JP4770726B2 - Method for determining operating conditions of membrane filtration device, membrane filtration device using the same - Google Patents

Description

  The present invention relates to a method for determining the operating conditions of a membrane filtration device for filtering a liquid through a separation membrane, a method for operating a membrane filtration device using the same, and a membrane filtration device, and more particularly, activated sludge, microbial culture solution, etc. A method for determining the operating conditions of a membrane filtration device characterized by separating the solid and liquid by membrane filtration, the operating method of the membrane filtration device based thereon, and the membrane filtration having means for controlling the operation based on the operating conditions Relates to the device.

  In a membrane filtration apparatus that performs membrane filtration of a liquid with a separation membrane, it is a great merit that the membrane filtration flow rate per unit area of the separation membrane, that is, the membrane filtration flux can be constantly kept high. This is because the membrane area necessary for obtaining a membrane filtrate having a constant flow rate is reduced, and the installation area of the membrane filtration device can be saved and the membrane equipment cost can be reduced. However, if the membrane filtration flux is set high or the membrane filtration pressure, which is the membrane filtration driving force, is set high to maintain the high membrane filtration flux, the non-membrane permeable substance contained in the liquid to be filtrated is applied to the membrane surface. It is easy to accumulate, and the accumulated substance causes an increase in film resistance. As membrane resistance increases, it is necessary to increase the pressure required to filter the membrane, and when obtaining membrane filtrate at a constant membrane filtration flow rate (constant membrane filtration flux) When the pressure rises and the membrane filtrate is obtained at a constant membrane filtration pressure, the membrane filtration flow rate (membrane filtration flux) will be reduced. When such an increase in transmembrane pressure or a decrease in the membrane filtration flow rate occurs, it becomes difficult to operate the membrane filtration device, so that it is necessary to wash or replace the separation membrane. That is, if the membrane filtration flux is set high, or if the membrane filtration pressure that is the membrane filtration driving force is set high in order to maintain the high membrane filtration flux, the frequency of the chemical washing and replacement of the separation membrane increases. The above-mentioned merits may be lost due to an increase in replacement costs.

  In order to solve the above problems, by injecting a modifier for the filtrate to be filtrated, it is difficult to accumulate non-membrane permeable substances contained in the filtrate to be filtrated on the membrane surface, and the membrane filtration flux is set high. Even if the membrane filtration pressure, which is the membrane filtration driving force, is set high to maintain a high membrane filtration flux, the increase in membrane resistance and transmembrane differential pressure, or the decrease in membrane filtration flow rate (membrane filtration flux) is suppressed. Can be able to. As a result, the frequency of chemical washing and replacement of the separation membrane is reduced, and as a result, the replacement cost of the membrane can be kept low. However, if the amount of the modifying agent to be used is excessive, the cost of the modifying agent increases. Therefore, it is very important to adjust the proper amount of modifying agent to be injected.

In Patent Document 1, the amount of the modifier to be injected is preferably 0.1 to 1% per solid content weight in the membrane filtration device, and particularly preferably 0.2 to 0.8%. It is preferred. And if it is less than 0.1%, the aggregation effect of the non-membrane permeable substance contained in the liquid to be filtered is insufficient, and the high membrane filtration flux cannot be maintained for a long time with a low transmembrane pressure difference. Even if the injection amount exceeds 1%, the aggregation effect of the non-membrane permeable substance contained in the liquid to be filtered does not change so much and the cost of the modifying agent is high, which is not preferable.
JP-A-8-332483

  The appropriate injection amount of the modifying agent may vary greatly depending on the membrane filtration properties of the liquid to be filtered, the membrane filtration conditions, and the water temperature conditions. It only shows the amount of modifier to be injected, so for example, given the properties of a certain liquid to be filtered, membrane filtration conditions, and water temperature conditions, how much modifier should be injected. However, there is a problem that it is impossible to determine whether the operation cost can be minimized while maintaining the high membrane filtration flux for a long time.

  The present invention has been made in response to this problem, and its purpose is to determine the membrane filtration properties of the liquid to be filtered, membrane filtration conditions, and the amount of the modifying agent injected suitable for the water temperature, By operating the membrane filtration device based on the determined value, it is possible to reduce the operating cost, the method for determining the operating condition of the membrane filtration device that contributes to the reduction of the burden on the membrane filtration device, and the operation method of the membrane filtration device based on the method, The present invention also provides a membrane filtration device having means for controlling the operation based on the operating conditions.

  In order to solve the above problem, it is essential that the method for determining the operating condition of the membrane filtration device of the present invention has any of the following configurations.

  (1) Using the membrane filtration apparatus having at least a separation membrane, a modifier injecting means for filtrate, and a modifier injection amount adjusting means for adjusting the injection amount of the modifier for filtrate A method for determining the operating conditions of a membrane filtration device that filters the liquid through the separation membrane and adjusts the amount of the modifying agent injected simultaneously or sequentially with the filtration, and the types and configurations of the components of the liquid to be filtered From the data representing the amount of components, the amount of the modifying agent injected, and the time-series change in the membrane filtration flow rate, the fluctuation of the component amount adhering to the separation membrane, the fluctuation of the membrane filtration pressure, or the membrane resistance A method for determining operating conditions of a membrane filtration device, comprising predicting fluctuations and determining operating conditions of the modifier injection amount adjusting means based on the prediction results.

  (2) At least using a membrane filtration apparatus having a separation membrane, a modifier injection means for filtrate, and a modifier injection amount adjusting means for adjusting the injection amount of the filtrate modifier. A method for determining the operating conditions of a membrane filtration device that filters the liquid through the separation membrane and adjusts the amount of the modifying agent injected simultaneously or sequentially with the filtration, and the types and configurations of the components of the liquid to be filtered Fluctuations in the amount of constituents adhering to the separation membrane, fluctuations in membrane filtration flow rate or membrane filtration flux based on data representing time-series changes in the amount of components, the amount of modifier injected, and membrane filtration pressure Alternatively, a method for determining the operating condition of the membrane filtration device, comprising predicting a fluctuation of the membrane resistance and determining an operating condition of the modifying agent injection amount adjusting means based on the prediction result.

  According to the present invention, according to various kinds of filtrate to be filtrated, an appropriate operating condition of the adjusting means for adjusting the injection amount of the filtrate modifier is determined, and based on the decision result, membrane filtration is performed. By operating the device, it is possible to suppress an increase in membrane resistance, membrane filtration pressure and membrane filtration flux over a long period of time, or to use a separation membrane for a long period of time. Costs are kept low. It is also possible to minimize the overall operating cost, including the cost of the necessary modifiers and membrane replacement. For example, depending on the respective conditions, such as when the membrane filtration flux is very large or small, when intermittent filtration is performed, when continuous filtration is performed, or when the cleaning effect on the membrane surface is large or small, the membrane resistance or long-term In order to suppress an increase in membrane filtration pressure and a decrease in membrane filtration flux, or to use a separation membrane for a long period of time, the appropriate modifier injection amount should be different. Therefore, it is possible to determine an appropriate amount of modifier injection.

  Hereinafter, the present invention will be described in detail and specifically.

  The present invention uses at least a membrane filtration apparatus having a separation membrane, a modifier injection means for filtrate, and a modifier injection amount adjusting means for adjusting the injection amount of the filtrate modifier. The present invention relates to a method for determining operating conditions of a membrane filtration device that filters a filtrate through the separation membrane and adjusts the amount of the modifier injected simultaneously or sequentially with the filtration.

  Here, the separation membrane has a function of capturing a substance having a certain particle diameter or more contained in the filtrate by applying pressure to the filtrate or sucking from the permeate side. There are dynamic filtration membranes, microfiltration membranes, ultrafiltration membranes, nanofiltration membranes, reverse osmosis membranes, etc., depending on the difference in trapped particle size. The separation membrane used in the present invention is preferably a dynamic filtration membrane, a microfiltration membrane, or an ultrafiltration membrane. In addition, examples of the shape of the separation membrane include a flat membrane and a hollow fiber membrane, but the shape is not particularly limited.

  In addition, the filtrate to be subjected to membrane filtration is generally a liquid containing suspended solids and is not particularly limited. For example, in the case of a liquid containing microorganisms, Since microorganisms and microbial metabolites as substances are present in a relatively high concentration, it is difficult to predict fluctuations such as resistance when the filtrate is subjected to membrane filtration. It is possible to predict fluctuations such as resistance when the liquid is subjected to membrane filtration, and accurately determine the operating conditions of the adjustment means (modifier injection amount adjustment means described later) in the membrane filtration device for membrane filtration of the liquid to be filtered. It becomes possible to do. In this respect, the liquid to be filtered in the present invention is preferably a liquid containing microorganisms such as a microorganism culture solution and activated sludge. Moreover, in this invention, the to-be-filtered liquid whose density | concentration of a suspended solid is 100 mg / L or more is preferable.

