KR101108031B1 - Wastewater treatment system and method of treating wastewater based on pridicton control - Google Patents

Abstract

The present invention detects this before the disturbance of the influent water affects the process in advance and applies the prefeeding control so that the sewage and wastewater treatment process can be stably operated in real time, and the control variable is controlled to the optimal value through the preliminary prediction control. Therefore, it provides sewage and wastewater treatment system and sewage and wastewater treatment method which can reduce the operating cost. In addition, the sewage and wastewater treatment system and the sewage and wastewater treatment method of the present invention can prevent a control operation driving error by preselecting an adjustment variable necessary for eliminating disturbance and deriving a variation thereof, even in the absence of a driver. It is a ubiquitous state-of-the-art technology that can control process variations and enable operators to obtain information about the process anytime, anywhere.

Sewage & Wastewater Treatment, Feeding Control, Predictive Control

Description

Wastewater treatment system and method of treating wastewater based on pridicton control}

The present invention relates to a sewage and wastewater treatment system and a wastewater treatment method based on predictive control.

The sewage and wastewater treatment process has a great difficulty of stably treating influent with daily, weekly and seasonal fluctuations to meet the discharged water quality standards. The most optimal way to cope with fluctuations in inflow flow or load changes of contaminants in influent is to predict such fluctuations in advance and take steps to prevent them from affecting the process. To monitor and confirm. In this respect, conventional feedback control (FB control) techniques that mitigate the effects of disturbance after a disturbance, such as DO control, which is traditionally used in sewage and wastewater treatment processes, are applied after a control action is applied to the process. There is a drawback that the delay time occurs until the control operation affects the process, so that the complete elimination of disturbance cannot be performed.

In contrast, the feedforward control (FF control) technique, which detects disturbances affecting the process in advance and performs a control operation that can cancel the disturbances in advance, has a preliminary effect on the process. It is easier to secure process stability than the FB control technique in that it is blocked. In order to apply the FF control technique to the sewage and wastewater treatment process, the precondition that disturbance must be measured or predicted in advance is required.

Control technologies for sewage and wastewater treatment processes have been mentioned in various ways, ranging from simple DO control to real-time control technologies using automatic nutrient analyzers. However, in the form of existing technologies that have been applied in the latest automation control systems, most of them have a feedback control (FB) structure that controls the effects of disturbances in the process, which does not prevent the effects of disturbances in advance. There have been frequent cases where difficulties in stable process operation have occurred. Recently, a method of predicting and controlling a COD or TN item has been reported, but it is limited to unit item control and does not have an FF control structure based on a pre-scheduled scheduling. There was a limit that the value was not clear and the disturbance effect was not fully controlled.

The present invention is to solve the above problems, by using the artificial neural network as a statistical model in advance to predict the future time fluctuations of the influent in advance, using the activated sludge model as a mathematical process model of the runoff according to the influent fluctuations predicted The purpose of the present invention is to provide a sewage / wastewater treatment system and a sewage / wastewater treatment method capable of blocking the effects of disturbances in advance by predicting the change of the predicted value and then applying the control action in advance at the appropriate time when the predicted result is required to control. It is done.

As a means for achieving the above object,

The present invention is a sewage and wastewater treatment system in which the inflow water is introduced into the reaction tank and then treated in the reaction tank and discharged into the effluent, which is installed in at least one of the inflow water, the effluent and the reaction tank, and measures the concentration of the pollutant in real time or at predetermined time intervals. Pollutant concentration measuring sensor; A database storing concentration values measured from the pollutant concentration sensor; An influent concentration prediction unit for predicting an influent contamination concentration value of a predetermined future time using the concentration value data of the database; An effluent concentration predicting unit predicting an effluent pollutant concentration value of a predetermined future time by using a future influent pollutant concentration value predicted by the influent concentration predicting unit; And a treatment process control unit controlling the sewage treatment process by selecting a control variable when the future pollutant concentration value predicted by the runoff concentration prediction unit is out of a standard concentration range.

