WO2014157751A1 - Energy-saving prediction system and method based on diagnosis of state of energy consumption in sewage treatment plant - Google Patents

Energy-saving prediction system and method based on diagnosis of state of energy consumption in sewage treatment plant Download PDF

Info

Publication number
WO2014157751A1
WO2014157751A1 PCT/KR2013/002580 KR2013002580W WO2014157751A1 WO 2014157751 A1 WO2014157751 A1 WO 2014157751A1 KR 2013002580 W KR2013002580 W KR 2013002580W WO 2014157751 A1 WO2014157751 A1 WO 2014157751A1
Authority
WO
WIPO (PCT)
Prior art keywords
energy consumption
data
energy
diagnosis
sewage treatment
Prior art date
Application number
PCT/KR2013/002580
Other languages
French (fr)
Korean (ko)
Inventor
김창원
김예진
김효수
김민수
박문화
신원재
Original Assignee
부산대학교 산학협력단
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 부산대학교 산학협력단 filed Critical 부산대학교 산학협력단
Publication of WO2014157751A1 publication Critical patent/WO2014157751A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Electricity, gas or water supply

Definitions

  • the present invention relates to an energy saving prediction system and method according to the diagnosis of energy consumption in sewage treatment plants.
  • it performs qualitative diagnosis of energy consumption caused by the operation of equipment such as pumps and blowers, which are essential for the operation of sewage treatment plants, and provides a set of operating conditions that can save energy according to the diagnosis results. Therefore, the present invention relates to an energy saving prediction system and method according to the diagnosis of energy consumption status of a sewage treatment plant, which can provide a practical operation setpoint for reducing energy use by confirming the reduction of energy consumption accordingly.
  • sewage treatment plant Since the sewage treatment plant is responsible for the final treatment in a series of water treatment facilities, it has been generally recognized as a resident hate and energy consumption facility because sewage generated after human activities and industrial systems are driven is a major inflow factor. Against this background, when untreated sewage is discharged to the discharge system, it causes water pollution and, in the past, causes a reduction in the human resources available to humans. Maintaining treatment performance has been set as the primary operational goal for sewage treatment plants.
  • renewable energy technologies that can make full use of the sites covered by sewage treatment plants.
  • Renewable energy technologies such as solar, hydropower and biogas are those that can produce enough energy even with a small amount of land in a sewage treatment plant or with a slight modification of the treatment stream.
  • a high-efficiency device there is also a way to reduce the energy consumption by ensuring higher performance even at the same power.
  • the initial investment costs for installation and replacement of facilities are incurred, in the long term, the introduction of renewable energy technology and high efficiency devices can be considered as one of the efficient methods for energy saving.
  • the energy consumption of the sewage treatment plant is reduced by appropriately adjusting the set values of operating variables within the range that does not affect the effluent quality.
  • the present invention has been made to solve such a problem, the present invention is a system and method for predicting the quantitative energy saving value by diagnosing the state of consumption of energy in the sewage treatment plant and deriving the energy saving operating conditions for it
  • the present invention relates to a method for analyzing energy consumption by performing multivariate statistical analysis on data on operating variables and weather conditions associated with major devices related to energy used in sewage treatment plants.
  • the energy consumption status of sewage treatment plant that can provide the operating conditions to reduce the energy consumption of the plant by calling the operating conditions that can save energy and applying them to the regression analysis model that can predict the energy consumption.
  • Energy saving prediction system by diagnosis To provide a method has its purpose.
  • the data call unit for calling data necessary for the diagnosis of energy consumption from the database for storing the inflow / outflow water quality data and process operation data of the sewage treatment plant Receives the data called from the data caller and applies the multivariate statistical analysis method to the called data to extract information on the energy consumption state inherent in the called data and to extract the information on the energy consumption state.
  • An energy consumption state diagnosis unit for deriving an energy consumption state diagnosis result from the;
  • An energy saving operation condition calling unit for calling a preset energy saving operation condition according to a group representing a diagnosis result derived from the energy consumption state diagnosis unit;
  • An energy consumption reduction prediction unit for predicting an energy saving amount by applying a polynomial regression analysis method according to the operation condition called from the energy saving operation condition calling unit;
  • energy saving operation condition applying unit for applying an operating condition predicted by the energy consumption reduction amount estimator to an actuator including a pump and a blower in the sewage treatment plant; Provide a prediction system.
  • the data call unit is characterized in that the main data automatically associated with the energy consumption of all the data accumulated in the database that is operated separately in the individual sewage treatment plant.
  • the energy consumption state diagnosis unit for the called data for extracting the information on the energy consumption state inherent in the called data for reducing the called data;
  • the energy consumption state diagnosis result of the called data is derived using the discrimination function derived by the discrimination function extracting unit.
  • the energy saving operation condition call unit DO dissolved oxygen
  • sludge conveyed flow rate and surplus for the group with a higher energy consumption state than the other groups according to the group means the diagnosis result derived from the energy consumption state diagnosis unit It is characterized in that a change in the operating conditions is called by setting in advance an operating condition capable of saving energy, including reducing the sludge flow rate.
  • the polynomial regression analysis technique is a model technique that utilizes the same operation variable and the data called by the data call unit, and predicts the energy consumption by dividing the operation variable into independent and dependent variables.
  • a data call step of calling data necessary for the diagnosis of energy consumption from the database for storing the inflow / outflow water quality data and process operation data of the sewage treatment plant Receives the called data and applies the multivariate statistical analysis method to the called data to extract information on the energy consumption state inherent in the called data and to extract the energy consumption state from the information on the extracted energy consumption state.
  • An energy consumption state diagnosis step of deriving a diagnosis result An energy saving operation condition calling step of calling an operation condition capable of saving energy in advance according to the group representing the derived diagnosis result;
  • the step of diagnosing the energy consumption state to reduce the called data to extract information on the energy consumption state inherent in the called data for the called data, and the current energy using the reduced data
  • the abbreviated function is grouped using Fisher's linear discriminant analysis to be used as a means to determine which group the new data belongs to. And deriving an energy consumption state diagnosis result of the called data using the derived discrimination function.
  • the present invention proposes an energy saving prediction model in addition to providing information on qualitative energy use state, thereby providing a numerical value that can be quantitatively reduced from the amount of energy currently used by using an adjustable operating variable setpoint as an input variable.
  • the energy saving prediction model according to the present invention has a model formula consisting of process operation variables associated with the amount of energy used in the sewage treatment plant, so that the process operator can arbitrarily change the numerical value of the operation variable to simulate the effect. In this way, it is possible to provide an educational aspect to process operators about process operation methods that can save energy, and to evaluate process behavior in advance, thereby enabling more efficient energy saving operation.
  • Figure 1 is a block diagram showing an energy saving prediction system according to the diagnosis of energy consumption state sewage treatment plant according to an embodiment of the present invention.
  • FIG. 2 is a flow chart showing a method for predicting energy saving according to the diagnosis of energy consumption in a sewage treatment plant according to an embodiment of the present invention.
  • FIG. 3 is a graph showing the results of the sewage treatment plant energy saving prediction model according to an embodiment of the present invention.
  • FIG. 4 is a graph for verifying the results of the sewage treatment plant energy saving prediction model of FIG. 3.
  • FIG. 5 is a graph showing a result of predicting the energy consumption before and after the change in the operating conditions when the energy consumption state of the sewage treatment plant according to the embodiment of the present invention is medium.
  • FIG. 6 is a graph showing a result of predicting the energy consumption before and after the change in the operating conditions when the energy consumption state of the sewage treatment plant according to an embodiment of the present invention is high.
  • the data call unit for calling data necessary for the diagnosis of energy consumption from the database for storing the inflow / outflow water quality data and process operation data of the sewage treatment plant Receives the data called from the data caller and applies the multivariate statistical analysis method to the called data to extract information on the energy consumption state inherent in the called data and to extract the information on the energy consumption state.
  • An energy consumption state diagnosis unit for deriving an energy consumption state diagnosis result from the;
  • An energy saving operation condition calling unit for calling a preset energy saving operation condition according to a group representing a diagnosis result derived from the energy consumption state diagnosis unit;
  • An energy consumption reduction prediction unit for predicting an energy saving amount by applying a polynomial regression analysis method according to the operation condition called from the energy saving operation condition calling unit;
  • energy saving operation condition applying unit for applying an operating condition predicted by the energy consumption reduction amount estimator to an actuator including a pump and a blower in the sewage treatment plant; Provide a prediction system.
  • FIG. 1 is a block diagram showing an energy saving prediction system according to the sewage treatment plant energy consumption diagnosis according to an embodiment of the present invention
  • Figure 2 is an energy saving prediction method according to the diagnosis of energy consumption state sewage treatment plant according to an embodiment of the present invention.
  • 3 is a graph showing the result of the sewage treatment plant energy saving prediction model according to an embodiment of the present invention
  • FIG. 4 is a graph for verifying the result of the sewage treatment plant energy saving prediction model of FIG. 3, and FIG.
  • FIG 5 is a graph showing the prediction result of the energy consumption before and after the change in the operating conditions when the energy consumption state of the sewage treatment plant according to the embodiment of the present invention is medium
  • Figure 6 is the energy consumption state of the sewage treatment plant according to the embodiment of the present invention In the case of high, it is a graph showing the prediction result of energy consumption before and after the change of operating conditions.
  • the energy saving prediction system 10 according to the sewage treatment plant energy consumption diagnosis according to the present invention includes a data call unit 100, an energy consumption state diagnosis unit 200, and an energy saving operation condition call unit 300. ), The energy consumption reduction amount prediction unit 400 and the energy saving operation condition applying unit 500.
  • the data call unit 100 serves to call data necessary for diagnosing an energy consumption state from a database storing inflow / outflow water quality data and process operation data of the sewage treatment plant.
  • the inflow / outflow water quality data includes inflow / outflow flow rate and inflow / outflow component concentration (BOD 5 , COD Mn , SS, TN, TP, etc.).
  • BOD 5 inflow / outflow flow rate
  • inflow / outflow component concentration BOD 5 , COD Mn , SS, TN, TP, etc.
  • the inflow / outflow water quality data is usually measured once a day and recorded, or measured several times a day, and as an average value of the result, all of the inflow and outflow water of the sewage treatment plant existed as one value per day. Recorded data of water quality items and inflows.
  • the inflow / outflow water quality data may include weather factors such as temperature, water temperature, rainfall, sunshine, and humidity that may affect the treatment performance of the sewage treatment plant.
  • the process operation data includes at least one of aeration amount, sludge waste amount, sludge conveyance amount, chemical injection amount, sedimentation capacity and concentration of suspended solids in the reactor.
  • process operation data are typically modified once a day by the operator and applied to the process, and the results are typically the operating factors of the sewage treatment plant where the results are present as records, aeration volume, DO concentration and sludge in the reactor. Recorded results of operator factors including one or more of conveyed volume, sludge waste volume and sedimentation capacity (SVI, SV30). Therefore, the data caller 100 may select data necessary for diagnosing the actual energy consumption state from the various input data in advance and select daily data.
  • the data caller 100 may automatically call key data related to energy consumption among all data accumulated in a database operated separately in an individual sewage treatment plant.
  • the energy consumption state diagnosis unit 200 receives the data called from the data caller 100 and applies a multivariate statistical analysis method to the called data to determine the energy consumption state inherent in the called data. Extracting information serves to derive an energy consumption state diagnosis result from the extracted information on the energy consumption state.
  • the energy consumption state diagnosis unit 200 includes a data reduction unit 210, a data grouping unit 220, and a determination function extracting unit 230.
  • the data condensing unit 210 condenses the called data to extract information on the energy consumption state inherent in the called data for the called data.
  • the data grouping unit 220 groups the reduced data to derive a diagnosis result regarding a current energy consumption state using the reduced data.
  • the determination function deriving unit 230 derives a determination function to be used as a means for determining which group the new data belongs to with respect to the grouped data using Fisher's linear discrimination analysis. Therefore, the energy consumption state diagnosis unit 200 derives an energy consumption state diagnosis result of the called data by using the determination function derived by the determination function derivation unit 230.
  • the principal component analysis is applied to reduce the dimension of a total of 10 data sets to analyze four principal components, and the energy consumption state is performed by performing a K-means clustering analysis for data grouping.
  • the number of datasets that can be used may vary to reflect the characteristics of each sewage treatment plant, but it is desirable to use operating variables associated with major energy sources such as pumps and blowers. It is also desirable to reduce the size of these datasets to three to five dimensions.
  • the classified groups will also vary depending on the characteristics of the sewage treatment plant, but three to five groups are most preferred.
  • Fisher's linear discrimination analysis it is possible to apply other multivariate statistical analytical methods that can enable group identification.
  • the energy saving driving condition call unit 300 serves to call a driving condition that can be set in advance according to a group meaning a diagnosis result derived from the energy consumption state diagnosis unit 200.
  • the energy saving operation condition call unit 300 is DO (dissolved oxygen) for a group having a higher energy consumption state than other groups according to a group meaning a diagnosis result derived from the energy consumption state diagnosis unit 200.
  • DO dissolved oxygen
  • the amount of change in the operating condition is called by setting in advance an operating condition capable of energy saving, including the reduction of the sludge conveying flow rate and the excess sludge flow rate.
  • sludge conveying flow rate In case of medium energy consumption, sludge conveying flow rate, second hand waste sludge flow rate, surplus sludge flow rate and SRT can be called as operating variables.
  • DO dissolved oxygen
  • SRT excess sludge Flow rate
  • centrifugal tank draw flow In case of high energy consumption, sludge conveying flow rate, second hand waste sludge flow rate, surplus sludge flow rate and SRT can be called as operating variables.
  • DO dissolved oxygen
  • SRT excess sludge Flow rate
  • centrifugal tank draw flow centrifugal tank draw flow
  • anaerobic digester inflow can be called as operating variables.
  • the set value of the called operation variable for each state can be derived by comparing the inflow flow rate when the energy consumption state is low with the inflow flow rate for each state to determine the strength of the load.
  • the set values of the operating variables to be called for each state will be described in detail. It would be desirable to adjust the quantitative value to reflect
  • the energy consumption savings predicting unit 400 serves to predict energy savings by applying a polynomial regression analysis method according to the operating conditions called from the energy saving driving condition calling unit 300.
  • the polynomial regression analysis technique utilizes the same operating variables as the data called by the data caller 100 and is a model technique for predicting energy usage by dividing the operating variables into independent and dependent variables. Therefore, a model for predicting the amount of energy consumption can be used by inputting the set value of the called energy-saving operation condition. Among them, it is preferable to apply the polynomial regression analysis technique.
  • the independent variables in the polynomial regression analysis method may be influent flow rate, water temperature, SRT, DO concentration, sludge conveying flow rate, second hand waste sludge flow rate, surplus sludge flow rate, centrifugal concentrated sludge drawing flow rate and digester inflow sludge flow rate, etc.
  • the energy saving driving condition applying unit 500 serves to apply the driving conditions predicted by the energy consumption savings predicting unit 400 to an actuator including a pump and a blower in a sewage treatment plant. If the energy consumption state is medium or high, it is confirmed that the energy consumption can be reduced by applying the set value of the called operation variable to an actuator including a pump and a blower in the sewage treatment plant. Therefore, if the sewage treatment plant driver wants to operate the sewage treatment plant under the same operating conditions after confirming the quantitative savings value, the sewage treatment plant operator inputs the set value to each driver and checks the result of the day. It would be desirable to apply the reduced operating conditions called after the sewage plant plant decision rather than being automatically reflected in the plant operation.
  • FIG. 2 describes the energy saving prediction method according to the diagnosis of energy consumption state sewage treatment plant according to the present invention.
  • the first step is a data call step of calling data necessary for diagnosing an energy consumption state from a database storing inflow / outflow water quality data and process operation data of the sewage treatment plant (S110).
  • the second step is to receive the called data and to apply the multivariate statistical analysis method to the called data to extract information on the energy consumption state inherent in the called data to the information on the extracted energy consumption state
  • the energy consumption state diagnosing step (S120) may abbreviate the called data to extract information on the energy consumption state that is inherent in the called data with respect to the called data, and use the reduced data to present Using Fisher's linear discriminant analysis, a discriminant function is used to group the abbreviated data to derive a diagnosis result regarding energy consumption and to be used as a means for determining which group the new data belongs to for the grouped data. Derivation of the energy consumption state of the called data is derived using the derived discrimination function.
  • the third step is an energy saving operation condition call step of calling an operation condition that can be set in advance according to the group representing the derived diagnosis result (S130).
  • the fourth step is an energy consumption savings prediction step of predicting an energy saving amount by applying a polynomial regression analysis method according to the called operation condition (S140).
  • the fifth step is an energy saving operation condition application step of applying the predicted operation condition to an actuator including a pump and a blower in the sewage treatment plant (S150).
  • the inflow flow rate (Qin) is linked to the inflow flow pump operation, the water temperature (Temp) is the process operation variation according to the seasonal influence, the SRT is the waste sludge pump driving, the DO concentration is the blower driving, the sludge conveying flow rate (Qras) is the return sludge Pump driving, second hand waste sludge flow rate (Qfirst), second hand waste sludge pump drive, surplus sludge flow rate (Qsecond), waste sludge pump drive, centrifugal concentration sludge drawing (Qthickener) centrifugal condenser drive, digester inflow sludge flow rate (Qdin) In connection with the energy savings associated with methane generation through anaerobic digestion, these sewage treatment plants were selected and called.
  • Energy consumption status diagnosis was performed using multivariate statistical analytical techniques on the called data.
  • principal component analysis was performed, and as a result, four principal components (PC1, PC2, PC3, and PC4) were selected through principal component analysis. This means that the 10-dimensional data set is transformed into a new 4-dimensional data set.
  • main component values are finally calculated through each operation variable value and the principal component coefficient value.
  • PC1 0.848 * Qin + 0.512 * Temp-0.697 * SRT + 0.069 * DO + 0.424 * Qras-0.373 * Qfirst-0.084 * Qsecond-0.1 * Qthickener-0.058 * Qdin + 0.086 * Elec
  • PC2 0.06 * Qin + 0.019 * Temp + 0.169 * SRT + 0.075 * DO + 0.511 * Qras + 0.401 * Qfirst + 0.747 * Qsecond + 0.155 * Qthickener + 0.791 * Qdin-0.061 * Elec
  • PC3 -0.153 * Qin-0.747 * Temp + 0.249 * SRT + 0.892 * DO + 0.173 * Qras + 0.352 * Qfirst + 0.159 * Qsecond + 0.017 * Qthickener-0.212 * Qdin + 0.094 * Elec
  • PC4 -0.038 * Qin-0.2 * Temp + 0.314 * SRT-0.121 * DO-0.259 * Qras-0.439 * Qfirst + 0.019 * Qsecond + 0.893 * Qthickener + 0.282 * Qdin + 0.141 * Elec
  • K-average cluster analysis was used to derive a total of three groups that can classify the current energy consumption. The characteristics of each group derived are shown in [Table 1].
  • Table 1 Item Qin Elec Temp SRT DO Qras Qfirst Qsecond Qthickener Qdin Group1 Low High Low Medium Low Medium Medium Mean 286,775.4 63,859.5 19.0 6.0 1.75 60,112.2 1,195.4 2,217.9 282.9 926.3 Group2 Medium High Medium High Low High Mean 328,707.2 66281.8 22.6 5.2 1.76 73,051.0 1,308.1 2,989.1 274.7 1,190.8 Group3 High Medium Low High Medium Low Medium High Low Mean 336,875.9 71206.0 19.5 4.5 2.51 69,694.1 1,086.4 2,290.0 310.9 899.7
  • the value of the function of each group is calculated by substituting the principal component values first derived into the discrimination function, and the group having the largest value is interpreted as a group representing the current energy consumption state.
  • groups 2 and 3 which are medium and high energy consumption groups, the operating conditions for saving energy are called. Except for weather conditions such as water temperature and normal operating conditions among the variables that affect the energy consumption state, the 2 groups are called sludge conveying flow rate, second hand waste sludge flow rate, surplus sludge flow rate and SRT as the operating variables. DO concentration, SRT, excess sludge flow rate, centrifugal condensate draw flow and anaerobic digester inflow flow will be called as operating parameters.
  • the quantitative setpoint of each called operation variable was derived in conjunction with the influent load. For example, in the case of the DO concentration of the three groups, the inflow increase of the group 1 and the group 3 in Table 1 was about 17%, but the DO concentration was increased by 45%. This is a high increase even when considering the inflow load, so 20% was derived as an appropriate value between 17% and 45%. In this way, the energy-saving operating condition settings of Groups 2 and 3 were derived as follows.
  • the present invention has developed a prediction model using a polynomial regression analysis technique.
  • the independent variables were the influent flow rate, water temperature, SRT, DO concentration, sludge conveying flow rate, second hand waste sludge flow rate, surplus sludge flow rate, centrifugal concentrated sludge drawing amount, and digester inflow sludge.
  • the total flow rate is 9, and the dependent variable is the total amount of power in the treatment plant representing the sewage treatment plant energy.
  • N is the total number of variables
  • Xm, i is the measured value
  • Xs, i is the simulation result.
  • the RMSE between the amount of power used in the sewage treatment plant and the model simulation results was found to be about 1,603 kWh with a low error of less than 5%.
  • the developed predictive model was used to check the amount of energy that could be saved by entering the set values of the energy saving operating conditions called in the case of energy consumption of group 2 and group 3.
  • the value of the operating variable is not adjusted for the two groups and when the set value of the new operation variable is increased by 30% increase in sludge conveying flow rate, 20% increase in second hand waste sludge flow rate, 30% decrease in excess sludge flow rate, and 30% increase in SRT.
  • the comparison of is shown in FIG.
  • the new set of operating parameters confirmed that sewage treatment plant energy consumption can be reduced from a minimum of 932.5 kWh to a maximum of 1,059.9 kWh.
  • the sewage plant operator's setting value is reflected to the actual operation through the driver's decision after checking the sewage treatment plant driver's decision, it is possible to reduce the amount of energy consumed in the sewage treatment plant by adjusting the driving of the pump and the blower.
  • the present invention performs a qualitative diagnosis of energy consumption caused by the operation of equipment such as pumps and blowers, which are essential for the operation of sewage treatment plants, and provides a set value of operating conditions that can save energy according to the diagnosis result. It can be widely used in sewage treatment plants by proactively identifying the reduction in energy usage and providing practical operating setpoints to reduce energy use.