  Moreover, the modifier for to-be-filtered liquid should just have the effect which changes the membrane filterability of to-be-filtered liquid, for example, a coagulant | flocculant, a biologic, a chemical | medical agent, etc. are mentioned. The flocculant has the effect of suppressing the increase in membrane resistance as a result of the effect of aggregating non-membrane permeable substances in the liquid to be filtered and increasing the water channels in the non-membrane permeable substances accumulated on the separation membrane surface. Inorganic flocculants such as polyaluminum chloride (PAC) and aluminum sulfate (sulfuric acid band), cation such as aminoalkyl (meth) acrylate quaternary salt (co) polymer, polyaminomethylacrylamide salt and chitosan Polymer flocculants, acrylamide / sodium acrylate copolymers, acrylamide / sodium acrylate / AMPS copolymers, anionic polymer flocculants such as polyacrylic acid soda, and nonionic polymer flocculants such as polyacrylamide There is. In addition, biologics contain (micro) organisms such as micrometazoans, protozoa, and bacteria, and as a result, by feeding on non-membrane permeable substances in the filtrate, membrane resistance It has the effect of suppressing the rise. Examples of minute metazoans include rotifers such as the genus Rotaria and Philodina, and examples of protozoa include ciliates such as the genus Monosiga, genus Biosoeca and genus Codosiga, and ciliates such as the genus Tokophrya, Campanella, and Vorticella. Examples of the bacteria include floc-forming bacteria that form flocs. Moreover, a chemical | medical agent has the effect which suppresses a raise of a membrane resistance as a result by decomposing | disassembling the non-membrane permeation | transmission substance in a to-be-filtered liquid, for example, ozone, sodium hypochlorite, etc. are mentioned.

  The modifier for injecting the filtrate is not particularly limited as long as it can inject the modifier for the filtrate into the membrane filtration device. For example, a pump is used. Thus, there is a method of injecting the modifier for the filtrate into the membrane filtration device.

  The modifier injection amount adjusting means is not particularly limited as long as it can adjust and change the injection amount of the modifier. For example, the amount of discharge (liquid feeding) in the pump, There is a method of adjusting / changing the injection amount of the modifying agent by changing the discharge (liquid feeding) time or the like.

  In addition, the membrane filtration device is characterized in that the separation membrane is immersed in the liquid to be filtered, and the membrane filtration device is characterized in that the liquid to be filtered is sent to the container containing the separation membrane while being filtered. And a rotary membrane filtration device characterized in that the separation membrane is rotated by immersing all or part of the separation membrane in the liquid to be filtered. The membrane filtration method may be a dead end type or a cross flow type. Among them, as an apparatus to which the present invention can be applied, there are a membrane separation type activated sludge method used for treatment of sewage, factory effluent, manure and the like, a microbiological substance production process including a membrane separation step, and the like.

  In the present invention, the component adhering to the separation membrane from the data representing the type and amount of component of the liquid to be filtered, the injection amount of the modifying agent, and the time-series change in the membrane filtration flow rate. A fluctuation in quantity, a fluctuation in membrane filtration pressure, or a fluctuation in membrane resistance is predicted (hereinafter referred to as prediction step 1). Alternatively, from the data representing the time-series change in the type and amount of constituents of the liquid to be filtered, the amount of the modifying agent injected, and the membrane filtration pressure, the amount of constituents adhering to the separation membrane The fluctuation, the fluctuation of the membrane filtration flow rate or the membrane filtration flux, or the fluctuation of the membrane resistance is predicted (hereinafter referred to as prediction step 2).

  Here, the component of the liquid to be filtered is a component that constitutes the liquid to be filtered.For example, the liquid to be filtered is fractionated by centrifugation, filtration with filter paper, filtration with a separation membrane, and the like. The liquid which becomes can be made into a structural component. Further, in the present invention, it is preferable to use a supernatant component, more preferably to use a supernatant component and a solid component, as a kind of constituent components of the liquid to be filtered, and to use a supernatant component, a solid component and a dissolved component. Most preferably, it is used. Here, the supernatant component is, for example, a centrifugal supernatant obtained by centrifuging the liquid to be filtered, a separation membrane having a larger pore size than the separation membrane in the membrane filtration device (for example, a microfiltration membrane as a separation membrane in the membrane filtration device) In the case of using an ultrafiltration membrane or the like, it can be used as a filtrate when the filtrate is filtered using a filter paper or the like, and the solid component is obtained by, for example, centrifuging the filtrate. Using a separation membrane having a larger pore size than the separation membrane in the membrane filtration device (for example, a filter paper or the like when a microfiltration membrane or an ultrafiltration membrane is used as the separation membrane in the membrane filtration device) It can be set as the non-membrane filtrate thing when filtering a to-be-filtered liquid. The dissolved component is preferably a permeate from a separation membrane in a membrane filtration device or a separation membrane made of the same material as the separation membrane.

  Further, the component amount of the liquid to be filtered is a substance amount such as a dissolved substance or a suspended substance contained in the component, for example, total organic carbon (TOC), chemical oxygen demand ( COD), biochemical oxygen demand (BOD), extracellular polymer substance amount (EPS), suspended solids concentration (MLSS), dry weight, suspended solids loss on ignition (MLVSS), etc. . In the present invention, in view of ease of measurement, accuracy, and automation, it is recommended to specify the amount of substance measured based on the TOC.

  Further, in the present invention, it is preferable to use the difference amount between the substance amount of the supernatant component and the substance amount of the dissolved component as the constituent component amount, or to use the solid component substance amount as the constituent component amount. More preferably, the difference between the amount of the substance and the amount of the dissolved component and the amount of the solid component are used as the constituent amount. The amount of difference between the amount of the supernatant component in the liquid to be filtered and the amount of the dissolved component, and the amount of the solid component cannot be actually permeated through the membrane, and are resisted by adhering to the membrane surface. This is because the amount of the substance contributes to. In particular, when membranes are filtered, the substances that do not permeate the separation membrane in the membrane filtration device among the supernatant components of the filtrate (the amount of the substance is the amount of the supernatant component and the dissolved component) In most cases, the membrane resistance is mainly formed, and the amount of the component of the supernatant and the amount of the dissolved component are By using the difference amount, the component component amount of the filtrate to be filtered attached to the separation membrane, the fluctuation of the membrane filtration pressure, the fluctuation of the membrane filtration flow rate, the fluctuation of the membrane filtration flux, or more accurately It becomes possible to predict a change in the value of the membrane resistance. Moreover, the whole to-be-filtered liquid can be defined comprehensively and easily by using only the amount of solid components as a component amount of to-be-filtered liquid. In addition, substances contained in the solid component of the liquid to be filtered contribute to resistance if attached to the membrane, and are separated from the membrane in comparison with substances that do not permeate the separation membrane in the membrane filtration device among the supernatant components. Therefore, there is an effect that a substance that does not permeate the separation membrane in the membrane filtration device is entrained in the supernatant component adhering to the membrane and peeled off together. Therefore, by using the difference between the amount of the supernatant component and the amount of the dissolved component and the amount of the solid component as the component amount, the amount of the component of the liquid to be filtered attached to the separation membrane The above phenomenon can be reproduced when predicting fluctuations, fluctuations in membrane filtration pressure, fluctuations in membrane filtration flow rate, fluctuations in membrane filtration flux, or changes in the value of membrane resistance.

  The membrane filtration flow rate is the flow rate of the permeate obtained from the membrane. Further, the value obtained by dividing the value of the membrane filtration flow rate by the membrane area is the membrane filtration flux (that is, the membrane filtration flux is the membrane filtration flow rate per unit area of the separation membrane). Data representing a time-series change in the membrane filtration flow rate is a time-series combination of values of the membrane filtration flow rate for an arbitrary time, and the value of the membrane filtration flow rate may vary over time, It may be constant without changing. Alternatively, the value of the membrane filtration flow rate may be converted into a membrane filtration flux, and the data representing the time series change of the membrane filtration flow rate may be replaced with the data representing the time series change of the membrane filtration flux.

  The membrane filtration pressure is calculated as the difference between the pressure on the filtrate side of the separation membrane and the pressure on the permeate side. When the membrane filtrate is obtained by applying pressure or suction with a pump, Can be calculated from the water head difference or the like when the membrane filtrate is obtained using the siphon principle such as pressurization or suction pressure. The data representing the time-series change in the membrane filtration pressure is a time-series combination of the values of the membrane filtration pressure for an arbitrary time, and the value of the membrane filtration pressure may vary over time, It may be constant without changing.

  Moreover, the component amount adhering to the separation membrane is the amount of substance contained in the component of the liquid to be filtered adhering to the membrane surface. Here, when using two or more kinds of substance amounts (for example, the difference between the substance amount of the supernatant component and the substance amount of the dissolved component, the substance amount of the solid component, etc.) as the constituent component amount of the liquid to be filtered For each type of component, it means the amount of component adhering to the separation membrane.

The membrane resistance is a resistance generated when the liquid to be filtered is filtered through a membrane, and is generally defined by the equation (1).
R = ΔP / (μJ) (1)

  Here, ΔP is the membrane filtration pressure [Pa], μ is the viscosity of the membrane filtrate [Pa · s], R is the membrane resistance [1 / m], and J is the membrane filtration flux [m / s]. Here, μ may directly measure the viscosity of the membrane filtrate, but may be converted from the temperature according to the equation (2).

μ × 10 ³ = F · exp [(1 + BT) / (CT + DT ² )] (2)

  Here, F = 0.01257187, B = −0.005806436, C = 0.001130911, D = −0.000005723952, and T is the absolute temperature [K]. That is, when the Celsius temperature is σ [° C.], it is expressed as T = σ + 273.15.

