge withdrawal time from the reaction tank was automatically adjusted taking into account a specified coefficient, as in the case of maintaining Mg / P:1.15. Figure 17 shows an example of the change in the PO4-P concentration of the input sludge, Figure 18 shows the measurement results of the Mg / P ratio in the reaction tank, and Figure 19 shows the measurement results of the seed crystal concentration in the reaction tank.

[0169] As shown in Figures 17, 18, and 19, by changing the amount of magnesium hydroxide supplied to the reaction tank as a chemical in response to fluctuations in PO4-P in the input sludge, it was possible to control the Mg / P ratio to a value close to 1.15 within a certain range, and at the same time, it was possible to control the seed crystal concentration in the reaction tank to a value close to 50 g / L.

[0170] Example 2 Following the 100-day data collection in Example 1, in Example 2, the same sludge and the same equipment were used, and three values, namely (1) "input sludge pH", (2) "pH in reaction tank 1", and (3) "difference between MAP recovery amount in cleaning device and assumed MAP recovery theoretical value", were used as input values, and a continuous test was conducted for another 100 days for the purpose of improving the accuracy of the target concentration management of "Mg / P ratio" and "MAP concentration in reactor". The correction adjustment numerical matrix due to the fluctuation of these three numerical values ​​was performed by a prior survey using the input sludge according to this embodiment and AI learning, and an example of automatic adjustment based on the conditions shown in Table 2 is shown. In Table 2, "pH increase ↑" indicates that the pH of (1) "input sludge pH" and (2) "pH in reaction tank 1" increases above 0.15, "pH maintained →" indicates that the pH is within + / - 0.15, and "pH decrease ↓" indicates that the pH is decreased below 0.15. "MAP amount (actual > theoretical)" indicates the case where the MAP recovery amount exceeds the theoretical value by 4% or more, "MAP amount (actual = theoretical)" indicates the case where the MAP recovery amount is within + / - 4% of the theoretical value, and "MAP amount (actual < theoretical)" indicates the case where the MAP recovery amount is less than 4% of the theoretical value. For example, in Table 2, if (3) the difference between the actual MAP recovery amount and the theoretical value is in the "actual > theoretical" relationship, (1) the pH of the input sludge is increasing, and (2) the pH in the reaction tank 1 is increasing, this means that the set value of the "Mg / P ratio" is reduced by 2% and the amount of extracted sludge is increased by 4%. If (3) the difference between the actual MAP recovery amount and the theoretical value is in the "actual = theoretical" relationship, (1) the pH of the input sludge is maintained within the appropriate range, and (2) the pH in the reaction tank 1 is maintained within the appropriate range, the Mg / P ratio and the amount of extracted sludge are maintained at the current level.

[0171] [Table 2]

[0172] As shown in Figures 17, 18, and 19, in the test period of 100 to 200 days in Example 2, although the fluctuation range of PO4-P in the input sludge was larger than that in Example 1 with a test period of 0 to 100 days, the Mg / P ratio and the seed MAP concentration in the reaction tank 1 were stable with small fluctuation ranges, and accurate tracking adjustment control operation was possible. It was possible to control the Mg / P ratio to a value closer to 1.15 than in Example 1, and at the same time, it was possible to control the seed crystal concentration in the reaction tank to a value closer to 50 g / L than in Example 1.

[0173] Example 3 After the end of Example 1, a test was conducted to confirm the correlation between the known MAP weight in the cleaning device and each measuring device using an ultrasonic method (ultrasonic sensor) and a weight measuring method (load cell) as a measuring device for measuring the amount of MAP present in the cleaning device from the interface height of the MAP layer after the cleaning process in the cleaning device, in addition to the image processing method for determining the interface height of the MAP layer, which was applied in this Example 1, and to draw a calibration curve. The test method was to fill the inside of the cleaning device to the fullest with tap water, then gradually add a predetermined amount of the recovered MAP room temperature dried material from the top, leave it to stand for a while, and then measure the interface level and weight of the MAP layer for each of the three methods, and draw each calibration curve. Figure 20 shows the results of the calibration curves for each method.

[0174] From Figure 20, it can be seen that the image processing method used in Examples 1 and 2 has a correlation coefficient of about 0.99, and is certainly practical. On the other hand, the ultrasonic method, which measures the interface level of the MAP layer in the same way as the image processing method, has a correlation coefficient of about 0.99, and can be used as an alternative measurement device to the image processing method. Also, with the load cell of the gravimetric measurement type, the weight of the room temperature dried MAP that is inserted is directly measured, and so it has an even higher correlation coefficient and is the most accurate than the image processing method and ultrasonic method. From these test results, it can be seen that the ultrasonic method and load cell method are applicable as alternative methods to the image processing method adopted in Example 1. [Explanation of symbols]

[0175] 1. Reactor 2. Drug supply device 3…Extraction device 4, 4a, 4b…separation device 5. Cleaning equipment 6. Measuring equipment 6a…Phosphate concentration measuring device 6b…Image judgment device 6c…Weight measuring device 6d…Ultrasonic measuring device 6e…pH change measuring device 7...Control device 8…Transfer device 10...Sludge supply device 11, 12...pH meter 40…Network 50…Cleaning tank 51…Window section 53…Plumbing 54…Plumbing 55…First water supply pipe 56…Second water supply pipe 58b, 58c, 58d...Detection device 60b... Processing section 60e...Calculation section 61a…Sampling section 61b…Photography department 61c...Weight measurement section 61d…Ultrasound diagnostic department 61e…Seed crystal concentration prediction section 62a...Solid-liquid separation section 62b...Image judgment section 62e…Mg / P ratio prediction section 63a…Dilution section 63b…Storage volume calculation section 63c…Storage volume calculation section 63d…Storage volume calculation section 64a…Phosphate concentration measuring section 64b…Learning Department 65a...Input load calculation section 70...Arithmetic section 71...Drug supply control unit 72... Withdrawal control section 73...Cleaning control section 74...Transport control section 1000a, 1000b... Phosphorus production facility 1001...Production management system 1001A…Information acquisition section 1001B…Analysis department 1001C…Operating condition determination section 1001D…Learning Department 1002...Server