Abstract

The present invention relates to an energy-saving prediction system and method based on the diagnosis of the state of energy consumption in a sewage treatment plant. More particularly, the present invention relates to an energy-saving prediction system and method based on the diagnosis of the state of energy consumption in a sewage treatment plant which are capable of: performing qualitative state diagnosis on energy consumption caused by operating equipment, such as a pump and blower, essential for operating the sewage treatment plant; providing a set value for an operating condition which allows energy saving according to the diagnosis result; confirming in advance a consequent decrease in the amount of energy used; and thereby providing a practical operational set value for energy use reduction. Provided according to the present invention is an energy-saving prediction system based on the diagnosis of the state of energy consumption in a sewage treatment plant, the system comprising: a data calling unit for calling data required for the diagnosis of the state of energy consumption from a database for storing influent/effluent water quality data and process operating data on the sewage treatment plant; an energy consumption state diagnosis unit for receiving the data called by the data calling unit, applying a multivariate statistical analysis technique to the called data, extracting information on the state of energy consumption inherent in the called data, and deriving a diagnosis result on the state of energy consumption from the extracted information on the state of energy consumption; an energy-saving operating condition calling unit for calling a preset operating condition which allows energy saving according to the group which indicates the diagnosis result derived by the energy consumption state diagnosis unit; an energy consumption reduction prediction unit for applying a polynomial regression analysis technique according to the operating condition called by the energy-saving operating condition calling unit and for predicting the amount of energy saved; and an energy-saving operating condition application unit for applying the operating condition predicted by the energy consumption reduction prediction unit to the actuator, including a pump and blower, of the sewage treatment plant.