  Prediction is input data (that is, in the case of prediction step 1, the type and amount of components of the liquid to be filtered, the injection amount of the modifying agent, and the time series of the membrane filtration flow rate). It is data representing a change, and in the case of the prediction step 2, it is data representing a time-series change in the type and amount of component of the liquid to be filtered, the injection amount of the modifying agent, and the membrane filtration pressure. ) To the target fluctuation (that is, in the case of the prediction step 1, the fluctuation of the component amount adhering to the separation membrane, the fluctuation of the membrane filtration pressure, or the fluctuation of the value of the membrane resistance). In the case of 2, it is a change in the amount of constituents adhering to the separation membrane, a change in the membrane filtration flow rate, a change in the membrane filtration flux, or a change in the value of the membrane resistance). It is.

  The means of the predicting step 1 or predicting step 2 includes a method of predicting based on the result of learning actual data using a neural network or the like, a membrane resistance value or a membrane filtration pressure as a function of input data. In the present invention, the data representing the time-series change in the component amount of the filtrate, the data representing the time-series change in the membrane filtration flow rate, or the membrane filtration It is preferable that the prediction method includes a calculation step 100 for calculating the value of the component amount adhering to the separation membrane based on data representing a time-series change in pressure. Accordingly, in the case of the prediction step 1, the fluctuation of the component amount adhering to the separation membrane, the fluctuation of the membrane filtration pressure, or the fluctuation of the membrane resistance value is changed. In the case of the prediction step 2, the separation membrane is changed. It is possible to more accurately predict fluctuations in the amount of constituents adhering to the membrane, fluctuations in the membrane filtration flow rate, fluctuations in the membrane filtration flux, or fluctuations in the value of the membrane resistance.

In addition, the calculation step 100 expresses the amount of change in the amount of the component adhering to the separation membrane as a difference between the rate at which the substance contained in the component of the liquid to be filtered adheres to the separation membrane and the rate at which the separation occurs. More preferably, the calculation step 101 is calculated based on the following calculation formula. Thus, it is possible to calculate the amount of change in the constituent component quantity attaching on the separation membrane, whereby, if the value of the constituent component quantity attaching on the separation membrane of any time t _i, a constant The value of the component amount adhering to the separation membrane at time t _{i + 1} can be calculated. In addition, even when the membrane filtration flow rate value, membrane filtration flux value, and membrane filtration pressure value are not constant and fluctuate, it is possible to calculate fluctuations in the amount of membrane adhesion of constituents of the liquid to be filtered. It becomes possible. Further, the rate at which the substances contained in the constituents of the liquid to be filtered peel from the separation membrane is determined based on a higher-order expression of the second or higher order of the amount of the constituent components adhering to the separation membrane. More preferably. Thus, in the case of the prediction step 1, the fluctuation of the component amount adhering to the separation membrane, the fluctuation of the membrane filtration pressure, or the fluctuation of the membrane resistance value is changed. In the case of the prediction step 2, the separation membrane is changed. It is possible to more accurately predict fluctuations in the amount of constituents adhering to the membrane, fluctuations in the membrane filtration flow rate, fluctuations in the membrane filtration flux, or fluctuations in the value of the membrane resistance.

  Further, the calculation formula expressing the amount of change in the amount of the supernatant component adhering to the separation membrane includes the value of the amount of the solid component adhering to the separation membrane, or the separation Most preferably, the calculation expression expressing the amount of change in the amount of the solid component adhering to the membrane is a calculation formula including the value of the amount of the supernatant component adhering to the separation membrane. As a result, the phenomenon that the substance contained in the solid component adhering to the separation membrane entrains the non-membrane-permeable substance contained in the supernatant component adhering to the membrane and peels off together (the solid of the liquid to be filtered If the component adheres to the membrane, it contributes to resistance, and it can be predicted that the component is easy to peel off from the membrane as compared to the non-membrane-permeable substance contained in the supernatant component. Is the fluctuation of the amount of the component adhering to the separation membrane, the fluctuation of the membrane filtration pressure, or the fluctuation of the value of the membrane resistance. It is possible to more accurately predict fluctuations, fluctuations in membrane filtration flow rate, fluctuations in membrane filtration flux, or fluctuations in the value of membrane resistance.

  Here, there are the following procedures as calculation procedures in the prediction step 1 and the prediction step 2.

  The prediction step 1 and the prediction step 2 include a calculation step 10 for calculating the value of the membrane filtration flux or the value of the membrane filtration pressure at an arbitrary time, and the value of the component amount adhering to the separation membrane at an arbitrary time. A calculation step 20 for calculating the film resistance, and a calculation step 30 for calculating the value of the membrane resistance at an arbitrary time. The calculation step 10, the calculation step 20 and the calculation step 30 are repeated while updating the time. Thus, in the prediction step 1 or the prediction step 2, it is possible to obtain a temporal change (that is, variation) of a value that is a prediction target.

  Here, in the calculation step 10, it is preferable to follow the equation (1). At this time, the expression (1) may be converted into the following expression (1 ').

R = AΔP / (μQ) (1 ′)
Here, Q is the membrane filtration flow rate [m ³ / s], and A is the membrane area [m ² ].

  By using the expression (1) or (1 ′), the value of the membrane filtration pressure at an arbitrary time in the case of the prediction step 1 or the value of the membrane filtration flow rate at an arbitrary time in the case of the prediction step 2 The value of the membrane filtration flux can be calculated.

  Further, in the calculation step 20, as described above, the amount of change in the amount of the constituent component adhering to the separation membrane is determined based on the rate at which the substance contained in the constituent component of the liquid to be filtered adheres to the separation membrane. It is preferable that the calculation step 101 is calculated based on a calculation expression expressed as a difference between and the calculation expression expressing the amount of change in the amount of substance of the supernatant component adhering to the separation membrane, A calculation formula that includes the value of the amount of the solid component adhering to the separation membrane or the amount of change in the value of the amount of the solid component adhering to the separation membrane is expressed by the separation membrane. More preferably, the calculation formula includes the value of the amount of the supernatant component adhering to the surface. Examples of such calculation formulas include the following formulas (3) and (4). In the present invention, it is recommended to follow the formulas (3) and (4). However, the scope of the present invention is not limited to the equations (3) and (4).

dXm / dt = X · J−γx · (τ−λx · ΔP) · (ηx · Xm + ηp · Pm) · Xm (3)
dPm / dt = P · J−γp · (τ−λp · ΔP) · (ηx · Xm + ηp · Pm) · Pm (4)

However, (τ−λx · ΔP) ≧ 0 and (τ−λp · ΔP) ≧ 0.

Here, X is the amount of the substance contained in the solid component [gC / m ³ ], P is the amount of the non-membrane permeable substance contained in the supernatant component [gC / m ³ ], and Xm is the separation per unit membrane area The amount of substance contained in the solid component adhering to the membrane [gC / m ² ], Pm is the amount of substance of the non-membrane permeable substance contained in the supernatant component adhering to the separation membrane per unit membrane area [gC / m ² ] , T is time [s], τ is membrane detergency [−], γx is the peel coefficient [1 / m / s] of the substance contained in the solid component, and γp is the non-membrane permeable substance contained in the supernatant component. Peel coefficient [1 / m / s], λx is the friction coefficient [1 / Pa] of the substance contained in the solid component, λp is the friction coefficient [1 / Pa] of the non-membrane permeable substance contained in the supernatant component, ηx the inverse of the density of the substance contained in the solid component [m 3 ^/ gC], ηp is the reciprocal of the density of non-membrane permeable substance contained in the supernatant component [ It is a ³ / gC]. Here, the first term on the right side of the equations (3) and (4) is the speed at which the substance contained in the constituent component adheres to the separation membrane, and the second term is the substance in which the constituent component is contained in the separation membrane. It shows the speed of peeling from. Further, by multiplying both sides of the formulas (3) and (4) by the membrane area A, the left side of the formula (3) is the amount of change in the amount of substance contained in the solid component attached to the separation membrane, and the formula (4) It is possible to convert the membrane filtration flux J on the right side into the membrane filtration flow rate Q to the amount of change in the amount of the non-membrane permeable substance contained in the supernatant component attached to the separation membrane on the left side.

  Further, as described above, the amount of change in the amount of the constituent component adhering to the separation membrane is expressed as the difference between the rate at which the substance contained in the constituent component of the liquid to be filtered adheres to the separation membrane and the rate at which the separation is performed. Is expressed as a differential equation related to the amount of constituents of the liquid to be filtered attached to the separation membrane. At that time, as a method for solving this differential equation, the Euler method, the Runge-Kutta method, the Runge method, and the like. -Kutta-Gill (RKG) method.

  By performing the calculation step 20 as described above, the value of the component amount adhering to the separation membrane at an arbitrary time can be calculated.

  In the calculation step 30, for example, the value of the membrane resistance at an arbitrary time can be calculated by following the equation (5).

  R = Rm + αxXm + αpPm (5)

  Here, Rm is the initial film resistance value [1 / m], αx is the resistance generation amount per unit amount of Xm [m / gC], and αp is the resistance generation amount per unit amount of Pm [m / gC]. ].