In addition, the influent concentration prediction unit provides a sewage and wastewater treatment system, characterized in that for predicting the concentration of the influent using an artificial neural network model.

In addition, the effluent concentration prediction unit provides a sewage and wastewater treatment system, characterized in that for predicting the concentration of the effluent using the activated sludge model.

In addition, the effluent concentration prediction unit provides a sewage and wastewater treatment system, characterized in that it predicts the effluent concentration by using a combination model of activated sludge model No. 3 and EAWAG Bio-P module for phosphorus behavior simulation.

In addition, the pollutant provides a sewage and wastewater treatment system, characterized in that at least one selected from NH4 ⁺ -N, NO _X -N and PO ₄ ^3- -P.

In addition, the pollutant concentration measurement sensor provides a sewage and wastewater treatment system, characterized in that the measurement by 0.5 to 2 hours.

In addition, the influent concentration prediction unit provides a sewage and wastewater treatment system, characterized in that for predicting the influent pollutant concentration value of the future time in the 12 to 36 hours range.

In addition, the effluent concentration prediction unit provides a sewage and wastewater treatment system, characterized in that for predicting the effluent pollutant concentration value of the future time in the 12 to 36 hours range.

In addition, the control parameter provides a sewage / wastewater treatment system, characterized in that at least one of an external carbon source flow rate, a reaction tank internal transfer flow rate, a sludge transfer flow rate, a waste sludge flow rate, and an aerobic bath dissolved oxygen concentration are selected.

In addition, the treatment process control unit provides a sewage and wastewater treatment system, characterized in that for selecting the control parameters including the external carbon source flow rate, the internal conveyance flow rate when the concentration value of NO _x -N exceeds the standard concentration range.

In addition, the treatment process control unit provides a sewage and wastewater treatment system, characterized in that for selecting a control parameter including the internal transport flow rate, aerobic dissolved oxygen concentration when the concentration value of NH4 ⁺ -N exceeds the standard concentration range.

In addition, the treatment process control unit selects a control variable including the internal transport flow rate, sludge return flow rate, and aerobic dissolved oxygen concentration when the concentration value of PO ₄ ^3- -P exceeds the reference concentration range Provide a wastewater treatment system.

In addition, a sewage and wastewater treatment system in which the inflow water is introduced into the reaction tank and then treated in the reaction tank and discharged into the effluent, which is installed in at least one of the inflow water, the effluent and the reaction tank to measure the concentration of the pollutant in real time or at predetermined time intervals. Pollutant concentration sensor; An effluent concentration predicting unit predicting an effluent pollutant concentration value of a predetermined future time by using the concentration value measured by the pollutant concentration measuring sensor; And a process control unit for controlling the wastewater treatment process by selecting a control variable when the future pollutant concentration value predicted by the runoff concentration predictor is outside the standard concentration range, wherein the control variable is an external carbon source to the reactor. It provides a sewage and wastewater treatment system, characterized in that at least one or more of the flow rate, the reaction tank internal conveyance flow rate, sludge conveyance flow rate, waste sludge flow rate, and aerobic tank dissolved oxygen concentration is selected.

In addition, the treatment process control unit selects a control variable including the external carbon source flow rate, the internal conveyance flow rate when the concentration value of NO _X -N of the pollutant exceeds the standard concentration range, the NH4 ⁺ -N concentration value of the pollutant is the reference concentration When the range is exceeded, the control parameters are selected, including the internal return flow rate and the aerobic dissolved oxygen concentration.When the concentration of PO ₄ ^3- -P in the pollutant exceeds the standard concentration range, the internal return flow rate, sludge return flow rate, and aerobic tank It provides a sewage and wastewater treatment system characterized by selecting a control variable including the dissolved oxygen concentration.