Description

하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측시스템 및 방법Energy Saving Prediction System and Method by Diagnosis of Energy Consumption in Sewage Treatment Plant
본 발명은 하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측시스템 및 방법에 관한 것이다. 보다 상세하게 설명하면, 하수처리장 운전에 있어 필수적인 펌프와 송풍기와 같은 기기의 구동으로 인해 발생되는 에너지 소비에 대한 정성적인 상태 진단을 수행하며, 진단 결과에 따라 에너지 절감이 가능한 운전 조건의 설정치를 제공하여 이에 따른 에너지 사용량의 감소를 사전에 확인하여 에너지 사용 절감을 위한 실질적인 운전 설정치를 제공할 수 있는 하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측시스템 및 방법에 관한 것이다.The present invention relates to an energy saving prediction system and method according to the diagnosis of energy consumption in sewage treatment plants. In more detail, it performs qualitative diagnosis of energy consumption caused by the operation of equipment such as pumps and blowers, which are essential for the operation of sewage treatment plants, and provides a set of operating conditions that can save energy according to the diagnosis results. Therefore, the present invention relates to an energy saving prediction system and method according to the diagnosis of energy consumption status of a sewage treatment plant, which can provide a practical operation setpoint for reducing energy use by confirming the reduction of energy consumption accordingly.
하수처리장은 일련의 수처리 설비에서 마지막 처리를 담당하고 있기에, 인간의 활동 및 산업 시스템의 구동 후 발생되는 하수가 주요한 유입 인자이므로, 일반적으로 주민 혐오시설 및 에너지 소비시설로 인식되어 왔다. 이와 같은 인식의 배경에는 완벽히 처리되지 않은 하수가 방류수계로 배출될 경우, 수계 수질 오염을 유발하고, 종래에는 인간이 사용할 수 있는 수자원의 감소를 유발하기에, 처리에 막대한 비용 및 에너지가 소비되더라도 안정적인 처리 성능 유지가 하수처리장의 가장 핵심적인 운전 목표로 설정되어 왔기 때문이다. Since the sewage treatment plant is responsible for the final treatment in a series of water treatment facilities, it has been generally recognized as a resident hate and energy consumption facility because sewage generated after human activities and industrial systems are driven is a major inflow factor. Against this background, when untreated sewage is discharged to the discharge system, it causes water pollution and, in the past, causes a reduction in the human resources available to humans. Maintaining treatment performance has been set as the primary operational goal for sewage treatment plants.
하지만, 최근 급격한 인구 증가에 따른 에너지 소비량의 증가로 인해, 기존 자원을 절약해야 할 뿐만 아니라 현재 과다하게 소비되고 있는 에너지 사용 시설에 대한 정부차원에서의 규제가 강화되고 있는 시점에서, 대표적인 에너지 소비 시설로 인식되어 왔던 하수처리장에 대해서도 다양한 에너지 절감 방법론이 제기되고 있다. However, due to the recent increase in energy consumption due to rapid population growth, not only should we save existing resources, but at the time government regulations on energy use facilities that are excessively consumed are being tightened. Various sewage methodologies have been proposed for sewage treatment plants that have been recognized.
대표적으로, 하수처리장이 포괄하고 있는 부지를 전체적으로 활용할 수 있는 다양한 신재생에너지 기술들이 있다. 태양광, 소수력, 바이오가스 등 신재생에너지 기술들은 하수처리장의 여유 부지 또는 처리 흐름의 약간의 수정으로도 충분히 에너지 생산이 가능한 기술들이다. 이와 같은 기술들과 함께, 최근에는 고효율의 장치를 도입함으로써, 동일한 전력으로도 더 높은 성능을 확보하여 에너지 사용량을 저감할 수 있는 방안도 있다. 비록 시설 설치 및 교체를 위한 초기 투자비가 발생하는 단점이 있지만, 장기적인 측면에서 이와 같은 신재생에너지 기술 및 고효율 장치 도입은 에너지 절감을 위한 효율적인 방법들 중 하나라고 판단할 수 있다. Representatively, there are various renewable energy technologies that can make full use of the sites covered by sewage treatment plants. Renewable energy technologies such as solar, hydropower and biogas are those that can produce enough energy even with a small amount of land in a sewage treatment plant or with a slight modification of the treatment stream. In addition to these technologies, recently, by introducing a high-efficiency device, there is also a way to reduce the energy consumption by ensuring higher performance even at the same power. Although the initial investment costs for installation and replacement of facilities are incurred, in the long term, the introduction of renewable energy technology and high efficiency devices can be considered as one of the efficient methods for energy saving.
상기 언급된 방법들은 하수처리장의 운영과는 전혀 연계되지 않은 새로운 기술들의 도입이다. 따라서, 에너지 절감을 위해서 처리장 운영 방안에 대한 새로운 대안을 제시하지 않는다면, 최적의 에너지 소비 상태로써 하수처리장이 운영되고 있다고 판단할 수 없다. 이는 하수처리장 운영에 있어 과다하게 소요되고 있는 에너지가 분명히 존재하며, 이와 같은 소비 에너지를 절감하는 것은 언급된 방법들과 같은 추가적인 초기 투자비가 필요하지 않기에, 경제적으로도 큰 기대효과를 얻을 수 있는 방안으로 판단된다.  The above mentioned methods are the introduction of new technologies that are not at all linked to the operation of sewage treatment plants. Therefore, unless a new alternative to the plant operation plan is proposed to save energy, it cannot be determined that the sewage treatment plant is operated at the optimum energy consumption. It is obvious that there is an excessive amount of energy consumed in the operation of the sewage treatment plant, and this saving of energy consumption does not require additional initial investment such as the mentioned methods, which can bring about economic expected effects. Judging by the way.
대부분 하수처리장의 운전자는 자신의 경험과 지식에 기반하여 하수처리장의 운영 상태를 진단하고, 안정적인 하수 처리를 최우선으로 고려하기 때문에, 에너지 소비 측면을 고려하지 않고 운전 조건을 과다하게 설정하여 운영하는 경우가 빈번하다. 일례로, 하수처리장의 단위 장치 중 가장 큰 부분을 차지하고 있는 송풍기의 경우, 부하 변동에 따라 적절한 DO(용존산소) 설정값을 제시하고, P, PI 및 PID(Proportional Integral Differential) 제어기를 사용하는 일련의 제어를 적용하여 효율적으로 송풍기를 구동하는 사례는 극히 드물다. 대부분의 하수처리장은 고정된 송풍량을 주입하고 있으며, 이로 인해 불필요한 소요 동력이 발생하여 에너지 사용량이 증가되는 경우가 빈번하다. Most drivers of sewage treatment plants diagnose their operating conditions based on their own experience and knowledge, and consider stable sewage treatment as the top priority. Therefore, if the operating conditions are set excessively without considering energy consumption aspects, Is frequent. For example, a blower that occupies the largest portion of a unit in a sewage treatment plant, a series of devices using P, PI, and Proportional Integral Differential (PID) controllers to present the appropriate DO (Dissolved Oxygen) settings according to load fluctuations. It is extremely rare to drive a blower efficiently by applying the control of. Most sewage treatment plants inject a fixed amount of blown air, which causes unnecessary power generation and often increases energy consumption.
따라서, 현재 하수처리장의 에너지 소비 상태에 대한 정확한 진단을 수행한 이후 도출된 진단 결과에 따라 하수처리장 유출수질에 영향을 주지 않는 범위 내에서 운전 변수의 설정치를 적절하게 조정함으로써 하수처리장 에너지 소비량을 절감할 수 있는 방안이 필요한 상황이다.Therefore, according to the diagnosis result, after the accurate diagnosis of the current consumption status of the sewage treatment plant, the energy consumption of the sewage treatment plant is reduced by appropriately adjusting the set values of operating variables within the range that does not affect the effluent quality. There is a need for a solution.
본 발명은 이와 같은 문제점을 해결하기 위해 안출된 것으로서, 본 발명은 하수처리장에서 소요되는 에너지의 소비 상태를 진단하고 이에 대한 에너지 절감 운전 조건을 도출하여 정량적인 에너지 절감 수치를 예측하기 위한 시스템 및 방법에 관한 것으로, 하수처리장에서 사용되는 에너지와 관련된 주요 기기들과 연관된 운전 변수 및 기상 조건의 자료를 대상으로 다변량 통계분석들을 수행하여 에너지 소비 상태를 진단하고, 이 결과에 따라 에너지 소비 상태가 높은 경우에 대하여 에너지 절감이 가능한 운전 조건을 호출하여 이를 에너지 소비 예측이 가능한 회귀분석 모델에 적용하여 그 결과를 확인함으로써, 처리장의 에너지 소비를 절감할 수 있는 운전 조건을 제공할 수 있는 하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측시스템 및 방법을 제공하는데 그 목적이 있다. The present invention has been made to solve such a problem, the present invention is a system and method for predicting the quantitative energy saving value by diagnosing the state of consumption of energy in the sewage treatment plant and deriving the energy saving operating conditions for it The present invention relates to a method for analyzing energy consumption by performing multivariate statistical analysis on data on operating variables and weather conditions associated with major devices related to energy used in sewage treatment plants. The energy consumption status of sewage treatment plant that can provide the operating conditions to reduce the energy consumption of the plant by calling the operating conditions that can save energy and applying them to the regression analysis model that can predict the energy consumption. Energy saving prediction system by diagnosis To provide a method has its purpose.
본 발명에 의하면, 하수처리장의 유입/유출 수질 데이터 및 공정운영 데이터를 저장하는 데이터베이스로부터 에너지소비상태 진단에 필요한 데이터를 호출하는 데이터호출부; 상기 데이터호출부로부터 호출된 데이터를 전달받아 상기 호출된 데이터에 대해 다변량 통계 분석기법을 적용하여 상기 호출된 데이터에 내재되어 있는 에너지소비상태에 관한 정보를 추출하여 상기 추출된 에너지소비상태에 관한 정보로부터 에너지소비상태 진단결과를 도출하는 에너지소비상태 진단부; 상기 에너지소비상태 진단부로부터 도출된 진단결과를 의미하는 그룹에 따라 사전에 설정된 에너지절감이 가능한 운전조건을 호출하는 에너지절감 운전조건 호출부; 상기 에너지절감 운전조건 호출부로부터 호출된 운전조건에 따라 다항회귀분석기법을 적용하여 에너지 절감량을 예측하는 에너지소비 절감량 예측부; 및 상기 에너지소비 절감량 예측부에 의해 예측된 운전조건을 하수처리장의 펌프 및 송풍기를 포함하는 구동기(actuator)에 적용하는 에너지절감 운전조건 적용부;를 포함하는 하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측시스템를 제공한다. According to the present invention, the data call unit for calling data necessary for the diagnosis of energy consumption from the database for storing the inflow / outflow water quality data and process operation data of the sewage treatment plant; Receives the data called from the data caller and applies the multivariate statistical analysis method to the called data to extract information on the energy consumption state inherent in the called data and to extract the information on the energy consumption state. An energy consumption state diagnosis unit for deriving an energy consumption state diagnosis result from the; An energy saving operation condition calling unit for calling a preset energy saving operation condition according to a group representing a diagnosis result derived from the energy consumption state diagnosis unit; An energy consumption reduction prediction unit for predicting an energy saving amount by applying a polynomial regression analysis method according to the operation condition called from the energy saving operation condition calling unit; And energy saving operation condition applying unit for applying an operating condition predicted by the energy consumption reduction amount estimator to an actuator including a pump and a blower in the sewage treatment plant; Provide a prediction system.
한편, 상기 데이터호출부는 개별 하수처리장에서 별도로 운영되고 있는 데이터베이스에 축적되는 과거의 모든 데이터 중 에너지 소비와 연관성이 있는 주요 데이터들을 자동으로 호출하는 것을 특징으로 한다.On the other hand, the data call unit is characterized in that the main data automatically associated with the energy consumption of all the data accumulated in the database that is operated separately in the individual sewage treatment plant.
한편, 상기 에너지소비상태 진단부는 상기 호출된 데이터에 대해 상기 호출된 데이터에 내재되어 있는 에너지소비상태에 관한 정보를 추출하기 위해 상기 호출된 데이터를 축약하는 데이터축약부; 상기 축약된 데이터를 이용하여 현재 에너지소비상태에 관한 진단결과를 도출하기 위해 상기 축약된 데이터를 그룹화하는 데이터그루핑부; 및 상기 그룹화된 데이터에 대하여 새로운 데이터가 어떠한 그룹에 속하는지를 판별하기 위한 수단으로 사용될 판별함수를 Fisher의 선형판별분석을 이용하여 도출하는 판별함수도출부;를 포함하되, 상기 에너지소비상태 진단부는 상기 판별함수도출부에 의해 도출된 판별함수를 이용하여 상기 호출된 데이터의 에너지소비상태 진단결과를 도출하는 것을 특징으로 한다.On the other hand, the energy consumption state diagnosis unit for the called data for extracting the information on the energy consumption state inherent in the called data for reducing the called data; A data grouping unit for grouping the reduced data to derive a diagnosis result regarding a current energy consumption state using the reduced data; And a discrimination function derivation unit for deriving a discrimination function to be used as means for determining which group the new data belongs to with respect to the grouped data by using Fisher's linear discrimination analysis. The energy consumption state diagnosis result of the called data is derived using the discrimination function derived by the discrimination function extracting unit.
한편, 상기 에너지절감 운전조건 호출부는 상기 에너지소비상태 진단부로부터 도출된 진단결과를 의미하는 그룹에 따라, 에너지소비상태가 다른 그룹에 비해 높은 그룹에 대하여 DO(용존산소), 슬러지반송유량 및 잉여슬러지유량의 절감을 포함한 에너지절감이 가능한 운전조건을 사전에 설정하여 상기 운전조건의 변화량을 호출하는 것을 특징으로 한다.On the other hand, the energy saving operation condition call unit DO (dissolved oxygen), sludge conveyed flow rate and surplus for the group with a higher energy consumption state than the other groups according to the group means the diagnosis result derived from the energy consumption state diagnosis unit It is characterized in that a change in the operating conditions is called by setting in advance an operating condition capable of saving energy, including reducing the sludge flow rate.
한편, 상기 다항회귀분석기법은 상기 데이터호출부에 의해 호출된 데이터와 동일한 운전변수를 활용하며, 상기 운전변수에 대해 독립변수와 종속변수로 나누어 에너지 사용량 예측을 하는 모델기법인 것을 특징으로 한다. On the other hand, the polynomial regression analysis technique is a model technique that utilizes the same operation variable and the data called by the data call unit, and predicts the energy consumption by dividing the operation variable into independent and dependent variables.
또한 본 발명에 의하면, 하수처리장의 유입/유출 수질 데이터 및 공정운영 데이터를 저장하는 데이터베이스로부터 에너지소비상태 진단에 필요한 데이터를 호출하는 데이터호출단계; 상기 호출된 데이터를 전달받아 상기 호출된 데이터에 대해 다변량 통계 분석기법을 적용하여 상기 호출된 데이터에 내재되어 있는 에너지소비상태에 관한 정보를 추출하여 상기 추출된 에너지소비상태에 관한 정보로부터 에너지소비상태 진단결과를 도출하는 에너지소비상태 진단단계; 상기 도출된 진단결과를 의미하는 그룹에 따라 사전에 설정된 에너지절감이 가능한 운전조건을 호출하는 에너지절감 운전조건 호출단계; 상기 호출된 운전조건에 따라 다항회귀분석기법을 적용하여 에너지 절감량을 예측하는 에너지 소비 절감량 예측단계; 및 상기 예측된 운전조건을 하수처리장의 펌프 및 송풍기를 포함하는 구동기(actuator)에 적용하는 에너지절감 운전조건 적용단계;를 포함하는 하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측방법를 제공한다. In addition, according to the present invention, a data call step of calling data necessary for the diagnosis of energy consumption from the database for storing the inflow / outflow water quality data and process operation data of the sewage treatment plant; Receives the called data and applies the multivariate statistical analysis method to the called data to extract information on the energy consumption state inherent in the called data and to extract the energy consumption state from the information on the extracted energy consumption state. An energy consumption state diagnosis step of deriving a diagnosis result; An energy saving operation condition calling step of calling an operation condition capable of saving energy in advance according to the group representing the derived diagnosis result; An energy consumption savings predicting step of predicting an energy saving amount by applying a polynomial regression analysis method according to the called driving condition; And an energy saving operation condition applying the predicted operating condition to an actuator including a pump and a blower in the sewage treatment plant.