  As described above, it is possible to calculate the fluctuation of the value to be predicted by the three calculation steps of the calculation step 10, the calculation step 20, and the calculation step 30, but only the fluctuation of the film resistance value is calculated. In the case of prediction, the calculation step 10 may be omitted, and the calculation step 20 and the calculation step 30 may be used. For example, in the calculation step 20, in the case of the prediction step 1, when a calculation formula including the value of the membrane filtration pressure (for example, the formula (3) or the formula (4)) is used, the membrane filtration pressure is calculated in the calculation step 10. By using, in the calculation step 20, a calculation formula in which a calculation formula (for example, formula (1)) used for calculation of the value is substituted into a calculation formula including the value of the membrane filtration pressure in the calculation step 20, the membrane The calculation step for calculating the value of the filtration pressure can be omitted, and in the case of the prediction step 2, when using a calculation formula (for example, formula (3) or formula (4)) including the value of the membrane filtration flow rate or the membrane filtration flux The calculation formula (for example, the formula (1)) used for calculating the value of the membrane filtration flow rate and the membrane filtration flux in the calculation step 10 is changed to the value of the membrane filtration flow rate and the membrane filtration flux in the calculation step 20. Including formula By using the assignment was equation to the calculation step 20, it can be omitted calculating step of calculating the value of the membrane filtration pressure. Similarly, in the prediction step 1, only the fluctuation of the value of the membrane filtration pressure is predicted, and in the case of the prediction step 2, only the fluctuation of the value of the membrane filtration flow rate or the membrane filtration flux is predicted, the calculation step 30 may be omitted, and the calculation step 10 and the calculation step 20 may be configured by two calculation steps. For example, in the calculation step 10, when a calculation formula including the value of the membrane filtration resistance (for example, the equation (1)) is used, a calculation formula (for example, (5) used for the calculation of the value of the membrane resistance in the calculation step 30. The calculation step 30 can be omitted by using, in the calculation step 10, a calculation formula obtained by substituting the formula ()) into the calculation formula including the film resistance value in the calculation step 10. As described above, calculation step 10 or calculation step 30 can be omitted as appropriate, but it is recommended not to omit calculation step 20.

  Further, instead of data representing the injection amount of the modifying agent, the parameter that depends on the injection amount of the modifying agent is used as the input of the prediction step 1 or the prediction step 2, and the parameter and the modifying agent are used. It is preferable to determine the correlation with the injection amount in advance. By using these parameters as data serving as the input of the prediction step 1 or the prediction step 2 and by determining in advance the correlation between these parameters and the modifier injection amount, the modifier injection amount, and Based on this, it becomes possible to appropriately determine the operating conditions of the modifier injection amount adjusting means. Further, the parameter is any one of the resistance coefficient (αx, αp) in the equation (5), the peeling coefficient (γx, γp) and / or the friction coefficient (λx, λp) in the equations (3) and (4). Is preferably 1 or more, more preferably 2 or more of them. The resistance coefficient is a value of the membrane resistance generated per unit amount of the constituent component adhering to the separation membrane, as can be seen from the equation (5). Moreover, the peeling coefficient is a parameter representing the ease of peeling from the separation membrane of the component adhering to the separation membrane, as can be seen from the equations (3) and (4). The coefficient of friction is a parameter representing the degree to which the membrane filtration pressure hinders the separation of the constituent components adhering to the separation membrane from the separation membrane, as can be seen from the equations (3) and (4). By using these parameters that reflect the membrane filtration properties of the liquid to be filtered, in the case of the prediction step 1, fluctuations in the amount of components adhering to the separation membrane, fluctuations in membrane filtration pressure, or values of membrane resistance In the case of the prediction step 2, the fluctuation of the amount of constituents adhering to the separation membrane, the fluctuation of the membrane filtration flow rate, the fluctuation of the membrane filtration flux, or the fluctuation of the value of the membrane resistance is predicted with higher accuracy. It becomes possible to do.

  In addition, the value of the parameter is obtained by filtering the filtrate to be filtered through a test separation membrane, and based on changes in the value of membrane filtration pressure, membrane filtration flux or membrane filtration flow rate, and membrane resistance in the filtration process. Preferably it is determined. Thus, not only can the value of the parameter be determined with high accuracy, but also the value of the parameter can be determined easily without actually operating the membrane filtration device.

As the test, for example, there is a method using the membrane filtration test apparatus 400 shown in FIG. The membrane filtration test apparatus 400 is an apparatus that pressurizes the liquid to be filtered using nitrogen gas 405 (the pressure is measured by the pressure gauge 411) and filters the liquid to be filtered with the separation membrane 402. Here, the area of the separation membrane 402 is preferably 0.01 m ² or less, and more preferably a separation membrane having the same material and shape as the separation membrane of the membrane filtration device. Accordingly, since the separation membrane to be used is small, it is possible to easily determine the value of the parameter. In addition, by using a separation membrane having the same material and shape as the separation membrane of the membrane filtration device, the value of the parameter can be determined with high accuracy. As the liquid to be filtered, pure water, a liquid to be filtered collected from a membrane filtration device with the modifier added at an arbitrary concentration (hereinafter referred to as a modifier-added liquid to be filtered), and its supernatant component It is preferable to use it. This makes it possible to determine the value of the parameter with high accuracy.

  When pure water is used as the liquid to be filtered, for example, pure water is accommodated in the pure water chamber 410, and the pure water chamber 410 is pressurized with nitrogen gas 405, whereby the pure water is stirred by a cell 401 (for example, Millipore Corporation). ) Amicon 8010), and a method of filtering pure water with the separation membrane 402 installed in the membrane fixing holder 406 is conceivable. The membrane filtration pressure is preferably 5 to 20 kPa in consideration of the time required for the test, the accuracy of data obtained, and the safety aspect of the membrane filtration test apparatus.

  When the modifier-added filtered liquid or the supernatant component thereof is used as the liquid to be filtered, the pure water chamber 410 is removed and the dotted line 415 in FIG. By adding the filtrate to be added or the supernatant component thereof, and pressurizing the stirring cell 401 with nitrogen gas 405, the separation membrane 402 installed in the membrane fixing holder 406 is used to add a modifier addition filtrate. Or the method of filtering the supernatant component can be considered. Here, it is preferable that the stirring type cell 401 has a structure in which a stirring bar 404 is attached so that the liquid to be filtered can be stirred by the magnetic stirrer 403. As a result, an effect of peeling off the non-membrane-permeable component in the modifier-added filtrate to be attached to the separation membrane 402 or its supernatant component can be obtained. Can be determined with high accuracy. The amount of the modifier-added filtrate to be added or the amount of the supernatant component is preferably 100 ml or less. As a result, since a small amount of liquid to be used for filtration is required, the value of the parameter can be easily determined. In addition, it is preferable that the liquid to be filtered is diluted by 1.1 times or more and charged into the stirring cell 401. This makes it possible to shorten the time required for the test. The membrane filtration pressure is preferably 5 to 20 kPa in consideration of the time required for the test, the accuracy of data obtained, and the safety aspect of the membrane filtration test apparatus. Further, during the test, it is preferable to include one or more periods during which filtration is suspended for 10 seconds or more by closing the valve 414. This makes it possible to determine the value of the parameter with high accuracy.

  In each test, it is preferable that the presence or absence of pressurization of each part of the membrane filtration test apparatus can be adjusted by opening and closing the valve 412, the valve 413, and the valve 414. The membrane permeate is preferably received by a beaker 407 placed on an electronic balance 408, the amount of the membrane permeate is measured by the electronic balance 408, and the measured value is taken into the personal computer 409. This makes it easy to acquire data and increases the accuracy of the obtained data, so that the parameter values can be determined with high accuracy. Data showing the relationship between the filtration time taken into the personal computer 409 and the amount of filtrate is processed as follows. First, the membrane filtration flux at an arbitrary filtration time is calculated using the differential coefficient of the filtrate amount at an arbitrary filtration time. Next, from the membrane filtration flux at the arbitrary filtration time, the membrane filtration resistance at the arbitrary filtration time is calculated according to the equation (1) using the membrane filtration pressure. From the result calculated as described above, a relationship between the total filtrate amount per unit membrane area and the membrane filtration resistance is created.

  As a method for determining the value of the parameter from the relationship between the total amount of filtrate per unit membrane area and the membrane filtration resistance obtained in this way, various values are given to the parameter and the prediction step 2 is executed. Therefore, there is a method for predicting the fluctuation of the membrane filtration resistance and setting the parameter value that minimizes the difference between the prediction result and the actually obtained fluctuation of the membrane filtration resistance as the parameter value to be determined. is there. Here, the prediction step 2 includes three calculation steps of the calculation step 10, the calculation step 20, and the calculation step 30, and further, the calculation step 10 can be expressed by the equation (1) or (1) ′. It is preferable to use the equation (3) and the equation (4) as the calculation step 20 and use the equation (5) as the calculation step 30. This makes it possible to determine the value of the parameter with high accuracy. In order to predict the fluctuation of the membrane filtration resistance using the equation (1) (or the equation (1) ′), the equation (3), the equation (4), and the equation (5), the initial membrane resistance Rm, the membrane cleaning power τ It is necessary to give the value of Among these, for the initial membrane resistance Rm, the membrane filtration resistance is almost constant from the relationship between the total filtrate amount per unit membrane area and the membrane filtration resistance created from the result of the test conducted using the pure water as the filtration target liquid. It is preferable that the average value of the portion is Rm. Further, it is preferable to give a certain value (for example, 1) for the film cleaning power τ.