In addition, sewage and wastewater treatment method in which the inflow water flows into the reaction vessel and is treated in the reaction vessel and flows out into the effluent, the real-time or at predetermined time intervals from the source concentration measurement sensor installed in at least one of the inflow, effluent and reaction vessel Measuring the concentration; Predicting an influent pollutant concentration value at a predetermined future time using the measured concentration value data; Predicting a effluent pollutant concentration value at a predetermined future time using the predicted future influent pollutant concentration value; And controlling the wastewater treatment process by selecting a control variable when the predicted future time effluent pollutant concentration value is out of a standard concentration range.

Further, the runoff of the source pixel value of a predicted future time when the concentration values of the NO _X -N concentration exceeds the reference range, select the control parameters, including the external carbon source flow rate, internal recycle flow rate and the concentration values of the NH4 ⁺ -N reference When the concentration range is exceeded, a control parameter is selected, including the internal transport flow rate and the aerobic dissolved oxygen concentration, and the process control unit controls the internal transport flow rate and the sludge return when the concentration value of PO ₄ ^3-- P exceeds the standard concentration range. It provides a sewage and wastewater treatment method characterized by selecting a control variable including the flow rate, and aerobic dissolved oxygen concentration.

The present invention detects this before the disturbance of the influent water affects the process in advance and applies the prefeeding control so that the sewage and wastewater treatment process can be stably operated in real time, and the control variable is controlled to the optimal value through the preliminary prediction control. Therefore, the operation cost can be reduced. In addition, it is possible to prevent a control operation driving error by preselecting an adjustment variable necessary for eliminating disturbance and deriving a variation thereof. The series of preliminary predictive control techniques proposed by the present invention can control the process variation even in the absence of an operator through automation, and can be referred to as a ubiquitous advanced technology that enables the operator to obtain information on a process anytime and anywhere.

Hereinafter, the present invention will be described in more detail with reference to the drawings and examples. The following descriptions are for specific examples of the present invention, but are not intended to limit the scope of the rights set forth in the claims, even if there is an assertive or limited expression.

1 is a schematic configuration diagram of a sewage and wastewater treatment system according to an embodiment of the present invention, and FIG. 2 is a data flow and configuration diagram of a preliminary predictive control technique of a sewage and wastewater treatment system according to an embodiment of the present invention. .

The sewage / wastewater treatment system according to an embodiment of the present invention is a sewage / wastewater treatment system in which inflow water flows into a reaction vessel and then is treated in a reaction vessel and flows out into effluent, and is installed in at least one of the inflow water, effluent water and a reaction vessel in real time. Or a source concentration measurement sensor for measuring the concentration of the source at predetermined time intervals, a database storing the concentration value measured from the source concentration measurement sensor, and an influent water source at a predetermined future time using the concentration value data of the database. An influent concentration predictor for predicting a concentration value, an effluent concentration predictor for predicting a effluent pollutant concentration value at a predetermined future time using a future influent pollutant concentration value predicted by the influent concentration predictor, and the effluent concentration predictor When the estimated future source concentration value is outside the standard concentration range Selecting a right control variable is characterized in that it comprises a process control unit for controlling the wastewater treatment process.

At least one or more of the influent, the effluent and the reaction tank is provided with a pollutant concentration measuring sensor capable of measuring the concentration of the pollutant in real time or at predetermined time intervals. The reactor can be divided into a number and a variety of known reactors can be selected and are not limited. As an example, as shown in FIGS. 1 and 2, as the first settler and the biological reaction tank which is a general nutrient removal process, when the sewage from which the solid matter is first submerged in the first settler is removed and flows into the oil reactor. , Anaerobic tanks for the purpose of releasing phosphorus by microorganisms, one or more anoxic tanks for the denitrification of nitrites using carbon sources contained in the raw sewage, and organic carbon materials used for denitrification and ammonia contained in raw water. One or more aerobic tanks for the purpose of oxidation to nitrites, secondary settling tanks for solid-liquid separation of microorganisms and treated water, and the like. In addition, as shown, the inner conveyance for returning the nitrite produced in the aerobic tank may be performed. The pollutant concentration measuring sensor may be installed in each reactor, or may be selectively installed. The source concentration measurement sensor may measure in real time, and may measure the concentration of the source at a predetermined time, for example, at intervals of minutes or hours. Preferably, the pollutant concentration sensor is preferably measured in units of 0.5 to 2 hours. The concentration value of the pollutant measured from the pollutant concentration sensor may be stored in a database.