한편, 상기 에너지소비상태 진단단계는 상기 호출된 데이터에 대해 상기 호출된 데이터에 내재되어 있는 에너지소비상태에 관한 정보를 추출하기 위해 상기 호출된 데이터를 축약하고, 상기 축약된 데이터를 이용하여 현재 에너지소비상태에 관한 진단결과를 도출하기 위해 상기 축약된 데이터를 그룹화하고, 상기 그룹화된 데이터에 대하여 새로운 데이터가 어떠한 그룹에 속하는지를 판별하기 위한 수단으로 사용될 판별함수를 Fisher의 선형판별분석을 이용하여 도출하고, 상기 도출된 판별함수를 이용하여 상기 호출된 데이터의 에너지소비상태 진단결과를 도출하는 것을 특징으로 한다.On the other hand, the step of diagnosing the energy consumption state to reduce the called data to extract information on the energy consumption state inherent in the called data for the called data, and the current energy using the reduced data The abbreviated function is grouped using Fisher's linear discriminant analysis to be used as a means to determine which group the new data belongs to. And deriving an energy consumption state diagnosis result of the called data using the derived discrimination function.
기존의 하수처리장에서 소요되는 에너지 사용에 대한 진단은 별도의 분석 없이 계절별로 높고 낮음만을 판단하여, 실제 절감 가능한 운전 조건에 대한 명확한 분석이 존재하지 않았지만, 본 발명에서는 다변량 통계 분석 기법들을 적용하여 운전 설정치들이 가지고 있는 정보에 기반하여 에너지 사용 상태가 높음, 중간 및 낮음으로 제시할 수 있으며, 높은 에너지 사용 상태에 대해서는 어떠한 운전 변수의 설정치가 문제가 될 수 있는지에 대한 원인을 제시가능하여, 공정 운전자가 쉽게 현재 사용되고 있는 에너지 상태에 대한 정보를 얻을 수 있는 효과가 있다.  The diagnosis of energy use in the existing sewage treatment plant is judged only high and low by season without any separate analysis, and there is no clear analysis on actual saving conditions. However, in the present invention, multivariate statistical analysis techniques are applied. Based on the information the setpoints have, energy use states can be presented as high, medium and low.For high energy use states, it is possible to suggest the cause of which operating variable setpoints may be a problem. Has the effect of easily obtaining information about the current energy state.
또한 본 발명은 정성적인 에너지 사용 상태에 대한 정보 제공에 추가하여, 에너지 절감 예측 모델을 제안함으로써, 조절 가능한 운전 변수의 설정치를 입력 변수로 활용하여 현재 사용되고 있는 에너지량보다 정량적으로 절감 가능한 수치를 제공할 수 있게 되어, 실제 공정에 운전 변수의 설정치를 반영함으로써, 유출수질이 안정적으로 유지됨과 동시에 에너지 사용 측면에서도 효율적인 공정 운영이 가능한 효과가 있다.  In addition, the present invention proposes an energy saving prediction model in addition to providing information on qualitative energy use state, thereby providing a numerical value that can be quantitatively reduced from the amount of energy currently used by using an adjustable operating variable setpoint as an input variable. By reflecting the set values of the operating variables in the actual process, it is possible to maintain the effluent quality stably and at the same time efficiently operate the process in terms of energy use.
또한 본 발명에 의한 에너지 절감 예측 모델은 하수처리장에서 사용되고 있는 에너지량과 연계된 공정 운전 변수로 모델식이 구성되어 있기 때문에, 공정 운전자가 임의적으로 운전 변수의 수치를 변경하여 그 효과를 모의할 수 있고, 이를 통해 에너지 절감이 가능한 공정 운전 방안들에 대해 공정 운전자에게 교육적인 측면을 제공할 수 있고, 공정 거동을 사전에 평가할 수 있어 보다 효과적인 에너지 절감 운영이 가능한 효과가 있다.In addition, the energy saving prediction model according to the present invention has a model formula consisting of process operation variables associated with the amount of energy used in the sewage treatment plant, so that the process operator can arbitrarily change the numerical value of the operation variable to simulate the effect. In this way, it is possible to provide an educational aspect to process operators about process operation methods that can save energy, and to evaluate process behavior in advance, thereby enabling more efficient energy saving operation.
도 1은 본 발명의 실시예에 따른 하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측시스템을 나타내는 구성도이다. Figure 1 is a block diagram showing an energy saving prediction system according to the diagnosis of energy consumption state sewage treatment plant according to an embodiment of the present invention.
도 2는 본 발명의 실시예에 따른 하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측방법을 나타내는 순서도이다. 2 is a flow chart showing a method for predicting energy saving according to the diagnosis of energy consumption in a sewage treatment plant according to an embodiment of the present invention.
도 3은 본 발명의 실시예에 따른 하수처리장 에너지 절감 예측 모델의 결과를 나타낸 그래프이다.3 is a graph showing the results of the sewage treatment plant energy saving prediction model according to an embodiment of the present invention.
도 4는 도 3의 하수처리장 에너지 절감 예측 모델의 결과를 검증하기 위한 그래프이다. 4 is a graph for verifying the results of the sewage treatment plant energy saving prediction model of FIG. 3.
도 5는 본 발명의 실시예에 따른 하수처리장 에너지소비상태가 중간인 경우 운전조건의 변경 전후의 에너지 소비량의 예측결과를 나타낸 그래프이다. 5 is a graph showing a result of predicting the energy consumption before and after the change in the operating conditions when the energy consumption state of the sewage treatment plant according to the embodiment of the present invention is medium.
도 6은 본 발명의 실시예에 따른 하수처리장 에너지소비상태가 높은 경우 운전조건의 변경 전후의 에너지 소비량의 예측결과를 나타낸 그래프이다. 6 is a graph showing a result of predicting the energy consumption before and after the change in the operating conditions when the energy consumption state of the sewage treatment plant according to an embodiment of the present invention is high.
본 발명에 의하면, 하수처리장의 유입/유출 수질 데이터 및 공정운영 데이터를 저장하는 데이터베이스로부터 에너지소비상태 진단에 필요한 데이터를 호출하는 데이터호출부; 상기 데이터호출부로부터 호출된 데이터를 전달받아 상기 호출된 데이터에 대해 다변량 통계 분석기법을 적용하여 상기 호출된 데이터에 내재되어 있는 에너지소비상태에 관한 정보를 추출하여 상기 추출된 에너지소비상태에 관한 정보로부터 에너지소비상태 진단결과를 도출하는 에너지소비상태 진단부; 상기 에너지소비상태 진단부로부터 도출된 진단결과를 의미하는 그룹에 따라 사전에 설정된 에너지절감이 가능한 운전조건을 호출하는 에너지절감 운전조건 호출부; 상기 에너지절감 운전조건 호출부로부터 호출된 운전조건에 따라 다항회귀분석기법을 적용하여 에너지 절감량을 예측하는 에너지소비 절감량 예측부; 및 상기 에너지소비 절감량 예측부에 의해 예측된 운전조건을 하수처리장의 펌프 및 송풍기를 포함하는 구동기(actuator)에 적용하는 에너지절감 운전조건 적용부;를 포함하는 하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측시스템를 제공한다.According to the present invention, the data call unit for calling data necessary for the diagnosis of energy consumption from the database for storing the inflow / outflow water quality data and process operation data of the sewage treatment plant; Receives the data called from the data caller and applies the multivariate statistical analysis method to the called data to extract information on the energy consumption state inherent in the called data and to extract the information on the energy consumption state. An energy consumption state diagnosis unit for deriving an energy consumption state diagnosis result from the; An energy saving operation condition calling unit for calling a preset energy saving operation condition according to a group representing a diagnosis result derived from the energy consumption state diagnosis unit; An energy consumption reduction prediction unit for predicting an energy saving amount by applying a polynomial regression analysis method according to the operation condition called from the energy saving operation condition calling unit; And energy saving operation condition applying unit for applying an operating condition predicted by the energy consumption reduction amount estimator to an actuator including a pump and a blower in the sewage treatment plant; Provide a prediction system.
이하, 본 발명의 바람직한 실시예를 첨부된 도면들을 참조하여 상세히 설명한다. 우선 각 도면의 구성요소들에 참조번호를 부가함에 있어서, 동일한 구성요소들에 대해서는 비록 다른 도면상에 표시되더라도 가능한 한 동일한 부호를 가지도록 하고 있음에 유의해야 한다. 또한 본 발명을 설명함에 있어, 관련된 공지 구성 또는 기능에 대한 구체적인 설명이 본 발명의 요지를 흐릴 수 있다고 판단되는 경우에는 그 상세한 설명은 생략한다. Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. First, in adding reference numerals to the components of each drawing, it should be noted that the same reference numerals are used as much as possible even if displayed on different drawings. In describing the present invention, when it is determined that the detailed description of the related well-known configuration or function may obscure the gist of the present invention, the detailed description thereof will be omitted.
도 1은 본 발명의 실시예에 따른 하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측시스템을 나타내는 구성도이고, 도 2는 본 발명의 실시예에 따른 하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측방법을 나타내는 순서도이고, 도 3은 본 발명의 실시예에 따른 하수처리장 에너지 절감 예측 모델의 결과를 나타낸 그래프이고, 도 4는 도 3의 하수처리장 에너지 절감 예측 모델의 결과를 검증하기 위한 그래프이고, 도 5는 본 발명의 실시예에 따른 하수처리장 에너지소비상태가 중간인 경우 운전조건의 변경 전후의 에너지 소비량의 예측결과를 나타낸 그래프이고, 도 6은 본 발명의 실시예에 따른 하수처리장 에너지소비상태가 높은 경우 운전조건의 변경 전후의 에너지 소비량의 예측결과를 나타낸 그래프이다. 1 is a block diagram showing an energy saving prediction system according to the sewage treatment plant energy consumption diagnosis according to an embodiment of the present invention, Figure 2 is an energy saving prediction method according to the diagnosis of energy consumption state sewage treatment plant according to an embodiment of the present invention. 3 is a graph showing the result of the sewage treatment plant energy saving prediction model according to an embodiment of the present invention, FIG. 4 is a graph for verifying the result of the sewage treatment plant energy saving prediction model of FIG. 3, and FIG. 5 is a graph showing the prediction result of the energy consumption before and after the change in the operating conditions when the energy consumption state of the sewage treatment plant according to the embodiment of the present invention is medium, Figure 6 is the energy consumption state of the sewage treatment plant according to the embodiment of the present invention In the case of high, it is a graph showing the prediction result of energy consumption before and after the change of operating conditions.
도 1을 참조하면, 본 발명에 의한 하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측시스템(10)은 데이터호출부(100), 에너지소비상태 진단부(200), 에너지절감 운전조건 호출부(300), 에너지소비 절감량 예측부(400) 및 에너지절감 운전조건 적용부(500)를 포함한다. Referring to FIG. 1, the energy saving prediction system 10 according to the sewage treatment plant energy consumption diagnosis according to the present invention includes a data call unit 100, an energy consumption state diagnosis unit 200, and an energy saving operation condition call unit 300. ), The energy consumption reduction amount prediction unit 400 and the energy saving operation condition applying unit 500.
상기 데이터호출부(100)는 하수처리장의 유입/유출 수질 데이터 및 공정운영 데이터를 저장하는 데이터베이스로부터 에너지소비상태 진단에 필요한 데이터를 호출하는 역할을 한다. 상기 유입/유출 수질 데이터는 유입/유출유량과 유입/유출성분농도(BOD5, CODMn, SS, TN, TP 등)를 포함한다. 좀 더 구체적으로 살펴보면 상기 유입/유출 수질 데이터는 통상적으로 1일 1회 측정되어 기록되거나, 1일 수회 측정되어 그 결과의 평균값으로서 1일에 1개의 값으로 존재하는 하수처리장의 유입수와 유출수의 모든 수질항목들과 유입유량의 기록된 데이터를 말한다. 상기 유입/유출 수질 데이터는 하수처리장의 처리성능에 영향을 줄 수 있는 기온, 수온, 강우, 일조량, 습도 등의 기상인자를 포함할 수 있다. 상기 공정운영 데이터는 하수처리장의 폭기량, 슬러지폐기량, 슬러지반송량, 약품주입량, 침전능 및 반응조 내 부유물질 농도 중 적어도 하나 이상을 포함한다. 좀 더 구체적으로 살펴보면, 공정운영 데이터는 통상적으로 1일 1회 운전자에 의해 수정되어 공정에 적용되며, 그 결과가 기록으로서 존재하는 하수처리장의 운전인자로서 통상적으로 폭기량, 반응조의 DO농도, 슬러지반송량, 슬러지폐기량, 침전능(SVI, SV30) 중 하나 이상을 포함하는 운전인자의 기록된 결과물을 말한다. 따라서 상기 데이터호출부(100)는 이와 같은 다양한 입력 데이터들로부터 실제 에너지소비상태 진단에 필요한 데이터들을 사전에 선정하여 일별 데이터들을 선정할 수 있는 것이다. The data call unit 100 serves to call data necessary for diagnosing an energy consumption state from a database storing inflow / outflow water quality data and process operation data of the sewage treatment plant. The inflow / outflow water quality data includes inflow / outflow flow rate and inflow / outflow component concentration (BOD 5 , COD Mn , SS, TN, TP, etc.). In more detail, the inflow / outflow water quality data is usually measured once a day and recorded, or measured several times a day, and as an average value of the result, all of the inflow and outflow water of the sewage treatment plant existed as one value per day. Recorded data of water quality items and inflows. The inflow / outflow water quality data may include weather factors such as temperature, water temperature, rainfall, sunshine, and humidity that may affect the treatment performance of the sewage treatment plant. The process operation data includes at least one of aeration amount, sludge waste amount, sludge conveyance amount, chemical injection amount, sedimentation capacity and concentration of suspended solids in the reactor. In more detail, process operation data are typically modified once a day by the operator and applied to the process, and the results are typically the operating factors of the sewage treatment plant where the results are present as records, aeration volume, DO concentration and sludge in the reactor. Recorded results of operator factors including one or more of conveyed volume, sludge waste volume and sedimentation capacity (SVI, SV30). Therefore, the data caller 100 may select data necessary for diagnosing the actual energy consumption state from the various input data in advance and select daily data.
또한 상기 데이터호출부(100)는 개별 하수처리장에서 별도로 운영되고 있는 데이터베이스에 축적되는 과거의 모든 데이터 중 에너지 소비와 연관성이 있는 주요 데이터들을 자동으로 호출할 수도 있다. In addition, the data caller 100 may automatically call key data related to energy consumption among all data accumulated in a database operated separately in an individual sewage treatment plant.
상기 에너지소비상태 진단부(200)는 상기 데이터호출부(100)로부터 호출된 데이터를 전달받아 상기 호출된 데이터에 대해 다변량 통계 분석기법을 적용하여 상기 호출된 데이터에 내재되어 있는 에너지소비상태에 관한 정보를 추출하여 상기 추출된 에너지소비상태에 관한 정보로부터 에너지소비상태 진단결과를 도출하는 역할을 한다. The energy consumption state diagnosis unit 200 receives the data called from the data caller 100 and applies a multivariate statistical analysis method to the called data to determine the energy consumption state inherent in the called data. Extracting information serves to derive an energy consumption state diagnosis result from the extracted information on the energy consumption state.