  The procedure for determining the values of the parameters (that is, resistance coefficients αx, αp, peeling coefficients γx, γp, friction coefficients λx, λp) is as follows. First, the values of αp, γp, and λp are determined from the test results obtained by using the supernatant component of the modifier-added liquid to be filtered as the liquid to be filtered. Specifically, the substance amount Xm contained in the solid component adhering to the separation membrane per unit membrane area is set to 0, and various values are given to αp, γp, and λp, and the equation (1) (or (1) ′ equation) , (4), and (5) are used to predict fluctuations in membrane filtration resistance. The values of αp, γp, and λp that minimize the difference between the predicted result and the actually obtained fluctuation in membrane filtration resistance are the values of αp, γp, and λp to be determined. Next, the values of αx, γx, and λx are determined from the test results obtained by using the modifier-added liquid to be filtered as the liquid to be filtered. Specifically, αp, γp, and λp are given the values determined by the above procedure, and αx, γx, and λx are given various values, and (1) formula (or (1) ′ formula), (3) The fluctuation | variation of membrane filtration resistance is estimated using Formula, Formula (4) Formula (5). The values of αx, γx, and λx that minimize the difference between the prediction result and the actually obtained fluctuation of the membrane filtration resistance are the values of αx, γx, and λx to be determined.

  In the present invention, based on the result predicted as described above, the modifying agent injection amount and the operating condition of the modifying agent injection amount adjusting means based thereon are determined. As a method for determining the operating condition, it is preferable to calculate the operating cost of the membrane filtration device based on the prediction result and to determine the operating condition of the appropriate modifier injection amount adjusting means based on the operating cost. . Here, the operating cost of the membrane filtration device is a cost necessary for operating the membrane filtration device. The effect of injecting the modifier into the membrane filtration device is that the non-membrane-permeating substance contained in the liquid to be filtered does not accumulate on the membrane surface. The increase in differential pressure can be suppressed, and even if the same membrane filtration pressure is set, a high membrane filtration flux can be obtained by injecting the modifier. The frequency of replacement is reduced, and as a result, membrane replacement costs are kept low. However, if the modifier is injected more than necessary, the cost of the modifier may exceed the reduction in the replacement cost of the separation membrane. Therefore, when the purpose is to minimize the operating cost of the membrane filtration device in this trade-off relationship, the operating cost may include the cost of the modifier and the replacement cost of the separation membrane. It is essential. In addition to the cost of the modifying agent and the replacement cost of the separation membrane, the operating cost of the membrane filtration device includes the blower for aeration, the power cost of the pump, and the sludge treatment cost. Therefore, it is preferable to search for a modifier injection amount that substantially minimizes the sum of the two costs. The cost of the modifier is calculated by the following equation (6), for example.

C _C = W _C × P _C / Q ′ (6)

Here, the cost of _{C C} is modifiers Circle ^/ m 3], _{W C} is the modifying agent injection amount [kg / day], the price of _{P C} is modifiers Circle / kg], Q 'is processed The amount of water [m ³ / day].

  On the other hand, the replacement cost of the separation membrane is calculated by the following equation (7), for example.

C _M = N _M × P _M / (Y _M × 365) / Q ′ (7)

Here, the _{C M} replacement cost of the separation membrane [yen ^/ m 3], _{N M} separation membrane sheets [sheets], _{P M} is the price of the separation membrane Circle / sheets], _{Y M} lifetime of the separation membrane Year ].

For example, when the amount of treated water in the membrane filtration device, the type and amount of components of the liquid to be filtered, and the membrane filtration flow rate (membrane filtration flux) are determined, the prediction step 1 and (6) to (7) ) Equation can be used to determine the modifier injection amount that minimizes the operating cost of the membrane filtration device and the operating conditions of the modifier injection amount adjusting means based on the modifier injection amount. For example, the membrane filtration pressure can be obtained by executing the prediction step 1 using the type and amount of components of the liquid to be filtered, the amount of optional modifier injected, and the membrane filtration flow rate (membrane filtration flux) as input data. Is calculated over time. Then, when the membrane filtration pressure reaches a predetermined value, the separation membrane is washed with chemicals. Here, let t (1) be the time to reach a predetermined value. Here, the number in parentheses represents the number of times of drug washing. When the chemical washing is performed, the non-membrane permeable substance accumulated in the separation membrane is removed, so that the membrane filtration pressure decreases. However, if the filtration is continued, the membrane filtration pressure increases and reaches a predetermined value again. Assuming that the time from when the chemical washing is performed until the membrane filtration pressure reaches a predetermined value is t (2), t (2) is smaller than t (1). This is because the non-membrane permeable substance accumulated in the separation membrane is not completely removed even after chemical washing. The above calculation is repeated, and the separation membrane is replaced when the time from when the (n-1) th chemical washing is performed until the membrane filtration pressure reaches a predetermined value falls below a predetermined time. Here, assuming that the time from when the chemical washing is performed until the membrane filtration pressure reaches a predetermined value is t (n), the lifetime Y _M of the separation membrane in the above equation (7) is t (1) + t (2) + ... + t (n) (see FIG. 5). Using the lifetime Y _M of the calculated separation membrane (7), the replacement cost C _M of the separation membrane was calculated, whereas using the modifier injection amount W _C a (6), the modifying agent the cost _{C C} is calculated, to calculate the sum of the _{C M} and _{C C.} Then, the modifying agent injection amount that minimizes the total cost is set as an appropriate value of the modifying agent injection amount and an operating condition of the appropriate modifying agent injection amount adjusting means. Here, a method of determining the predetermined value based on the maximum output value of the pump that generates the membrane filtration pressure, a method of determining the predetermined value based on the maximum energy consumption allowable value of the pump that generates the membrane filtration pressure, and the like are exemplified. In addition, there is a method of determining the predetermined time from labor required for chemical washing.

Further, if the correlation between the membrane filtration pressure and lifetime Y _M of the separation membrane is known in advance, in the calculations, predict the lifetime Y _M of the separation membrane from the calculated membrane filtration pressure, (7) using may calculate the replacement cost C _M of the separation membrane.

  Further, based on the result predicted as described above, it is determined whether or not the operating condition of the virtual modifier injection amount adjusting means (that is, the modifier injection amount) is right and wrong, and it is determined that the determination result is correct. From the determined operating conditions of the virtual modifier injection amount adjusting means, an appropriate operating condition of the modifier injection amount adjusting means is determined, and the determined appropriate modifier injection amount adjusting means Based on the operating conditions, the operating conditions of the modifier injection amount adjusting means may be determined. An example of such a determination method is the method shown in FIG. FIG. 6 is an example of a flowchart for carrying out the present invention. First, when using the prediction step 1, data representing a time-series change in the component amount of the liquid to be filtered, data representing a time-series change in the membrane filtration flow rate, and n arbitrary revisions. Prepare a value for the amount of material injection. The data representing the time-series change in the component amount of the liquid to be filtered and the data representing the time-series change in the membrane filtration flow rate are used. The value (modifier injection amount 1) is used. Based on these data and values, predict the fluctuation of the amount of constituents adhering to the separation membrane based on the prediction step 1, the fluctuation of the membrane filtration pressure, or the fluctuation of the membrane resistance, and based on the prediction result, Judge whether the amount of virtual modifier injection is right or wrong. Here, in the case where a parameter depending on the modifier injection amount is used instead of the modifier injection amount as input data for the prediction step 1, the correlation between the predetermined modifier injection amount and the parameter is used. The parameter value 1 for the modifier injection amount 1 is calculated, and the prediction step 1 is executed using the parameter value 1. Based on the prediction result, the virtual modifier injection amount is calculated. Judgment is made (Fig. 7). This determination result is referred to as determination result 1. Similarly, prediction is performed using the second to nth hypothetical modifier injection amounts, and whether the virtual modifier injection amount is right or wrong is determined based on the prediction result. ~ Obtain judgment result n. Among the determination results 1 to n obtained as described above, an appropriate modifier injection amount value is determined from the virtual modifier injection amount corresponding to the determination result determined to be good. .

  In addition, when using the prediction step 2, data representing a time-series change in the component amount of the liquid to be filtered, data representing a time-series change in the membrane filtration pressure, and n arbitrary modifications. Prepare a value for the amount of material injection. The data representing the time-series change in the component amount of the liquid to be filtered and the data representing the time-series change in the membrane filtration pressure are used. The value (modifier injection amount 1) is used. Based on these data and values, predict the fluctuation of the amount of constituents adhering to the separation membrane, the fluctuation of the membrane filtration flow rate, the fluctuation of the membrane filtration flux or the fluctuation of the membrane resistance based on the prediction step 2. Based on the prediction result, the right or wrong of the virtual modifier injection amount is determined. Here, in the case where a parameter depending on the modifier injection amount is used instead of the modifier injection amount as input data for the prediction step 1, the correlation between the predetermined modifier injection amount and the parameter is used. The parameter value 1 for the modifier injection amount 1 is calculated, and the prediction step 2 is executed using the parameter value 1. Based on the prediction result, the virtual modifier injection amount is calculated. Judgment is made (Fig. 7). This determination result is referred to as determination result 1. Similarly, prediction is performed using the second to nth hypothetical modifier injection amounts, and whether the virtual modifier injection amount is right or wrong is determined based on the prediction result. ~ Obtain judgment result n. Among the determination results 1 to n obtained as described above, an appropriate modifier injection amount value is determined from the virtual modifier injection amount corresponding to the determination result determined to be good. .