Although not limited to the source of contamination, preferably at least one of NH 4 ⁺ -N, NO _X -N and PO ₄ ^3- -P may be selected.

The influent concentration prediction unit predicts the influent contaminant concentration value at a predetermined future time using the concentration value data of the database. Thereafter, the effluent contaminant predicts the effluent contaminant value at a predetermined future time using the future influent contaminant concentration value predicted by the influent concentration predictor. Although not limited to the predetermined future time, it is good to predict the influent pollutant concentration value in the future time range of 12 to 36 hours.

The influent concentration predictor is not limited, but it is preferable to predict the influent concentration using an artificial neural network model, and the effluent concentration predictor may predict the concentration of the effluent using an activated sludge model. In particular, it is recommended to estimate the effluent concentration using a model that combines activated sludge model No. 3 and EAWAG Bio-P module for phosphorus behavior simulation.

The present inventor-based preliminary predictive control technology utilizes an artificial neural network, which is a statistical model, to predict the inflow water quality after one day in advance, and then applies it to the activated sludge model, which is a mathematical model, to predict the outflow water quality, and to advance the timing at which the control action is required. To detect. The type of control operation (selection of control variables) and their size are determined by a qualitative table between the control variables and the control variables, which are not only the optimal control variables associated with a single water quality change but also various water quality changes. By presenting the priority of the control variable that can cancel the, it is possible to provide the latest control logic form of the MIMO (Multiple Input Multiple Output) form, rather than the conventional single input single output (SISO) form of the simple control logic. The selected control variables are pre-evaluated for the control effect by various simulations that can minimize the effects of disturbance, and the control schecule is composed by selecting the control variables and the amount of change that resulted in the best effluent quality. As such, the control time and the amount of change in the quantified control variable are configured in one day unit (1 hour unit), thereby applying FF control, thereby minimizing the influence of disturbance on the process.

An artificial neural network used to predict the influent flow rate and concentration used as an example in the present invention is a type of black box model generally used for pattern learning using artificial intelligence, and the most widely applied model is a multilayer perceptron artificial neural network. The multilayer perceptron neural network is composed of an input layer, an output layer, and a hidden layer between them. Training done to calculate an estimate for the desired target variable proceeds from the input layer to the output layer. By calculating the output value for a given input, and passing the error between the calculated output value and the target output value in the reverse direction, by adjusting the weight and bias connecting the nodes between the layers, You will tune the neural network to calculate the output value for the pattern. By repeating this process, the training process is terminated when the error between the output value of the artificial neural network and the target output value for a given input becomes smaller than the given limit value. The completed neural network expresses the correlation between the input value and the output value used for learning, and outputs the output maintaining the proper correlation even for the input that is not used for learning.

The activated sludge models (ASMs) will be described below. Recently, in order to maximize the nutrient removal rate of the target process or to maintain stable effluent quality, a process control for adjusting operation factors such as aeration amount, sludge return, and internal return rate for the reaction tank is required. Determining the appropriate process control method is based on experienced operator knowledge or difficult traditional control theory. However, since activated sludge process has strong biological characteristics, it is different from the control theory applied to physicochemical process control. Difficulties arise in applying control to the plant. At this time, a more rational control method can be calculated by a mathematical model representing the behavior of the removal reaction by the microorganisms in the process. Mathematical models, which have been actively studied since the 1980s, are actively used for process behavior prediction and process design using ASMs (ASM1, ASM2, ASM2d, ASM3) developed at IWA.

In order to use mathematical models for process behavior prediction and control, a model optimization process that involves model selection and model parameter estimation and verification of actual data among various mathematical models is required. Models constructed in this way can be used to improve sewage treatment plants (Coen et al. , 1996) or to evaluate the performance of processes, such as assessing the potential for improved biological removal from current installations (Ladiges et al. , 1999) . Or the situation is widely used in design. The present invention is in need of automatic control and manager control of the sewage treatment plant for stable and economical removal of nutrients from the sewage treatment plant, and this model is useful as a tool for performing process diagnosis and process behavior prediction for manager control. Applied.