상기 에너지소비상태 진단부(200)는 데이터축약부(210), 데이터그루핑부(220) 및 판별함수도출부(230)를 포함한다. 상기 데이터축약부(210)는 상기 호출된 데이터에 대해 상기 호출된 데이터에 내재되어 있는 에너지소비상태에 관한 정보를 추출하기 위해 상기 호출된 데이터를 축약한다. 상기 데이터그루핑부(220)는상기 축약된 데이터를 이용하여 현재 에너지소비상태에 관한 진단결과를 도출하기 위해 상기 축약된 데이터를 그룹화한다. 상기 판별함수도출부(230)는 상기 그룹화된 데이터에 대하여 새로운 데이터가 어떠한 그룹에 속하는지를 판별하기 위한 수단으로 사용될 판별함수를 Fisher의 선형판별분석을 이용하여 도출한다. 따라서 상기 에너지소비상태 진단부(200)는 상기 판별함수도출부(230)에 의해 도출된 판별함수를 이용하여 상기 호출된 데이터의 에너지소비상태 진단결과를 도출하게 된다.The energy consumption state diagnosis unit 200 includes a data reduction unit 210, a data grouping unit 220, and a determination function extracting unit 230. The data condensing unit 210 condenses the called data to extract information on the energy consumption state inherent in the called data for the called data. The data grouping unit 220 groups the reduced data to derive a diagnosis result regarding a current energy consumption state using the reduced data. The determination function deriving unit 230 derives a determination function to be used as a means for determining which group the new data belongs to with respect to the grouped data using Fisher's linear discrimination analysis. Therefore, the energy consumption state diagnosis unit 200 derives an energy consumption state diagnosis result of the called data by using the determination function derived by the determination function derivation unit 230.
본 발명에서는 데이터축약을 위해 주성분분석을 적용하여 총 10개의 데이터셋의 차원을 축소하여 4개의 주성분을 분석할 것이며, 분석된 주성분에 대해 데이터그루핑을 위해 K-평균 군집분석을 수행하여 에너지소비상태를 3개의 그룹으로 분류할 것이다. 사용될 수 있는 데이터셋의 갯수는 각 하수처리장의 특성을 반영하여 달라질 수 있지만, 펌프 및 송풍기 등과 같은 주요한 에너지원과 연계된 운전변수들을 사용하는 것이 바람직하다. 또한 이들 데이터셋의 차원을 축소할 경우 3 ~ 5개로 차원축소하는 것이 바람직하다. 또한 분류된 그룹은 해당되는 하수처리장의 특성에 따라 달라지게 되지만, 3 ~ 5개의 그룹이 가장 바람직하다, 새로운 데이터셋에 대한 사전 분류된 그룹의 소속의 확인하기 위해 본 발명에서는 피셔의 선형판별분석을 수행하였으며, 그룹의 소속확인을 가능하게 할 수 있는 다른 다변량 통계 분석기법의 적용도 얼마든지 가능할 것이다. In the present invention, the principal component analysis is applied to reduce the dimension of a total of 10 data sets to analyze four principal components, and the energy consumption state is performed by performing a K-means clustering analysis for data grouping. Will be divided into three groups. The number of datasets that can be used may vary to reflect the characteristics of each sewage treatment plant, but it is desirable to use operating variables associated with major energy sources such as pumps and blowers. It is also desirable to reduce the size of these datasets to three to five dimensions. The classified groups will also vary depending on the characteristics of the sewage treatment plant, but three to five groups are most preferred. In the present invention, to determine the affiliation of pre-sorted groups for a new dataset, Fisher's linear discrimination analysis It is possible to apply other multivariate statistical analytical methods that can enable group identification.
상기 에너지절감 운전조건 호출부(300)는 상기 에너지소비상태 진단부(200)로부터 도출된 진단결과를 의미하는 그룹에 따라 사전에 설정된 에너지절감이 가능한 운전조건을 호출하는 역할을 한다. 상기 에너지절감 운전조건 호출부(300)는 상기 에너지소비상태 진단부(200)로부터 도출된 진단결과를 의미하는 그룹에 따라, 에너지소비상태가 다른 그룹에 비해 높은 그룹에 대하여 DO(용존산소), 슬러지반송유량 및 잉여슬러지유량의 절감을 포함한 에너지절감이 가능한 운전조건을 사전에 설정하여 상기 운전조건의 변화량을 호출한다. 에너지소비상태에 대한 진단이 수행된 이후, 에너지소비상태가 높음 및 중간인 경우, 각 상태에 대한 에너지절감 운전조건을 호출한다. 에너지소비상태가 중간인 경우, 슬러지반송유량, 초침 폐슬러지 유량, 잉여슬러지 유량 및 SRT 등이 운전변수로 호출될 수 있으며, 에너지소비상태가 높은 경우, DO(용존산소)농도, SRT, 잉여슬러지 유량, 원심농축조 인발유량 및 혐기성소화조 유입유량 등이 운전변수로 호출될 수 있다. 각 상태별 호출된 운전변수의 설정치는 에너지소비상태가 낮은 경우의 유입유량과 상태별 유입유량을 비교하여 부하의 강도를 판단함으로써 설정치가 도출될 수 있다. 아래의 실시예에서 삭 상태별 호출되는 운전변수의 설정치가 상세히 설명될 것이다. 각 상태별 호출된 운전변수의 설정치는 하수처리장의 특성을 반영하여 정량적 수치를 조절하는 것이 바람직할 것이다. The energy saving driving condition call unit 300 serves to call a driving condition that can be set in advance according to a group meaning a diagnosis result derived from the energy consumption state diagnosis unit 200. The energy saving operation condition call unit 300 is DO (dissolved oxygen) for a group having a higher energy consumption state than other groups according to a group meaning a diagnosis result derived from the energy consumption state diagnosis unit 200. The amount of change in the operating condition is called by setting in advance an operating condition capable of energy saving, including the reduction of the sludge conveying flow rate and the excess sludge flow rate. After the diagnosis of the energy consumption state is performed, if the energy consumption state is high and medium, the energy saving operation condition for each state is called. In case of medium energy consumption, sludge conveying flow rate, second hand waste sludge flow rate, surplus sludge flow rate and SRT can be called as operating variables.In case of high energy consumption, DO (dissolved oxygen) concentration, SRT, excess sludge Flow rate, centrifugal tank draw flow, and anaerobic digester inflow can be called as operating variables. The set value of the called operation variable for each state can be derived by comparing the inflow flow rate when the energy consumption state is low with the inflow flow rate for each state to determine the strength of the load. In the following embodiments, the set values of the operating variables to be called for each state will be described in detail. It would be desirable to adjust the quantitative value to reflect the setpoint of operating variables called for each state by reflecting the characteristics of the sewage treatment plant.
상기 에너지소비 절감량 예측부(400)는 상기 에너지절감 운전조건 호출부(300)로부터 호출된 운전조건에 따라 다항회귀분석기법을 적용하여 에너지 절감량을 예측하는 역할을 한다. 상기 다항회귀분석기법은 상기 데이터호출부(100)에 의해 호출된 데이터와 동일한 운전변수를 활용하며, 상기 운전변수에 대해 독립변수와 종속변수로 나누어 에너지 사용량 예측을 하는 모델기법이다. 따라서 호출된 에너지절감 운전조건의 설정치를 입력으로 하여 에너지소비 절감량을 예측하는 모델을 사용할 수 있는데, 그 중에서도 다항회귀분석기법을 적용하는 것이 바람직하다. 상기 다항회귀분석기법에서 독립변수는 유입유량, 수온, SRT, DO농도, 슬러지반송유량, 초침 폐슬러지 유량, 잉여슬러지 유량, 원심농축슬러지 인발유량 및 소화조 유입슬러지 유량 등이 될 수 있으며, 종속변수는 하수처리장의 에너지를 대표하는 하수처리장 전체 전력량이 될 수 있다. 아래의 실시예에서 다항회귀분석기법을 적용한 에너지소비 절감량 예측을 확인해 볼 수 있을 것이다.  The energy consumption savings predicting unit 400 serves to predict energy savings by applying a polynomial regression analysis method according to the operating conditions called from the energy saving driving condition calling unit 300. The polynomial regression analysis technique utilizes the same operating variables as the data called by the data caller 100 and is a model technique for predicting energy usage by dividing the operating variables into independent and dependent variables. Therefore, a model for predicting the amount of energy consumption can be used by inputting the set value of the called energy-saving operation condition. Among them, it is preferable to apply the polynomial regression analysis technique. The independent variables in the polynomial regression analysis method may be influent flow rate, water temperature, SRT, DO concentration, sludge conveying flow rate, second hand waste sludge flow rate, surplus sludge flow rate, centrifugal concentrated sludge drawing flow rate and digester inflow sludge flow rate, etc. Can be the total amount of electricity from the sewage treatment plant representing the energy of the sewage treatment plant. In the following examples, the energy consumption savings prediction using the polynomial regression analysis method will be confirmed.
상기 에너지절감 운전조건 적용부(500)는 상기 에너지소비 절감량 예측부(400)에 의해 예측된 운전조건을 하수처리장의 펌프 및 송풍기를 포함하는 구동기(actuator)에 적용하는 역할을 한다. 만약 에너지소비상태가 중간 또는 높은 경우에, 호출된 운전변수의 설정치를 하수처리장의 펌프 및 송풍기를 포함하는 구동기(actuator)에 적용하여 에너지소비량을 절감할 수 있음을 확인하게 된다. 따라서 하수처리장 운전자는 정량적인 절감수치를 확인한 이후 동일한 운전조건으로 하수처리장을 운전하길 원할 경우, 각 구동기에 설정치를 입력하고 당일의 적용결과를 확인하게 된다. 호출된 절감운전 조건은 자동적으로 처리장 운전에 반영되는 것보다 하수처리장 운전장의 의사결정 이후 적용되는 것이 바람직할 것이다.The energy saving driving condition applying unit 500 serves to apply the driving conditions predicted by the energy consumption savings predicting unit 400 to an actuator including a pump and a blower in a sewage treatment plant. If the energy consumption state is medium or high, it is confirmed that the energy consumption can be reduced by applying the set value of the called operation variable to an actuator including a pump and a blower in the sewage treatment plant. Therefore, if the sewage treatment plant driver wants to operate the sewage treatment plant under the same operating conditions after confirming the quantitative savings value, the sewage treatment plant operator inputs the set value to each driver and checks the result of the day. It would be desirable to apply the reduced operating conditions called after the sewage plant plant decision rather than being automatically reflected in the plant operation.
도 2를 참조하여 본 발명에 의한 하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측방법을 설명하면 다음과 같다. Referring to Figure 2 describes the energy saving prediction method according to the diagnosis of energy consumption state sewage treatment plant according to the present invention.
제 1단계는 하수처리장의 유입/유출 수질 데이터 및 공정운영 데이터를 저장하는 데이터베이스로부터 에너지소비상태 진단에 필요한 데이터를 호출하는 데이터호출단계이다(S110). The first step is a data call step of calling data necessary for diagnosing an energy consumption state from a database storing inflow / outflow water quality data and process operation data of the sewage treatment plant (S110).
제 2단계는 상기 호출된 데이터를 전달받아 상기 호출된 데이터에 대해 다변량 통계 분석기법을 적용하여 상기 호출된 데이터에 내재되어 있는 에너지소비상태에 관한 정보를 추출하여 상기 추출된 에너지소비상태에 관한 정보로부터 에너지소비상태 진단결과를 도출하는 에너지소비상태 진단단계이다(S120). 상기 에너지소비상태 진단단계(S120)는 상기 호출된 데이터에 대해 상기 호출된 데이터에 내재되어 있는 에너지소비상태에 관한 정보를 추출하기 위해 상기 호출된 데이터를 축약하고, 상기 축약된 데이터를 이용하여 현재 에너지소비상태에 관한 진단결과를 도출하기 위해 상기 축약된 데이터를 그룹화하고, 상기 그룹화된 데이터에 대하여 새로운 데이터가 어떠한 그룹에 속하는지를 판별하기 위한 수단으로 사용될 판별함수를 Fisher의 선형판별분석을 이용하여 도출하고, 상기 도출된 판별함수를 이용하여 상기 호출된 데이터의 에너지소비상태 진단결과를 도출하게 된다.  The second step is to receive the called data and to apply the multivariate statistical analysis method to the called data to extract information on the energy consumption state inherent in the called data to the information on the extracted energy consumption state The energy consumption state diagnosis step of deriving an energy consumption state diagnosis result from (S120). The energy consumption state diagnosing step (S120) may abbreviate the called data to extract information on the energy consumption state that is inherent in the called data with respect to the called data, and use the reduced data to present Using Fisher's linear discriminant analysis, a discriminant function is used to group the abbreviated data to derive a diagnosis result regarding energy consumption and to be used as a means for determining which group the new data belongs to for the grouped data. Derivation of the energy consumption state of the called data is derived using the derived discrimination function.
제 3단계는 상기 도출된 진단결과를 의미하는 그룹에 따라 사전에 설정된 에너지절감이 가능한 운전조건을 호출하는 에너지절감 운전조건 호출단계이다(S130). The third step is an energy saving operation condition call step of calling an operation condition that can be set in advance according to the group representing the derived diagnosis result (S130).
제 4단계는 상기 호출된 운전조건에 따라 다항회귀분석기법을 적용하여 에너지 절감량을 예측하는 에너지 소비 절감량 예측단계이다(S140). The fourth step is an energy consumption savings prediction step of predicting an energy saving amount by applying a polynomial regression analysis method according to the called operation condition (S140).
제 5단계는 상기 예측된 운전조건을 하수처리장의 펌프 및 송풍기를 포함하는 구동기(actuator)에 적용하는 에너지절감 운전조건 적용단계이다(S150).The fifth step is an energy saving operation condition application step of applying the predicted operation condition to an actuator including a pump and a blower in the sewage treatment plant (S150).
이하, 실시예를 기준으로 본 발명에서 언급하는 하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측방법을 설명하기로 한다. Hereinafter, the energy saving prediction method according to the diagnosis of the energy consumption state of the sewage treatment plant mentioned in the present invention will be described with reference to Examples.
본 실시예에서는 B시에 위치한 N하수처리장을 대상으로 하여 일별 데이터셋들이 수집되었다. 수집된 데이터는 2008년 1월 1일부터 2010년 12월 31일까지의 총 3년간의 데이터였고, 호출된 운전변수는 유입유량(Qin), 수온(Temp), SRT, DO(용존산소)농도, 슬러지반송 유량(Qras), 초침 폐슬러지 유량(Qfirst), 잉여슬러지 유량(Qsecond), 원심농축슬러지 인발량(Qthickener), 소화조 유입슬러지 유량(Qdin) 및 하수처리장 전체 사용 전력량(Elec)이었다.  In this example, daily data sets were collected for the N sewage treatment plant located in city B. The collected data were for three years from January 1, 2008 to December 31, 2010. The operating variables called were influent flow rate (Qin), water temperature (Temp), SRT, and DO (dissolved oxygen) concentration. , Sludge conveyance flow rate (Qras), second hand waste sludge flow rate (Qfirst), surplus sludge flow rate (Qsecond), centrifugal concentration sludge withdrawal amount (Qthickener), digester inflow sludge flow rate (Qdin), and total power consumption of sewage treatment plant (Elec).
유입유량(Qin)은 유입유량 펌프의 구동과 연계되며, 수온(Temp)은 계절적 영향에 따른 공정운전변동, SRT는 폐슬러지 펌프구동, DO농도는 송풍기 구동, 슬러지반송 유량(Qras)은 반송슬러지 펌프구동, 초침 폐슬러지 유량(Qfirst)은 초침 폐슬러지 펌프 구동, 잉여슬러지 유량(Qsecond)은 폐슬러지 펌프 구동, 원심농축슬러지 인발량(Qthickener)은 원심농축기 구동, 소화조 유입슬러지 유량(Qdin)은 혐기성 소화를 통한 메탄가스 발생에 따른 에너지절감과 연계가 되기 때문에, 해당 하수처리장의 경우 이와 같은 운전변수들이 선택되어 호출되었다.  The inflow flow rate (Qin) is linked to the inflow flow pump operation, the water temperature (Temp) is the process operation variation according to the seasonal influence, the SRT is the waste sludge pump driving, the DO concentration is the blower driving, the sludge conveying flow rate (Qras) is the return sludge Pump driving, second hand waste sludge flow rate (Qfirst), second hand waste sludge pump drive, surplus sludge flow rate (Qsecond), waste sludge pump drive, centrifugal concentration sludge drawing (Qthickener) centrifugal condenser drive, digester inflow sludge flow rate (Qdin) In connection with the energy savings associated with methane generation through anaerobic digestion, these sewage treatment plants were selected and called.
호출된 데이터들에 대해서 다변량 통계 분석기법들을 적용하여 에너지소비 상태 진단이 수행되었다. 먼저 주성분 분석을 수행하여, 주성분 분석을 통해 고유 값이 1이상인 성분을 선택한 결과, 4개의 주성분(PC1, PC2, PC3, PC4)이 선택되었다. 이는 10차원의 데이터셋이 새로운 4차원의 데이터셋으로 변형되었음을 의미하며, 새로운 데이터가 입력되었을 경우, 각 운전 변수값과 주성분 계수값을 통해 최종적으로 다음과 같은 주성분값이 계산하게 된다. Energy consumption status diagnosis was performed using multivariate statistical analytical techniques on the called data. First, principal component analysis was performed, and as a result, four principal components (PC1, PC2, PC3, and PC4) were selected through principal component analysis. This means that the 10-dimensional data set is transformed into a new 4-dimensional data set. When new data is input, the following main component values are finally calculated through each operation variable value and the principal component coefficient value.