  In the present invention, when it is determined in the determination result that the value of the virtual modifier injection amount is not, the value of the virtual modifier injection amount is increased by a predetermined amount, and then again. From the value of the hypothetical modifier injection amount, the fluctuation of the amount of constituents adhering to the separation membrane, the fluctuation of the membrane filtration pressure, the fluctuation of the membrane filtration flow rate, the fluctuation of the membrane filtration flux, or the membrane resistance Predicting fluctuations, and determining whether or not the value of the virtual modifier injection amount after the predetermined amount increase is based on the prediction result; and, in the determination result, the virtual modifier injection When it is determined that the amount value is good, it is preferable that the virtual modifier injection amount value is an appropriate value for the modifier injection amount.

  For example, FIG. 8 (when the modifier injection amount is used as input data for the prediction step 1 or prediction step 2) and FIG. 9 (parameters depending on the modifier injection amount instead of the modifier injection amount are predicted step 1). (When used as input data for prediction step 2), an example of a flowchart in the case of including a change step 40 for changing the value of the virtual modifier injection amount is shown. Here, first, a sufficiently small value of the virtual modifier injection amount is prepared, and prediction is performed by the prediction step 1 or the prediction step 2. A virtual modifier injection amount is determined from the prediction result, and when the determination result is not valid, the value of the virtual modifier injection amount is increased by a predetermined amount in change step 40, and the prediction is performed again. Step 1 or prediction step 2 and subsequent steps are repeated. At this time, the value of the virtual modifier injection amount used when the determination result is positive is set as an appropriate value of the modifier injection amount. If such a procedure is followed, the appropriate value of the modifier injection amount can be determined more efficiently than the procedure shown in FIG. 6 or FIG. In addition, in the changing step 40, a virtual modifier injection amount between the value of the virtual modifier injection amount for which the determination result is good and the value of the virtual modifier injection amount for which the determination result is not valid. There is a method to gradually reduce the range estimated as the appropriate value of the modifier injection amount by changing to the value of, but if such a method is followed, the appropriate value of the modifier injection amount is finally determined. In doing so, it is preferable to follow the procedure shown in FIG.

  In addition, as a method of determining whether or not the value of the virtual modifier injection amount is appropriate based on the prediction result, the value of the amount of the constituent component adhered to the separation membrane within a predetermined time, the value of the membrane resistance When the value of the membrane filtration pressure is equal to or less than a predetermined value, or when the value of the membrane filtration flux or the value of the membrane filtration flow rate is equal to or greater than the predetermined value, the virtual reforming is performed. When the value of the agent injection amount is determined to be good, or when the value of the component amount adhering to the separation membrane, the value of the membrane resistance or the value of the membrane filtration pressure converges in the prediction result, It is preferable to judge that the correct amount of modifier injection is good. This makes it possible to objectively determine the proper value of the modifier injection amount.

  Here, the predetermined value needs to be set in advance, but the predetermined value is determined based on the maximum output value of the pump that generates the membrane filtration pressure, and the energy consumption of the pump that generates the membrane filtration pressure. A method of determining based on the maximum allowable value, a method of determining based on the maximum allowable value of the membrane resistance value and the membrane filtration pressure value to increase the membrane life and cleaning interval, and the pre-planned membrane filtration flow rate In order to satisfy, the method of determining based on the permissible minimum value of the value of the membrane filtration flux and the value of the membrane filtration flow rate is exemplified. For example, if the predetermined value is the maximum output value of the pump that generates the membrane filtration pressure, when the membrane filtration conditions are judged to be non-existing, the initially scheduled membrane filtrate volume cannot be obtained. When the condition is determined to be good, the expected amount of membrane filtrate can be obtained. In addition, there is a method of determining the predetermined time from labor required for chemical washing.

  In the present invention, the value of the modifying agent injection amount and the operating condition of the modifying agent injection amount adjusting means based on the value of the modifying agent injection amount are determined based on the determined appropriate value of the modifying agent injection amount. At this time, the appropriate value of the modifier injection amount may be determined as it is as the value of the modifier injection amount, and in the membrane filtration apparatus targeted by the present invention, the modifier injection amount is necessary and sufficient. If the amount is smaller than the amount, there is a risk of causing rapid film clogging. Therefore, to make it safer, the appropriate value of the modifier injection amount may be multiplied by a certain value as a safety factor, or a certain value may be added. Good.

  Of the determination results 1 to n, among the hypothetical modifier injection amounts corresponding to the determination results determined to be correct, the membrane filtration device calculated by the equations (6) and (7) It is good also considering the modifier injection amount that the operation cost becomes the smallest as the appropriate modifier injection amount.

  Further, in the present invention, based on the operating conditions of the membrane filtration device determined as described above, the operating conditions of the modifier injection amount adjusting means are controlled, and the operation method of the membrane filtration device, , Controlling at least the separation membrane, the modifier injection means for the filtrate, the modifier injection amount adjusting means for adjusting the injection quantity of the filtrate modifier, and the operation of the modifier injection quantity adjusting means. A membrane filtration apparatus having a modifier injection amount control means, wherein the control means controls the operation of the modifier injection amount adjustment means based on the operating conditions of the membrane filtration apparatus determined as described above. A membrane filtration device is provided. Here, it is preferable to control the modifier injection amount adjusting means using a timer provided with a timer capable of changing the modifier injection time of the modifier injection amount adjusting means.

Hereinafter, the present invention will be described in detail, but the present invention is not limited to only these examples.
Example 1
An outline of the structure of the membrane filtration device 500 used in this example is shown in FIG. The membrane filtration device 500 is a submerged membrane separation activated sludge device, and is a wastewater treatment device for treating factory wastewater containing acetic acid as a main component. The factory wastewater mainly composed of acetic acid, which is the raw water 508, was intermittently charged into the filtrate storage tank 501 having an average flow rate of 2.3 m ³ / d and an effective capacity of 2.3 m ³ . Activated sludge is accommodated in the filtrate storage tank 501 as the filtrate 504, the separation membrane 502 is immersed in the filtrate 504, and an aeration tube 509 is installed below the separation membrane 502. The air diffuser 509 is structured such that air is supplied from an aeration blower 503 as a cleaning means. That is, in this apparatus, since the air bubbles supplied from the air diffuser 509 come into contact with the membrane surface, and the flow of activated sludge due to aeration also occurs at the same time, it is possible to obtain the effect that the adhered components on the membrane surface peel from the membrane. It will be. The aeration air volume was set at 300 L / min. The water temperature was 20 ° C. As the separation membrane 502, a flat membrane type microfiltration membrane (effective membrane area 7.8 m ² ) made of PVDF (polyvinylidene fluoride) was used. Moreover, about the filtration, the suction pump installed in the membrane permeate side was performed using the membrane filtrate acquisition means 513. Moreover, filtration was performed continuously for 8 minutes, and after that, it stopped for 2 minutes and repeated this. The average membrane filtration flux was set at 1.0 m / day. Moreover, the modifier storage tank 505 which accommodated the modifier 506 was prepared, and the modifier was supplied to the to-be-filtered liquid storage tank 501 using the pump as the modifier injection means 507. As the modifier, a cationic flocculant (manufactured by Sanyo Chemical Co., Ltd.) was used. With respect to the pump, the injection amount of the modifier can be adjusted by adjusting the discharge time. In this example, in such a membrane filtration device, the modifier is injected in an amount of 0%, 0.5%, 1.0%, 1.5%, 2.0% per activated sludge weight. The amount of the injection of the modifier was calculated based on the result of calculating whether the operation cost would be the lowest when the amount was injected.

In this example, in order to comprehensively and easily define the entire liquid to be filtered, only the solid component amount X [gC / m ³ ] was used as the component amount of the liquid to be filtered. X is determined by measuring MLSS [g / m ³ ] of the filtrate according to the sewerage test method (3260 [g / m ³ ]), and assuming that its characteristic formula is C ₅ H ₇ O ₂ N, It was calculated by multiplying the measured value by 60/113. As a result, X was calculated to be 1731 [gC / m ³ ].

  Further, in this embodiment, instead of the data representing the amount of modifying agent injected, the parameter depending on the amount of modifying agent injected is used as an input, and the membrane filtration flux calculated as described above, the liquid to be filtered The fluctuation of the value of the membrane filtration flux was predicted by executing the prediction step 1 using the value of the component amount of As the parameters, a resistance coefficient (αx), a peeling coefficient (γx), and a friction coefficient (λx) were used. In order to execute the prediction step 1 using a parameter depending on the amount of modifier injection as an input, it is necessary to determine the correlation between the amount of modifier injection and the parameter. In this example, such a relationship was obtained as follows.

First, 100 ml of the filtrate 504 (activated sludge) was collected from the filtrate storage tank 501. The collected activated sludge was diluted 5 times with tap water and then divided into four beakers. Among the three beakers, a modifier was added at 2.0%, 1.0% by weight of the activated sludge. % And 0.5%. No modifier was added to the remaining one beaker. The four types of activated sludge are mixed with a separation membrane 502 (separation membrane area 4.1 × 10 ⁻⁴ m ² ) having the same material and shape as the separation membrane 502 used in the membrane filtration device 500. 10 ml, Amicon 8010 manufactured by Millipore Corporation, volume 12 ml) was introduced, and a membrane filtration test was performed using the apparatus shown in FIG. The membrane filtration pressure was set at 20 kPa, and the stirring speed by the magnetic stirrer 403 was set at 600 rpm. For filtration, first, continuous filtration was performed until the total filtrate volume reached 5 ml, then the valve 414 was closed and the filtration was stopped for 2 minutes, and then the valve 414 was opened to start the filtration again. Filtration was performed until the filtrate volume reached 7 ml.