On the other hand, the treatment process control unit controls the wastewater treatment process by selecting a control variable when the future pollutant concentration value predicted by the effluent concentration prediction unit is out of the standard concentration range using the model or the like.

Control parameters of the sewage / wastewater treatment system are not limited, but as shown in FIG. 1, the external carbon source flow rate, the internal transfer flow rate of the reactor, the sludge transfer flow rate, the waste sludge flow rate, and the aerobic tank dissolved oxygen concentration (this is caused by a blower) Can be controlled).

In particular, the present invention derives the following as control variables that can be controlled by controlling the pollution process, that is, optimized for each pollution source, that is, control variables.

That is, it is preferable that the treatment process controller selects an external carbon source flow rate and an internal conveyance flow rate control variable when the NO _x -N concentration value in the predicted effluent concentration value exceeds the reference concentration range. That is, by increasing the external carbon source flow rate and the internal conveyance flow rate, it is possible to prevent the near future runoff concentration value from exceeding the reference range. It is also possible to include other control parameters, but at the expense of additional process time and cost.

When the concentration value of NH4 ⁺ -N exceeds the standard concentration range, it is preferable to select a control variable including the internal return flow rate and the aerobic dissolved oxygen concentration, and the concentration value of PO ₄ ^3- -P may exceed the standard concentration range. In this case, it is desirable to select a control variable including internal return flow rate, sludge return flow rate, and aerobic dissolved oxygen concentration.

If the pollutant exceeds two or more concentration values, it is better to combine the control parameters to control the process. The specific correlation between the control variable and the control variable is shown in Table 1 below.

TABLE 1

Item NOx NH4 PO4 NH4 NOx PO4 NH4 PO4 NOx PO4 NH4 NOx Qex, Qir, Qr, Qw, DO One Qir, Qr, DO One One Qex, Qir, Qr, DO 2 One Qir, DO One 2 2 Qex, Qir, DO 2 One Qex, Qir One Qr, DO 2 Qex, Qir 2 Qex, DO 2

Where Qex is the external carbon source flow rate, Qir is the internal return flow rate, Qr is the sludge return flow rate, Qw is the waste sludge flow rate, DO is the aerobic tank dissolved oxygen, number 1 is the priority control variable, and number 2 is the priority control variable. )

In the present invention, the limitation of "long-term stable process operation" caused by the application of simple feedback control, a form of post-control, which has traditionally been applied to sewage and wastewater treatment processes, does not completely eliminate the effects of disturbances on the process. In addition, it is possible to comprehensively control the correlation between various control variables (here, control variables refer to nutrients in the influent) beyond the control of a single control variable, which is a simple control form, and the control operation. By selecting a control variable and presenting an accurate change value of the selected control variable, it is possible to prevent an inappropriate selection of the control variable and a driving error of the control operation. In addition, by scheduling the series of control operation points in advance, the effects of inflow disturbances are completely eliminated in advance so that the sewage and wastewater treatment process can be normally operated even when the driver is absent.

The present invention also provides a sewage / wastewater treatment method in which inflow water flows into a reaction tank and then is treated in a reaction tank and flows out into an outflow water. Measuring the concentration of the contaminant, estimating the influent contaminant concentration value at a predetermined future time using the measured concentration value data, and using the predicted future influent contaminant concentration value, the effluent contaminant concentration at a predetermined future time It provides a sewage and wastewater treatment method comprising the step of predicting the value, and controlling the sewage treatment process by selecting a control variable if the predicted future effluent pollutant concentration value is out of the standard concentration range. The contaminant concentration sensor and the concentration measurement have been described in detail above and thus will be omitted. In addition, other processes may be added in the above process. For example, as described above, the concentration measurement may be stored in a database.