PC1 = 0.848*Qin + 0.512*Temp - 0.697*SRT + 0.069*DO + 0.424*Qras - 0.373*Qfirst - 0.084*Qsecond - 0.1*Qthickener - 0.058*Qdin + 0.086*Elec PC1 = 0.848 * Qin + 0.512 * Temp-0.697 * SRT + 0.069 * DO + 0.424 * Qras-0.373 * Qfirst-0.084 * Qsecond-0.1 * Qthickener-0.058 * Qdin + 0.086 * Elec
PC2 = 0.06*Qin + 0.019*Temp + 0.169*SRT + 0.075*DO + 0.511*Qras + 0.401*Qfirst + 0.747*Qsecond + 0.155*Qthickener + 0.791*Qdin - 0.061*Elec PC2 = 0.06 * Qin + 0.019 * Temp + 0.169 * SRT + 0.075 * DO + 0.511 * Qras + 0.401 * Qfirst + 0.747 * Qsecond + 0.155 * Qthickener + 0.791 * Qdin-0.061 * Elec
PC3 = -0.153*Qin - 0.747*Temp + 0.249*SRT + 0.892*DO + 0.173*Qras + 0.352*Qfirst + 0.159*Qsecond + 0.017*Qthickener - 0.212*Qdin + 0.094*Elec PC3 = -0.153 * Qin-0.747 * Temp + 0.249 * SRT + 0.892 * DO + 0.173 * Qras + 0.352 * Qfirst + 0.159 * Qsecond + 0.017 * Qthickener-0.212 * Qdin + 0.094 * Elec
PC4 = -0.038*Qin - 0.2*Temp + 0.314*SRT - 0.121*DO - 0.259*Qras - 0.439*Qfirst + 0.019*Qsecond + 0.893*Qthickener + 0.282*Qdin + 0.141*Elec PC4 = -0.038 * Qin-0.2 * Temp + 0.314 * SRT-0.121 * DO-0.259 * Qras-0.439 * Qfirst + 0.019 * Qsecond + 0.893 * Qthickener + 0.282 * Qdin + 0.141 * Elec
도출된 주성분값을 사용하여 K-평균 군집분석을 통해 현재의 에너지소비 상태를 분류할 수 있는 그룹을 총 3개로 도출하였다. 도출된 각 그룹별 특성을 [표 1]에 나타내었다. Using the derived principal component values, K-average cluster analysis was used to derive a total of three groups that can classify the current energy consumption. The characteristics of each group derived are shown in [Table 1].
표 1
항목 Qin Elec Temp SRT DO Qras Qfirst Qsecond Qthickener Qdin
Group1 Low High Low Medium Low Medium Medium
Mean 286,775.4 63,859.5 19.0 6.0 1.75 60,112.2 1,195.4 2,217.9 282.9 926.3
Group2 Medium High Medium High Low High
Mean 328,707.2 66281.8 22.6 5.2 1.76 73,051.0 1,308.1 2,989.1 274.7 1,190.8
Group3 High Medium Low High Medium Low Medium High Low
Mean 336,875.9 71206.0 19.5 4.5 2.51 69,694.1 1,086.4 2,290.0 310.9 899.7
Table 1
Item Qin Elec Temp SRT DO Qras Qfirst Qsecond Qthickener Qdin
Group1 Low High Low Medium Low Medium Medium
Mean 286,775.4 63,859.5 19.0 6.0 1.75 60,112.2 1,195.4 2,217.9 282.9 926.3
Group2 Medium High Medium High Low High
Mean 328,707.2 66281.8 22.6 5.2 1.76 73,051.0 1,308.1 2,989.1 274.7 1,190.8
Group3 High Medium Low High Medium Low Medium High Low
Mean 336,875.9 71206.0 19.5 4.5 2.51 69,694.1 1,086.4 2,290.0 310.9 899.7
도출된 각 그룹별 에너지소비상태 특성을 통한 진단결과는 다음과 같은 정성적인 정보로 운전자에게 제시될 수 있다. Diagnosis results through energy consumption status characteristics of each group can be presented to the driver with the following qualitative information.
- 1 그룹:낮은 유입 부하에 따라 수온이 낮음에도 DO 농도를 낮게 유지할 수 있어 송풍 전력비가 적게 소요되며, 긴 SRT에 따른 잉여슬러지 펌프의 저가동으로 에너지 소비가 낮음.-1 group: It can keep the DO concentration low even though the water temperature is low due to the low inflow load, and the air consumption cost is low, and the energy consumption is low due to the low cost of surplus sludge pump according to the long SRT.
- 2 그룹:높은 수온에도 유입 부하에 따라 반송슬러지 유량이 증가되어 펌프 가동이 증가되며, 초침 및 잉여슬러지 유량 증가에 따른 펌프 구동에 따라 에너지 소비는 중간임.-Group 2: The pumping operation is increased due to the increase of the return sludge flow rate according to the inflow load, even at high water temperature, and the energy consumption is medium by the operation of the pump with the increase of the second hand and surplus sludge flow rate.
- 3 그룹:유입 부하가 높고 이에 따라 DO 농도를 높게 유지하게 되어 송풍 전력비가 크게 소요되며, 짧은 SRT에 따른 잉여슬러지 펌프의 가동 증가, 원심농축조 효율 저하로 인한 구동 시간 증가 및 낮은 혐기성 소화조 유입 유량으로 메탄 가스 발생량이 낮아 에너지 소비가 높음.-Group 3: High influent load, thus maintaining high DO concentration, which greatly increases the blowing power cost, increases the operation of surplus sludge pumps due to short SRT, increases the operating time due to declining efficiency of centrifugal condenser, and lowers the flow rate of the anaerobic digester High energy consumption due to low methane gas generation.
새로운 데이터셋에 대하여 이와 같은 에너지상태 그룹 중 어디에 소속되는 지를 판단하기 위해 Fisher의 선형 판별분석이 수행된다. 판별분석을 통해 각 그룹별 판별함수식이 다음과 같이 도출되었다.  Fisher's linear discriminant analysis is performed to determine which of these groups of energy states belongs to the new dataset. Through discriminant analysis, the discriminant function for each group was derived as follows.
Group 1 = -2.217*PC1 - 1.472*PC2 - 0.679*PC3 - 0.018*PC4 - 1.991 Group 1 = -2.217 * PC1-1.472 * PC2-0.679 * PC3-0.018 * PC4-1.991
Group 2 = 0.849*PC1 + 2.962*PC2 - 0.881*PC3 - 0.701*PC4 - 3.379 Group 2 = 0.849 * PC1 + 2.962 * PC2-0.881 * PC3-0.701 * PC4-3.379
Group 3 = 2.976*PC1 - 0.477*PC2 + 2.077*PC3 + 0.748*PC4 - 3.778 Group 3 = 2.976 * PC1-0.477 * PC2 + 2.077 * PC3 + 0.748 * PC4-3.778
상기의 판별함수에 먼저 도출된 주성분값을 대입하여 각 그룹의 함수의 값을 계산하게 되고, 가장 큰 값을 가지게 되는 그룹이 현재의 에너지 소비 상태를 대변하는 그룹으로 해석되었다. The value of the function of each group is calculated by substituting the principal component values first derived into the discrimination function, and the group having the largest value is interpreted as a group representing the current energy consumption state.
에너지 소비상태가 중간 및 높은 그룹인 2 그룹과 3 그룹의 경우, 에너지 절감이 가능한 운전조건을 호출하게 된다. 에너지 소비 상태에 영향을 주는 변수 중 수온과 같은 기상 조건 및 정상적인 운전 조건 변수를 제외하여, 2 그룹은 슬러지반송 유량, 초침 폐슬러지 유량, 잉여슬러지 유량 및 SRT가 운전 변수로 호출되며, 3 그룹은 DO농도, SRT, 잉여슬러지 유량, 원심농축조 인발유량 및 혐기성소화조 유입 유량이 운전 변수로 호출되게 된다. 호출된 각 운전 변수의 정량적 설정치는 유입 부하와 연계하여 그 적정값을 도출하였다. 한 예로, 3 그룹의 DO농도의 경우, [표 1]에서 그룹 1과 그룹 3의 유입 유량 증가는 약 17%였으나, DO농도는 45%가 증가하였다. 이는 유입 부하를 고려하더라도 높은 수치의 증가이므로, 17%와 45% 사이의 적정 수치로써 20%를 적정값으로 도출하였다. 이와 같은 방법으로 다음과 같이 2 그룹과 3 그룹의 에너지 절감 운전 조건 설정치를 도출하였다. In groups 2 and 3, which are medium and high energy consumption groups, the operating conditions for saving energy are called. Except for weather conditions such as water temperature and normal operating conditions among the variables that affect the energy consumption state, the 2 groups are called sludge conveying flow rate, second hand waste sludge flow rate, surplus sludge flow rate and SRT as the operating variables. DO concentration, SRT, excess sludge flow rate, centrifugal condensate draw flow and anaerobic digester inflow flow will be called as operating parameters. The quantitative setpoint of each called operation variable was derived in conjunction with the influent load. For example, in the case of the DO concentration of the three groups, the inflow increase of the group 1 and the group 3 in Table 1 was about 17%, but the DO concentration was increased by 45%. This is a high increase even when considering the inflow load, so 20% was derived as an appropriate value between 17% and 45%. In this way, the energy-saving operating condition settings of Groups 2 and 3 were derived as follows.
- 2 그룹 운전 조건 설정치(중간 에너지 소비 상태):슬러지반송 유량 30% 증가, 초침 폐슬러지 유량 20% 증가, 잉여슬러지 유량 30% 감소, SRT 30% 증가 -Group 2 operating condition set point (medium energy consumption): 30% increase in sludge conveying flow rate, 20% increase in second hand waste sludge flow rate, 30% decrease in excess sludge flow rate, 30% increase in SRT
- 3 그룹 운전 조건 설정치(높은 에너지 소비 상태):DO 농도 20% 감소, SRT 30% 증가, 잉여슬러지 유량 30% 감소, 원심농축조 인발유량 15% 증가, 혐기성소화조 유입 유량 20% 증가 3 group operating condition set point (high energy consumption): 20% decrease in DO concentration, 30% increase in SRT, 30% decrease in excess sludge flow rate, 15% increase in centrifugal condensate draw flow, 20% increase in inflow rate of anaerobic digester
호출된 에너지 절감 운전 조건의 설정치를 입력으로 하여 에너지 절감 예측량을 정량적으로 도출하기 위해 본 발명에서는 다항회귀분석 기법을 활용하여 예측 모델을 개발하였다. 에너지소비 상태진단과 동일한 운전변수를 활용하여 다항 회귀분석에서 독립변수는 유입 유량, 수온, SRT, DO농도, 슬러지반송 유량, 초침 폐슬러지 유량, 잉여슬러지 유량, 원심농축슬러지 인발량 및 소화조 유입 슬러지 유량으로 총 9개이며, 종속변수는 하수처리장 에너지를 대표하는 처리장 전체 전력량이  In order to quantitatively derive the energy saving prediction quantity by inputting the set value of the called energy saving operation condition, the present invention has developed a prediction model using a polynomial regression analysis technique. In the multinomial regression analysis, the independent variables were the influent flow rate, water temperature, SRT, DO concentration, sludge conveying flow rate, second hand waste sludge flow rate, surplus sludge flow rate, centrifugal concentrated sludge drawing amount, and digester inflow sludge. The total flow rate is 9, and the dependent variable is the total amount of power in the treatment plant representing the sewage treatment plant energy.
된다. 총 300개의 일별 데이터 셋을 활용하여 예측 모델의 개발에 사용하였으며, 이 결과를 도 3에 나타내었다. 개발 결과, 하수처리장에서 소요되고 있는 전력량의 거동을 개발된 예측 모델이 성공적으로 모사할 수 있음을 확인할 수 있었으며, 보다 정량적인 분석을 위해 RMSE(Root Mean Square Error)값을 도출하였다. 계산에 사용된 RMSE 분석식은 다음과 같다.do. A total of 300 daily data sets were used to develop the predictive model, and the results are shown in FIG. 3. As a result of the development, it was confirmed that the developed prediction model can successfully simulate the behavior of power consumption in sewage treatment plant, and the root mean square error (RMS) value was derived for more quantitative analysis. The RMSE equation used for the calculation is as follows.
[규칙 제26조에 의한 보정 21.05.2013]　
Figure WO-DOC-67
[Revision 21.05.2013 under Rule 26]
Figure WO-DOC-67
(여기서, N은 변수의 총 수, Xm,i은 측정값, Xs,i은 시뮬레이션 결과값을 나타냄.) Where N is the total number of variables, Xm, i is the measured value, and Xs, i is the simulation result.
예측 모델 개발에서 하수처리장에서 사용된 전력량과 모델 시뮬레이션 결과 사이의 RMSE는 약 1,603 kWh로 5% 이하의 낮은 오차로 확인되었다.  In the development of the predictive model, the RMSE between the amount of power used in the sewage treatment plant and the model simulation results was found to be about 1,603 kWh with a low error of less than 5%.
개발된 예측 모델의 검증을 위해 총 95개의 일별 데이터 셋이 활용되었으며, 이 결과를 도 4에 나타내었다. 검증 결과, RMSE는 약 1,818 kWh로 모델 개발에서의 RMSE값보다는 높지만 역시 5% 이하의 오차로써 예측 모델이 성공적으로 검증되었음을 확인하였다. A total of 95 daily data sets were used to verify the developed prediction model, and the results are shown in FIG. 4. As a result, the RMSE was about 1,818 kWh, which is higher than the RMSE value in the model development but also less than 5% of the error.
개발된 예측 모델을 활용하여 2 그룹과 3 그룹의 에너지 소비 상태의 경우에서 호출된 에너지 절감 운전 조건의 설정치를 입력하여 절감 가능한 에너지량을 확인해 보았다. 2 그룹에 대하여 운전 변수의 값을 조정하지 않은 경우와 슬러지반송 유량 30% 증가, 초침 폐슬러지 유량 20% 증가, 잉여슬러지 유량 30% 감소 및 SRT를 30% 증가한 새로운 운전 변수의 설정치를 입력한 경우의 비교를 도 5에 나타내었다. 새로운 운전 변수의 설정치를 통해 하수처리장 에너지 사용량은 최소 932.5 kWh에서 최대 1,059.9 kWh까지 절감될 수 있음을 확인하였다.  The developed predictive model was used to check the amount of energy that could be saved by entering the set values of the energy saving operating conditions called in the case of energy consumption of group 2 and group 3. When the value of the operating variable is not adjusted for the two groups and when the set value of the new operation variable is increased by 30% increase in sludge conveying flow rate, 20% increase in second hand waste sludge flow rate, 30% decrease in excess sludge flow rate, and 30% increase in SRT. The comparison of is shown in FIG. The new set of operating parameters confirmed that sewage treatment plant energy consumption can be reduced from a minimum of 932.5 kWh to a maximum of 1,059.9 kWh.
3 그룹에 대하여 운전 변수의 값을 조정하지 않은 경우와 DO 농도 20% 감소, SRT 30% 증가, 잉여슬러지 유량 30% 감소, 원심농축조 인발유량 15% 증가 및 혐기성소화조 유입 유량을 20% 증가한 운전 변수의 설정치를 입력한 경우의 비교를 도 6에 나타내었다. 3 그룹에서의 경우, 운전 변수의 설정치 변경을 통해 최소 734 kWh에서 최대 866.6 kWh까지 에너저 소비량이 절감될 수 있음을 확인하였다. Operational variables with no adjustment of operating variables for group 3 and with a 20% decrease in DO concentration, 30% increase in SRT, 30% reduction in excess sludge flow rate, 15% increase in centrifugal condensate draw flow, and 20% increase in anaerobic digester influent flow. The comparison in the case of inputting the set value of is shown in FIG. In group 3, it was found that changing the setpoint of operating variables could reduce energy consumption from a minimum of 734 kWh to a maximum of 866.6 kWh.
도출된 운전 변수의 설정치를 하수처리장 운전자가 확인 후 운전자 의사결정을 통해 실제 운영에 반영한다면, 펌프 및 송풍기의 구동을 조절하게 됨으로써, 하수처리장에서 소비되는 에너지량을 절감할 수 있을 것으로 판단된다. If the sewage plant operator's setting value is reflected to the actual operation through the driver's decision after checking the sewage treatment plant driver's decision, it is possible to reduce the amount of energy consumed in the sewage treatment plant by adjusting the driving of the pump and the blower.
이상의 설명은 본 발명을 예시적으로 설명한 것에 불과한 것으로, 본 발명이 속하는 기술분야에서 통상의 지식을 가지는 자라면 본 발명의 본질적인 특성에서 벗어나지 않는 범위에서 다양한 변형이 가능할 것이다. 따라서 본 명세서에 개시된 실시예들은 본 발명을 한정하기 위한 것이 아니라 설명하기 위한 것이고, 이러한 실시예에 의하여 본 발명의 사상과 범위가 한정되는 것은 아니다. 본 발명의 범위는 아래의 청구범위에 의하여 해석되어야 하며, 그와 동등한 범위 내에 있는 모든 기술은 본 발명의 권리범위에 포함되는 것으로 해석되어야 할 것이다.The above description is merely illustrative of the present invention, and those skilled in the art to which the present invention pertains may various modifications without departing from the essential characteristics of the present invention. Therefore, the embodiments disclosed herein are not intended to limit the present invention but to describe the present invention, and the spirit and scope of the present invention are not limited by these embodiments. It is intended that the scope of the invention be interpreted by the following claims, and that all descriptions within the scope equivalent thereto shall be construed as being included in the scope of the present invention.
본 발명은 하수처리장 운전에 있어 필수적인 펌프와 송풍기와 같은 기기의 구동으로 인해 발생되는 에너지 소비에 대한 정성적인 상태 진단을 수행하며, 진단 결과에 따라 에너지 절감이 가능한 운전 조건의 설정치를 제공하여 이에 따른 에너지 사용량의 감소를 사전에 확인하여 에너지 사용 절감을 위한 실질적인 운전 설정치를 제공함으로서 하수처리장에 널리 사용될 수 있다. The present invention performs a qualitative diagnosis of energy consumption caused by the operation of equipment such as pumps and blowers, which are essential for the operation of sewage treatment plants, and provides a set value of operating conditions that can save energy according to the diagnosis result. It can be widely used in sewage treatment plants by proactively identifying the reduction in energy usage and providing practical operating setpoints to reduce energy use.