From the relationship between the total filtrate amount per unit membrane area obtained by the above test and the membrane filtration resistance, the values of αx, γx, and λx were determined as follows. That is, by performing prediction step 2 with various values given to the three parameters, the fluctuation of the membrane filtration resistance is predicted, and the difference between the prediction result and the fluctuation of the actually obtained membrane filtration resistance is minimized. The value of the parameter is determined as the value of the parameter to be determined. In the prediction step 2, the equation (1) (here, μ uses the equation (2)) as the calculation step 10, and the equation (3) (provided that Pm = 0) is used as the calculation step 20. ) And the above equation (5) (where Pm = 0) was used as the calculation step 30. Regarding the initial membrane resistance Rm in the formula (5), the membrane filtration resistance is determined from the relationship between the total filtrate amount per unit membrane area and the membrane filtration resistance created from the result of the test conducted by filtering pure water. The average value (8.5 × 10 ¹⁰ [1 / m]) of the substantially constant portion was given. Further, a constant value 1 was given for the film cleaning power τ in the equation (3). Moreover, in the present Example, the membrane filtration prediction program was created using numerical calculation software MATLAB (made by USA Mathworks), and the said calculation was performed. Here, the Runge-Kutta method was used as the integration method.

  Table 1 shows the values of the resistance coefficient αx, the peeling coefficient γx, and the friction coefficient λx with respect to each modifier addition amount obtained as described above.

  From the results in Table 1, the correlation between the modifier addition amount and each parameter was formulated by exponential approximation as shown in the following equations (9), (10), and (11).

αx = 2.5 × 10 ¹¹ × exp (−1.7295 × W _P ) (9)
γx = 2.0 × 10 ³ (10)
λx = 4.2 × 10 ⁻⁵ × exp (0.0958 × W _P ) (11)

Here, W _P is a modifier addition amount per activated sludge wt%. In this example, by using the correlation between the modifier addition amount obtained as described above and the parameter, by executing the prediction step 1, the modifier is 0% per activated sludge weight, Variations in membrane filtration pressure values were predicted when added in amounts of 0.5%, 1.0%, 1.5%, and 2.0%. Here, the calculation step 10 uses the above formula (1) (here, μ uses the above formula (2)), and the calculation step 20 uses the above formula (3) (provided that Pm = 0). As the calculation step 30, the above formula (5) (where Pm = 0) was used. The calculation period was 3600 seconds. About the initial film | membrane resistance Rm in Formula (5), the value (8.5 * 10 < ¹⁰ > [1 / m]) calculated | required by the said test was given. Further, 2.0 × 10 ⁶ [m ³ / gC] was given as the reciprocal ηx of the density of the substance contained in the solid component.

  Further, the film cleaning power τ in the equation (3) was determined as follows. First, the membrane filtration apparatus 500 has a structure in which a tube is connected to the filtrate side of the separation membrane 502 and the membrane filtrate can be obtained by a water head difference using the siphon principle. In addition, a flow meter 510 (digital type) is installed in the middle of the tube so that a change with time in the indicated value of the flow meter 510 can be continuously recorded on the recorder 515. In addition, a valve 512 is installed downstream of the flow meter 510 so that intermittent operation can be performed manually. Then, the height of the tube outlet connected to the filtrate side of the separation membrane 502 was changed over time, and the membrane was intermittently filtered by opening and closing the valve 512 intermittently. Regarding the membrane filtration pressure, the difference between the water level of the filtrate in the filtrate receiving tank 501 and the height of the tube outlet (see FIG. 11) was ΔH, and the membrane filtration pressure was calculated from ΔH. The time-dependent change of the membrane filtration flow rate obtained under such membrane filtration pressure conditions was predicted using the prediction step 2. In the prediction step 2, the equation (1 ′) is used as the calculation step 10 (where μ is the equation (2)), and the equation (3) (provided that Pm = 0) as the calculation step 20. The equation (5) was used as the calculation step 30 by using a value obtained by substituting J = Q / A into The resistance coefficient αx, the peeling coefficient γx, and the friction coefficient λx were given the values in Table 1 without addition of a modifier. Then, the integral value of the difference between the calculation result of the change over time of the membrane filtration flow rate value and the change over time of the actual membrane filtration flow rate value at an arbitrary virtual membrane cleaning power value is calculated, and the integral value is minimized. As a result of obtaining the value of the film cleaning power to be 0.3, it was 0.3.

When the modifier is added in an amount of 0%, 0.5%, 1.0%, 1.5%, 2.0% per activated sludge weight by the above-described method, What predicted the fluctuation | variation of a value is shown in FIG. In this example, the average value of the membrane filtration pressure in the calculation period is calculated for each addition amount, and the correlation between the membrane filtration pressure determined in advance and the life Y _{M of the} separation membrane (hereinafter expressed by the equation (12)). From the above, a lifetime Y _M [year] of the separation membrane was calculated, and a replacement cost C _M [yen / m ³ ] of the separation membrane was further calculated using the equation (7).

Y _M = 7.7 × exp (−0.09 × ΔPave) (12)
Here, ΔPave is the average value [kPa] of the membrane filtration pressure during the calculation period.

On the other hand, the cost C _C [yen / m ³ ] of the modifier was calculated using the equation (6). However, since the agglomeration effect disappears in about 14 days after the modifier is injected into the membrane filtration apparatus, the modifier is injected every 14 days. The price _{P C} of the modifying agent was 1000 yen / kg]. Table 2 shows the separation membrane replacement cost C _M [yen / m ³ ], the modifier cost C _C [yen / m ³ ], and the total calculated for each addition amount in this manner.

  From this result, it was found that when the modifier was injected in an amount of 1.5% per activated sludge weight, the operating cost of the membrane filtration device was the lowest, so in this example, the above injection amount was obtained. Thus, it was decided to adjust the modifier injection amount.

(Example 2)
The structural outline of the membrane filtration device used in this example is exactly the same as the membrane filtration device 500 used in Example 1. In this example, in such a membrane filtration apparatus, the average membrane filtration flux is 1.0 [m / day] (8-minute filtration, 2-minute rest), and the allowable maximum value of the membrane filtration pressure is 10 [kPa]. When set, the percentage of the modifier to be injected per activated sludge weight (hereinafter referred to as the minimum modifier injection amount) was calculated, and the injection amount of the modifier was determined based on the result.

In this example, as in Example 1, in order to comprehensively and easily define the entire liquid to be filtered, only the solid component amount X [gC / m ³ ] is used as the component amount of the liquid to be filtered, and X = 1733 It was set as [gC / m < ³ >].

  Further, in this embodiment, instead of the data representing the amount of modifying agent injected, the parameter depending on the amount of modifying agent injected is used as an input, and the membrane filtration flux determined as described above, the liquid to be filtered The fluctuation of the value of the membrane filtration pressure was predicted by executing the prediction step 1 using the value of the constituent component amount of. As the parameters, a resistance coefficient (αx), a peeling coefficient (γx), and a friction coefficient (λx) were used. The liquid to be filtered and the separation membrane used in this example are the same as those used in Example 1. Therefore, in the present embodiment, the correlation between the modifier addition amount determined in the embodiment 1, the resistance coefficient αx, the peeling coefficient γx, and the friction coefficient λx (that is, the expressions (9), (10), and (11) ), The value of the initial film resistance Rm, the value of the inverse ηx of the density of the substance contained in the solid component, and the value of the film cleaning power τ were used as they were.

  Calculation of the minimum injection amount of the modifier was performed as follows. First, the injection amount of the modifier was set to 0% per activated sludge weight (that is, no modifier was added), the prediction step 1 was executed, and the fluctuation of the membrane filtration pressure value was predicted. In this case, since the membrane filtration pressure during the filtration time exceeded the allowable maximum value of the membrane filtration pressure, the predicting step 1 was performed again by increasing the injection amount of the modifier by 0.2% per activated sludge weight. Run and predict the variation of the membrane filtration pressure value. Also in this case, since the membrane filtration pressure during the filtration time exceeded the allowable maximum value of the membrane filtration pressure, the amount of modifier injected was further increased by 0.2% per weight of the activated sludge and predicted step 1 again. And the fluctuation of the membrane filtration pressure value was predicted. Repeating this calculation process, when searching for the amount of modifier injected with the membrane filtration pressure below the allowable maximum value of the membrane filtration pressure during the filtration time, when 1.2% was injected per activated sludge weight, For the first time, the membrane filtration pressure fell below the maximum allowable membrane filtration pressure (FIG. 13). Based on this, in this example, an injection amount of 1.2% per weight of activated sludge was set as the minimum injection amount of the modifier, and the modifier injection amount was adjusted so as to be the injection amount.