3 is a flow chart of a wastewater treatment method according to an embodiment of the present invention. As shown, first, the measured concentration value data is used to predict the influent pollutant concentration value at a predetermined future time. The predicted future influent source concentration value is then used to predict the effluent source concentration value at a predetermined future time. Then, if the estimated future time runoff contaminant concentration values fall outside the standard concentration range, the control parameters are selected according to a predetermined schedule. Then, the wastewater treatment process is controlled by performing a feed-back control process schedule according to the control parameters of each pollutant.

FIG. 4 is a diagram illustrating the inflow and outflow concentrations in this manner, and a time point for applying the control. FIG. 5 is a view illustrating the inflow and outflow concentrations after the control is applied. As shown, the preliminary prediction of the concentration can be seen to prevent the concentration of the effluent from exceeding the reference value.

1 is a schematic diagram of a sewage and wastewater treatment system according to an embodiment of the present invention;

2 is a data flow and configuration diagram of a predictive control technique of the sewage and wastewater treatment system according to an embodiment of the present invention;

3 is a flow chart of the sewage and wastewater treatment method according to an embodiment of the present invention,

4 is a graph showing the time point of applying the control through the influent prediction and the effluent prediction and pre-prediction,

FIG. 5 is a graph illustrating a case where the sewage and wastewater are stably discharged through the predictive control.

Claims (16)

delete delete delete delete delete delete delete delete delete delete delete delete delete A sewage and wastewater treatment system in which inflow water flows into a reaction tank and then is treated in a reaction tank and flows out into an outflow water, A source concentration measurement sensor installed in at least one of the inflow water, the outflow water, and the reaction tank to measure the concentration of the pollution source in real time or at predetermined time intervals; An effluent concentration predicting unit predicting an effluent pollutant concentration value of a predetermined future time by using the concentration value measured by the pollutant concentration measuring sensor; And And a treatment process controller configured to control the wastewater treatment process by selecting a control variable when the future pollutant concentration value predicted by the effluent concentration prediction unit is out of a standard concentration range. The control parameter is characterized in that at least one selected from the flow rate of the external carbon source to the reactor, the internal transport flow rate of the reactor, sludge transport flow rate, waste sludge flow rate, and aerobic dissolved oxygen concentration, The process control unit selects a control variable including an external carbon source flow rate and an internal return flow rate when the NO _x -N concentration value of the pollutant exceeds the standard concentration range, and the NH4 ⁺ -N concentration value of the pollutant is the reference concentration range. When exceeded, control parameters are selected, including internal return flow rate and aerobic dissolved oxygen concentration, and when the concentration of PO ₄ ^3- -P in the source exceeds the standard concentration range, internal return flow rate, sludge return flow rate, and aerobic dissolved oxygen concentration A sewage and wastewater treatment system comprising selecting a control variable including concentration. delete Sewage and wastewater treatment method in which inflow water flows into the reaction tank and then is treated in the reaction tank and flows into the effluent, Measuring the concentration of the pollutant in real time or at predetermined time intervals from a pollutant concentration sensor installed in at least one of the influent, the effluent and the reaction tank; Predicting an influent pollutant concentration value at a predetermined future time using the measured concentration value data; Predicting a effluent pollutant concentration value at a predetermined future time using the predicted future influent pollutant concentration value; And And controlling the wastewater treatment process by selecting a control variable when the predicted future time effluent pollutant concentration value is out of a standard concentration range. If the concentration value of NO _X -N exceeds the standard concentration range among the estimated future runoff pollutant concentration values, select the control variable including the external carbon source flow rate and the internal return flow rate, and the NH4 ⁺ -N concentration value is the reference concentration range. The control parameters are selected, including the internal transport flow rate and the aerobic dissolved oxygen concentration when exceeding, and the process control unit controls the internal transport flow rate, the sludge return flow rate, and when the concentration of PO ₄ ^3- -P exceeds the standard concentration range. And an aerobic tank dissolved oxygen concentration for selecting a control variable.

Images