Claims

1. a reaction tank for crystallizing MAP particles in a crystallization reaction solution in which seed crystals of magnesium ammonium phosphate are present by mixing input sludge containing phosphorus in the crystallization reaction solution; A chemical supply device that supplies a chemical containing magnesium into the reaction tank; an extracting device for extracting the crystallization reaction liquid containing the MAP particles from the reaction tank; A separation device for separating the MAP particles from the crystallization reaction solution; a washing device that stores and washes the MAP particles separated by the separation device; A measuring device capable of measuring fluctuations in the phosphorus input load of the input sludge; A system for producing magnesium ammonium phosphate comprising:

2. 2. The system for producing magnesium ammonium phosphate according to claim 1, wherein the measuring device includes a phosphate concentration measuring device that automatically measures the phosphate concentration of the input sludge.

3. The magnesium ammonium phosphate production system according to claim 1, wherein the measuring device includes an image assessment device that measures the amount of MAP particles stored in the cleaning device by image assessment of the height of the interface of the MAP particle layer stored in the cleaning device.

4. 2. The magnesium ammonium phosphate generating system according to claim 1, wherein the measuring device includes a weight measuring device that measures the amount of the MAP particles stored in the cleaning device by measuring the weight of the cleaning device.

5. The magnesium ammonium phosphate production system of claim 1, wherein the measuring device includes an ultrasonic measuring device that measures the amount of MAP particles stored in the cleaning device by using ultrasonic waves to measure the height of the interface of the MAP particle layer stored in the cleaning device.

6. The magnesium ammonium phosphate production system of claim 1, wherein the measuring device includes an image assessment device that image-assessssses the height of the interface of the MAP particle layer stored in the cleaning device, and a weight measuring device that measures the weight of the cleaning device, thereby measuring the amount of MAP particles stored in the cleaning device.

7. 2. The system for producing magnesium ammonium phosphate according to claim 1, wherein the measuring device includes a pH change measuring device that measures a change in pH over a predetermined time in the reaction tank.

8. The system for producing magnesium ammonium phosphate according to any one of claims 1 to 7, further comprising a chemical supply control unit that controls the amount of the chemical supplied from the chemical supply device based on the measurement result of the measurement device so that the Mg / P ratio of the crystallization reaction solution in the reaction tank is within an appropriate range.

9. The system for producing magnesium ammonium phosphate according to any one of claims 1 to 7, further comprising an extraction control unit that controls at least one of an amount or frequency of the crystallization reaction solution extracted by the extraction device based on a measurement result of the measurement device.

10. The system for producing magnesium ammonium phosphate according to any one of claims 1 to 7, further comprising a cleaning control unit that controls at least one of the flow rate or time of the cleaning solution supplied to the cleaning device based on the amount of impurities deposited on the MAP particle layer stored in the cleaning device.

11. A transfer device that transfers the MAP particles in the cleaning device to the outside of the cleaning device; A magnesium ammonium phosphate production system as described in any one of claims 1 to 7, further comprising a transport control unit that controls at least one of the amount or frequency of transport of the MAP particles by the transport device based on the measurement results of the measuring device that measures the amount of the MAP particles stored in the cleaning device.

12. a seed crystal concentration prediction unit that predicts a future seed crystal concentration in the crystallization reaction solution in the reaction tank based on a pH in the reaction tank and a measurement result of the amount of the MAP particles stored in the cleaning device; an Mg / P ratio prediction unit that predicts the supply amount of the chemical so that the Mg / P ratio of the crystallization reaction solution in the reaction tank in the future will be within an appropriate range based on the prediction result of the seed crystal concentration prediction unit; The system for producing magnesium ammonium phosphate according to any one of claims 1 to 7, further comprising:

13. an information acquisition unit that acquires MAP particle production plan information including information on the production amount of the MAP particles; an operating condition determination unit that determines at least one of the operating conditions, i.e., the Mg / P ratio in the crystallization reaction solution, the amount or frequency of withdrawal of the crystallization reaction solution from the reaction tank, or the input sludge flow rate, based on the MAP particle production plan information; The system for producing magnesium ammonium phosphate according to any one of claims 1 to 7, further comprising:

14. The phosphorus-containing input sludge is mixed in a reaction tank containing a crystallization reaction solution containing magnesium ammonium phosphate seed crystals, thereby crystallizing MAP particles in the crystallization reaction solution; extracting the crystallization reaction solution containing the MAP particles from the reaction tank to separate the MAP particles from the crystallization reaction solution, introducing the separated MAP particles into a washing device for washing, and transporting and recovering the MAP particles after washing, measuring a fluctuation in the phosphorus input load of the input sludge, and controlling at least one of an amount of a magnesium-containing chemical to be supplied into the reaction tank and an amount or frequency of the crystallization reaction liquid to be withdrawn from the reaction tank based on the measurement result; A method for producing magnesium ammonium phosphate, comprising:

Citation Information

Patent Citations

  • Organic sludge treating method

    JP1994071297A

  • Water quality controller

    JP1997117750A

  • Dephosphorizing apparatus

    JP1999290863A

  • Method of treating organic waste water and sludge, and treatment apparatus thereof

    JP2004160300A

  • Apparatus and method for treatment of organic effluent and organic sludge

    JP2018075555A