Claims (7)

  1. 하수처리장의 유입/유출 수질 데이터 및 공정운영 데이터를 저장하는 데이터베이스로부터 에너지소비상태 진단에 필요한 데이터를 호출하는 데이터호출부; A data caller for calling data necessary for diagnosing an energy consumption state from a database storing inflow / outflow water quality data and process operation data of the sewage treatment plant;
    상기 데이터호출부로부터 호출된 데이터를 전달받아 상기 호출된 데이터에 대해 다변량 통계 분석기법을 적용하여 상기 호출된 데이터에 내재되어 있는 에너지소비상태에 관한 정보를 추출하여 상기 추출된 에너지소비상태에 관한 정보로부터 에너지소비상태 진단결과를 도출하는 에너지소비상태 진단부; Receives the data called from the data caller and applies the multivariate statistical analysis method to the called data to extract information on the energy consumption state inherent in the called data and to extract the information on the energy consumption state. An energy consumption state diagnosis unit for deriving an energy consumption state diagnosis result from the;
    상기 에너지소비상태 진단부로부터 도출된 진단결과를 의미하는 그룹에 따라 사전에 설정된 에너지절감이 가능한 운전조건을 호출하는 에너지절감 운전조건 호출부; An energy saving operation condition calling unit for calling a preset energy saving operation condition according to a group representing a diagnosis result derived from the energy consumption state diagnosis unit;
    상기 에너지절감 운전조건 호출부로부터 호출된 운전조건에 따라 다항회귀분석기법을 적용하여 에너지 절감량을 예측하는 에너지소비 절감량 예측부; 및 An energy consumption reduction prediction unit for predicting an energy saving amount by applying a polynomial regression analysis method according to the operation condition called from the energy saving operation condition calling unit; And
    상기 에너지소비 절감량 예측부에 의해 예측된 운전조건을 하수처리장의 펌프 및 송풍기를 포함하는 구동기(actuator)에 적용하는 에너지절감 운전조건 적용부;를 포함하는 하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측시스템. Energy saving prediction according to the energy consumption state diagnosis of the sewage treatment plant, including; energy saving operation condition applying unit for applying the operating conditions predicted by the energy consumption reduction prediction unit to the actuator (actuator) including a pump and a blower of the sewage treatment plant system.
  2. 제 1항에 있어서,The method of claim 1,
    상기 데이터호출부는 개별 하수처리장에서 별도로 운영되고 있는 데이터베이스에 축적되는 과거의 모든 데이터 중 에너지 소비와 연관성이 있는 주요 데이터들을 자동으로 호출하는 것을 특징으로 하는 하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측시스템.The data caller is an energy saving prediction system according to the diagnosis of energy consumption status of the sewage treatment plant, which automatically calls the main data related to the energy consumption among all the data accumulated in the database operated separately in the individual sewage treatment plant. .
  3. 제 2항에 있어서,The method of claim 2,
    상기 에너지소비상태 진단부는 The energy consumption state diagnosis unit
    상기 호출된 데이터에 대해 상기 호출된 데이터에 내재되어 있는 에너지소비상태에 관한 정보를 추출하기 위해 상기 호출된 데이터를 축약하는 데이터축약부; A data condensing unit for condensing the called data to extract information on the energy consumption state inherent in the called data with respect to the called data;
    상기 축약된 데이터를 이용하여 현재 에너지소비상태에 관한 진단결과를 도출하기 위해 상기 축약된 데이터를 그룹화하는 데이터그루핑부; 및A data grouping unit for grouping the reduced data to derive a diagnosis result regarding a current energy consumption state using the reduced data; And
    상기 그룹화된 데이터에 대하여 새로운 데이터가 어떠한 그룹에 속하는지를 판별하기 위한 수단으로 사용될 판별함수를 Fisher의 선형판별분석을 이용하여 도출하는 판별함수도출부;를 포함하되,And a discrimination function derivation unit for deriving a discrimination function to be used as a means for determining which group the new data belongs to with respect to the grouped data using Fisher's linear discrimination analysis.
    상기 에너지소비상태 진단부는 상기 판별함수도출부에 의해 도출된 판별함수를 이용하여 상기 호출된 데이터의 에너지소비상태 진단결과를 도출하는 것을 특징으로 하는 하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측시스템. The energy consumption state diagnosis system according to the energy consumption state diagnosis of the sewage treatment plant, characterized in that for deriving the energy consumption state diagnosis results of the called data using the determination function derived by the determination function derivation unit.
  4. 제 3항에 있어서,The method of claim 3, wherein
    상기 에너지절감 운전조건 호출부는 상기 에너지소비상태 진단부로부터 도출된 진단결과를 의미하는 그룹에 따라, 에너지소비상태가 다른 그룹에 비해 높은 그룹에 대하여 DO(용존산소), 슬러지반송유량 및 잉여슬러지유량의 절감을 포함한 에너지절감이 가능한 운전조건을 사전에 설정하여 상기 운전조건의 변화량을 호출하는 것을 특징으로 하는 하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측시스템. The energy saving operation condition call unit may perform DO (dissolved oxygen), sludge conveyance flow rate, and excess sludge flow rate for a group having a higher energy consumption state than other groups according to a group meaning a diagnosis result derived from the energy consumption state diagnosis unit. Energy saving prediction system according to the diagnosis of energy consumption status of the sewage treatment plant, characterized in that to call the amount of change in the operating conditions in advance by setting the operating conditions capable of energy saving, including the reduction of.
  5. 제 4항에 있어서, The method of claim 4, wherein
    상기 다항회귀분석기법은 상기 데이터호출부에 의해 호출된 데이터와 동일한 The polynomial regression analysis method is the same as the data called by the data caller.
    운전변수를 활용하며, 상기 운전변수에 대해 독립변수와 종속변수로 나누어 에너지 사용량 예측을 하는 모델기법인 것을 특징으로 하는 하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측시스템. Energy saving prediction system according to the diagnosis of energy consumption status of sewage treatment plant, characterized in that using the operating variables, the model technique for predicting the energy consumption by dividing the independent and dependent variables for the operating variable.
  6. 하수처리장의 유입/유출 수질 데이터 및 공정운영 데이터를 저장하는 데이터베이스로부터 에너지소비상태 진단에 필요한 데이터를 호출하는 데이터호출단계; A data call step of calling data necessary for diagnosing an energy consumption state from a database storing inflow / outflow water quality data and process operation data of the sewage treatment plant;
    상기 호출된 데이터를 전달받아 상기 호출된 데이터에 대해 다변량 통계 분석기법을 적용하여 상기 호출된 데이터에 내재되어 있는 에너지소비상태에 관한 정보를 추출하여 상기 추출된 에너지소비상태에 관한 정보로부터 에너지소비상태 진단결과를 도출하는 에너지소비상태 진단단계; Receives the called data and applies the multivariate statistical analysis method to the called data to extract information on the energy consumption state inherent in the called data and to extract the energy consumption state from the information on the extracted energy consumption state. An energy consumption state diagnosis step of deriving a diagnosis result;
    상기 도출된 진단결과를 의미하는 그룹에 따라 사전에 설정된 에너지절감이 가능한 운전조건을 호출하는 에너지절감 운전조건 호출단계; An energy saving operation condition calling step of calling an operation condition capable of saving energy in advance according to the group representing the derived diagnosis result;
    상기 호출된 운전조건에 따라 다항회귀분석기법을 적용하여 에너지 절감량을 예측하는 에너지 소비 절감량 예측단계; 및 An energy consumption savings predicting step of predicting an energy saving amount by applying a polynomial regression analysis method according to the called driving condition; And
    상기 예측된 운전조건을 하수처리장의 펌프 및 송풍기를 포함하는 구동기(actuator)에 적용하는 에너지절감 운전조건 적용단계;를 포함하는 하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측방법. Energy saving prediction method according to the energy consumption state diagnosis of the sewage treatment plant comprising the; applying the energy saving operating conditions applying the predicted operating conditions to the actuator (actuator) including a pump and a blower in the sewage treatment plant.
  7. 제 6항에 있어서,The method of claim 6,
    상기 에너지소비상태 진단단계는 상기 호출된 데이터에 대해 상기 호출된 데이터에 내재되어 있는 에너지소비상태에 관한 정보를 추출하기 위해 상기 호출된 데이터를 축약하고, 상기 축약된 데이터를 이용하여 현재 에너지소비상태에 관한 진단결과를 도출하기 위해 상기 축약된 데이터를 그룹화하고, 상기 그룹화된 데이터에 대하여 새로운 데이터가 어떠한 그룹에 속하는지를 판별하기 위한 수단으로 사용될 판별함수를 Fisher의 선형판별분석을 이용하여 도출하고, 상기 도출된 판별함수를 이용하여 상기 호출된 데이터의 에너지소비상태 진단결과를 도출하는 것을 특징으로 하는 하수처리장 에너지소비상태 진단에 따른 에너지 절감 예측방법. The energy consumption state diagnosing step may abbreviate the called data to extract information on the energy consumption state inherent in the called data for the called data, and use the reduced data to present the current energy consumption state. Grouping the abbreviated data to derive a diagnosis result of, and using Fisher's linear discriminant analysis, a discrimination function to be used as a means for determining which group the new data belongs to the grouped data, The energy saving prediction method according to the energy consumption state diagnosis of the sewage treatment plant, characterized in that to derive the energy consumption state diagnosis result of the called data using the derived discrimination function.
PCT/KR2013/002580 2013-03-27 2013-03-28 Energy-saving prediction system and method based on diagnosis of state of energy consumption in sewage treatment plant WO2014157751A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
KR10-2013-0032982 2013-03-27
KR1020130032982A KR20140117966A (en) 2013-03-27 2013-03-27 System and method for predicting energy reduction according to energy consumption state diagnosis of wastewater treatment plant