It is a flowchart in case the prediction step 1 is included in this invention. It is a flowchart in case the prediction step 2 is included in this invention. In this invention, it is a flowchart in the case of including the prediction step 1 or the prediction step 2, and using the parameter depending on a modifier injection quantity instead of the data showing the injection quantity of a modifier. In the present invention, including the prediction step 1 or the prediction step 2, the operation cost of the membrane filtration device is calculated based on the prediction result, and the operation condition of the proper modifier injection amount adjusting means is determined based on the operation cost. It is a flowchart in the case of doing. (7) is a diagram showing the relationship between the elapsed time and the membrane filtration pressure when predicting the lifetime Y _M of the separation membrane in the equation. In the present invention, the prediction step 1 or the prediction step 2 is included. Based on the prediction result, whether the operating condition of the virtual modifier injection amount adjusting means is right or wrong is determined, and the determination result is determined to be good. It is a flowchart in the case of determining the driving | running condition of an appropriate modifier injection amount adjustment means from the driving | running conditions of a virtual modifier injection amount adjustment means. FIG. 6 is a flowchart in the case of using, as input data for the prediction step 1 or the prediction step 2, a parameter that depends on the amount of the modifier injection instead of the amount of the modifier injection. It is a flowchart in the case of determining the operating condition of an appropriate modifier injection amount adjusting means including the prediction step 1 or the prediction step 2 and using the change step 40 in the present invention. FIG. 9 is a flowchart in the case of using, as input data for the prediction step 1 or the prediction step 2, a parameter that depends on the amount of the modifier injection instead of the amount of the modifier injection. It is the schematic of the test apparatus for determining the value of the parameter depending on the said modifier injection quantity. 1 is a schematic diagram of a membrane filtration device used in Example 1. FIG. It is a figure which shows the prediction result of the time-sequential change of the membrane filtration pressure obtained in Example 1. FIG. It is a figure which shows the prediction result of the time-sequential change of the membrane filtration pressure obtained in Example 2. FIG.

Explanation of symbols

400: Membrane filtration test apparatus 401: Stirring cell 402: Separation membrane 403: Magnetic stirrer 404: Stirrer 405: Nitrogen gas 406: Membrane fixing holder 407: Beaker 408: Electronic balance 409: Personal computer 410: Pure water chamber 411: Pressure Total 412, 413, 414: Valve 500: Membrane filtration device 501: Filtration liquid storage tank 502: Separation membrane 503: Aeration blower 504: Filtration liquid 505: Modification agent storage tank 506: Modification agent 507: Modification agent Injection means 508: Raw water 509: Aeration pipe 510: Flow meter 511: Record meter 512: Valve 513: Membrane filtrate acquisition means

Claims (16)

At least, using a membrane filtration apparatus having a separation membrane, a modifier injecting means for filtrate, and a modifier injection amount adjusting means for adjusting the injection amount of the modifier for filtrate, the filtrate to be filtered Filtration through a separation membrane, and a method for determining the operating conditions of a membrane filtration device that adjusts the injection amount of a modifier simultaneously or sequentially with filtration, comprising the type and amount of components of the liquid to be filtered, Predict fluctuations in the amount of constituents adhering to the separation membrane, fluctuations in membrane filtration pressure, or fluctuations in membrane resistance from data representing the time-series changes in the amount of modifier injected and the membrane filtration flow rate. A method for determining the operating condition of the membrane filtration device, wherein the operating condition of the modifier injection amount adjusting means is determined based on the prediction result. Separating the filtrate to be filtered using a membrane filtration device having at least a separation membrane, a modifier injection means for filtrate, and a modifier injection amount adjusting means for adjusting the injection amount of the filtrate modifier Filtration through a membrane, a method for determining the operating conditions of a membrane filtration device that adjusts the amount of the modifier injected simultaneously or sequentially with filtration, comprising the type and amount of components of the liquid to be filtered, From the data representing the time-series changes in the amount of modifier injected and the membrane filtration pressure, fluctuations in the amount of components adhering to the separation membrane, fluctuations in membrane filtration flow rate or membrane filtration flux, or membrane resistance And determining the operating condition of the modifying agent injection amount adjusting means based on the prediction result. A parameter that depends on a modifier injection amount is used instead of data representing the modifier injection amount, and a correlation between the parameter and the modifier injection amount is determined in advance. Item 3. A method for determining operating conditions of a membrane filtration device according to Item 1 or 2. 4. The method for determining operating conditions of a membrane filtration device according to claim 3, wherein the parameter is any one or more of a resistance coefficient, a peeling coefficient, and a friction coefficient. The value of the parameter is determined based on the change in the value of the membrane filtration pressure, the value of membrane filtration flux or membrane filtration flow rate, the value of membrane resistance after filtration of the liquid to be filtered through a test separation membrane. The method for determining the operating conditions of the membrane filtration device according to claim 3 or 4. In the membrane filtration test for determining the value of the parameter, the area of the test separation membrane is 0.01 m ² or less, and the amount of the liquid to be filtered for filtration through the test separation membrane is 100 ml or less. In addition, the liquid to be filtered for filtration by the test separation membrane is filtered after being diluted by 1.1 times or more, and the filtration pause period of 10 seconds or more is included once or more during the filtration, The method for determining operating conditions of a membrane filtration device according to claim 5, wherein any one or more of the conditions is used. Based on the data representing the time-series change of the component amount of the filtrate to be filtered, the data representing the time-series change of the membrane filtration flow rate, or the data representing the time-series change of the membrane filtration pressure, The method for determining operating conditions of a membrane filtration device according to any one of claims 1 to 6, further comprising a calculation step 100 for calculating the amount of constituent components adhering to the separation membrane. The calculation step 100 determines the amount of change in the amount of the component adhering to the separation membrane, the rate at which the substance contained in the component of the liquid to be filtered adheres to the separation membrane, and the rate at which the substance is separated from the separation membrane. The method for determining the operating conditions of the membrane filtration device according to claim 7, wherein the calculation step 101 is calculated based on a calculation expression expressed as a difference between the two.
The rate at which the substances contained in the constituents of the liquid to be filtered peel from the separation membrane is determined based on a higher-order expression of the second or higher order of the amount of the constituent components adhering to the separation membrane. The method for determining the operating conditions of the membrane filtration device according to claim 7. The method for determining the operating conditions of the membrane filtration device according to any one of claims 1 to 9, wherein a supernatant component and / or a solid component of the liquid to be filtered is used as a kind of constituent components of the liquid to be filtered. . The calculation formula expressing the amount of change in the amount of the supernatant component adhering to the separation membrane is a calculation formula including the term of the amount of the solid component adhering to the separation membrane. And / or the calculation expression expressing the amount of change in the amount of the solid component adhering to the separation membrane includes a term of the amount of the substance amount of the supernatant component adhering to the separation membrane. It is a calculation formula, The determination method of the operating condition of the membrane filtration apparatus of Claim 10 characterized by the above-mentioned. From the operating conditions of the virtual modifier injection amount adjusting means, the component amount of the filtered liquid adhering to the separation membrane, the fluctuation of the membrane filtration pressure, the fluctuation of the membrane filtration flow rate or the membrane filtration flux Alternatively, the fluctuation of the membrane resistance is predicted, and based on the prediction result, it is determined whether or not the operating condition of the virtual modifier injection amount adjusting means is appropriate, and the virtual result determined to be good in the determination result From the operating conditions of the modifier injection amount adjusting means, determine the operating conditions of the appropriate modifier injection amount adjusting means, based on the determined operating conditions of the appropriate modifier injection amount adjusting means, The method for determining the operating condition of the membrane filtration device according to any one of claims 1 to 11, wherein the operating condition of the modifying agent injection amount adjusting means is determined. In the determination result, when it is determined that the operating condition of the virtual modifier injection amount adjusting unit is not, the virtual modifier injection corresponding to the operating condition of the virtual modifier injection amount adjusting unit The virtual modifier injection amount adjusting means is changed to the operating condition of the virtual modifier injection amount adjusting means so that the value of the virtual modifier injection amount becomes larger than the amount value. From these operating conditions, the fluctuation of the amount of constituent components adhering to the separation membrane, the fluctuation of the membrane filtration pressure, the fluctuation of the membrane filtration flow rate or the membrane filtration flux, or the fluctuation of the membrane resistance are predicted, and the prediction result And determining whether or not the operating condition of the virtual modifier injection amount adjusting means after the change is appropriate, and in the determination result, the operating condition of the virtual modifier injection amount adjusting means is When it is determined to be correct, the virtual modifier injection amount adjusting means is operated. Method of determining the operating conditions of the membrane filtration device according to claim 12, characterized in that the matter and the operating conditions of the appropriate modifier injection amount adjusting means. In the prediction, the prediction is performed using the operating condition of the virtual modifier injection amount adjusting means, and in the prediction result, the value of the amount of component adhering to the separation membrane within a predetermined time, the membrane When the resistance value or the membrane filtration pressure value is equal to or less than a predetermined value, or when the membrane filtration flow rate value or the membrane filtration flux value is equal to or greater than a predetermined value, the virtual method of determining the operating conditions of the membrane filtration device according to claim 12 or 13, characterized in that determining the operating condition of the modifying agent injection amount adjusting means and Shi. In the prediction, the prediction is performed using the operating condition of the virtual modifier injection amount adjusting means, and in the prediction result, the value of the component amount adhering to the separation membrane, the value of the membrane resistance, the membrane the value of the filtration pressure, membrane filtration flow rate value or when the value of the membrane filtration flux is converged, claims, characterized in that to determine the operating conditions of the virtual modifiers injection amount adjusting means and Shi, method of determining the operating conditions of the membrane filtration device according to 12 or 13. At least the separation membrane, the modifier injection means for the filtrate, the modifier injection amount adjustment means for adjusting the injection quantity of the filtrate modifier, and the modification for controlling the operation of the modifier injection quantity adjustment means. It is a membrane filtration apparatus which has a mass agent injection amount control means, Comprising : The said modifier injection quantity control means was determined in the determination method of the operating condition of the membrane filtration apparatus in any one of Claims 1-15 A membrane filtration apparatus for controlling the operation of a modifier injection amount adjusting means.

Images