Publications (1)

Publication Number Publication Date
WO2014157751A1 true WO2014157751A1 (en) 2014-10-02

Family

ID=51624703

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2013/002580 WO2014157751A1 (en) 2013-03-27 2013-03-28 Energy-saving prediction system and method based on diagnosis of state of energy consumption in sewage treatment plant

Country Status (2)

Country Link
KR (1) KR20140117966A (en)
WO (1) WO2014157751A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111915089A (en) * 2020-08-07 2020-11-10 青岛洪锦智慧能源技术有限公司 Method and device for predicting pump set energy consumption of sewage treatment plant

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105549388B (en) * 2015-12-12 2018-03-13 北京工业大学 A kind of sewage disposal process energy consumption Forecasting Methodology based on adaptive recurrence kernel function

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20090078502A (en) * 2008-01-15 2009-07-20 부산대학교 산학협력단 Apparatus and method for diagnosis of operating states in municipal wastewater treatment plant

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20090078502A (en) * 2008-01-15 2009-07-20 부산대학교 산학협력단 Apparatus and method for diagnosis of operating states in municipal wastewater treatment plant

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
KIM, HYO SOO ET AL.: "Development of Diagnosis Algorithm for Energy Consumption State with Data Mining Technologies in Wastewater Treatment Plant", JOURNAL OF 2010 COMMON CONFERENCE, 21 March 2012 (2012-03-21), pages 320 - 321 *
KIM, MIN JEONG ET AL.: "Design of a Wastewater Treatment Plant Upgrading to Advanced Nutrient Removal Treatment Using Modeling Methodology and Multivariate Statistical Analysis for Process Optimization", KOREAN CHEM. ENG. RES., vol. 48, October 2010 (2010-10-01), pages 589 - 597 *
LEE, BYEONG GUK: "Study on Expansion Plan of Automatic Facilities Large-scale Sewage Treatment Plants", KEI/2000 RESEARCH REPORT, December 2000 (2000-12-01), pages 1 - 57 *
OSAMU YAMANAKA ET AL.: "A Knowledge Discovery Assistance Method Using Multivariate Statistics for Efficient Wastewater Treatment Plant Operation", JOURNAL OF WATER AND ENVIRONMENT TECHNOLOGY, vol. 10, 2012, JAPAN, pages 87 - 99 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111915089A (en) * 2020-08-07 2020-11-10 青岛洪锦智慧能源技术有限公司 Method and device for predicting pump set energy consumption of sewage treatment plant

Also Published As

Publication number Publication date
KR20140117966A (en) 2014-10-08

Similar Documents

Publication Publication Date Title
CN103744293B (en) Wastewater treatment monitoring method and system based on fuzzy neural network
Rieger et al. Improving nutrient removal while reducing energy use at three Swiss WWTPs using advanced control
CN102491506B (en) Method and system for controlling aeration biofilter of sewage treatment plant
CN209583779U (en) A kind of sewage treatment accurate aeration system
WO2013125737A1 (en) Rule-based control apparatus and rule-based control method based on diagnosis of process state of sewage treatment plant
Günther et al. Species‐sorting and mass‐transfer paradigms control managed natural metacommunities
CN108088974B (en) Soft measurement method for effluent nitrate nitrogen in anaerobic simultaneous denitrification methanogenesis process
WO2012002713A2 (en) System and method for diagnosing the processes of a sewage and wastewater treatment plant
CN109592804A (en) A kind of sewage treatment approach follows excellent accurate aeration method
CN103792844B (en) A kind of sewage treatment control method based on orderly cluster
WO2014157751A1 (en) Energy-saving prediction system and method based on diagnosis of state of energy consumption in sewage treatment plant
CN111039383A (en) Carbon source adding method and system
CN114772873A (en) Sewage treatment system based on monitoring of Internet of things
KR101767368B1 (en) Supporting method and system of sewage disposal for energy saving
CN114565154A (en) Prediction method and optimization algorithm for carbon source adding amount of biochemical section of leachate
WO2015147351A1 (en) System for diagnosing operation state of anaerobic digestor of sewage treatment plant and predicting production amount of digestion gas and method therefor
WO2014157750A1 (en) Apparatus and method for providing causative factors for state of quality of effluent water from sewage treatment plant
CN204454747U (en) The miniature artificial swamp drop purification system of automatic monitoring water quality
CN110436609B (en) Intelligent sewage treatment aeration control method with self-learning function
CN105785951B (en) The processing automatic operating system of sewage is realized using statistical modeling technology
CN107344733A (en) One kind on-line monitoring protecting against shock system
CN114790039A (en) Intelligent denitrification regulation and control method and system for aquaculture wastewater
CN114149076B (en) Intelligent debugging system of anaerobic ammonia oxidation sewage treatment system
KR20230038038A (en) Energy efficiency evaluation method and energy saving decision-making system of sewage treatment plant based on water treatment energy index
KR20180076454A (en) Method of controlling electric energy in wastewater treatment operation for energy saving

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 13879981

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 13879981

Country of ref document: EP

Kind code of ref document: A1