JPH02280892A - Chlorine injection amount inference apparatus of water purifying plant - Google Patents

Chlorine injection amount inference apparatus of water purifying plant

Info

Publication number
JPH02280892A
JPH02280892A JP1103164A JP10316489A JPH02280892A JP H02280892 A JPH02280892 A JP H02280892A JP 1103164 A JP1103164 A JP 1103164A JP 10316489 A JP10316489 A JP 10316489A JP H02280892 A JPH02280892 A JP H02280892A
Authority
JP
Japan
Prior art keywords
amount
chlorine
correction
activated carbon
water
Prior art date
Legal status (The legal status 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 status listed.)
Pending
Application number
JP1103164A
Other languages
Japanese (ja)
Inventor
Tetsuo Kosuda
小須田 徹夫
Hiroshi Tsukura
津倉 洋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Meidensha Corp
Meidensha Electric Manufacturing Co Ltd
Original Assignee
Meidensha Corp
Meidensha Electric Manufacturing Co Ltd
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 Meidensha Corp, Meidensha Electric Manufacturing Co Ltd filed Critical Meidensha Corp
Priority to JP1103164A priority Critical patent/JPH02280892A/en
Publication of JPH02280892A publication Critical patent/JPH02280892A/en
Pending legal-status Critical Current

Links

Landscapes

  • Devices For Executing Special Programs (AREA)
  • Feedback Control In General (AREA)
  • Treatment Of Water By Oxidation Or Reduction (AREA)

Abstract

PURPOSE:To automatically operate a chlorine injection amount and to reduce the load of an operator by adding correction quantity U calculated by fuzzy inference based on the stagnation time or temp. of inflow water, a measured value of a chlorine demand and a consumption amount due to activated carbon. CONSTITUTION:The chrine demand U1 measured by a chlorine demand meter, a consumption amount U2 due to activated carbon and correction quantity U are added to set a chlorine injection amount inference value. The consumption amount U2 is set to a value calculated by multiplying an activated carbon inflow amount YAC by 0.25 and this operation is performed in an input data processing part 3. The correction quantity U is set to a value calculated by adding U1, U2, U3 inferred in a fuzzy inference part 4 and U1 is the first correction quantity based on the stagnation time T of inflow water in a sedimentation basin and quantity of solar radiation, U2 is the second correction quantity based on water temp. and U3 is the third correction value based on the residual chlorine amount at the outlet of the sedimentation basin and a change portion thereof.

Description

【発明の詳細な説明】 A、産業上の利用分野 本発明は浄水場の塩素注入量を推論する装置に関するも
のである。
DETAILED DESCRIPTION OF THE INVENTION A. Field of Industrial Application The present invention relates to a device for estimating the amount of chlorine injection in a water treatment plant.

B0発明の概要 本発明は浄水場の塩素注入量を推論するにあたって、塩
素要求量Ulと活性炭による消費量U、と補正量ΔUと
の加算値を推論値とする方法において、 日射量と流入水の滞留時間とにもとずいてΔUlを、水
温にもとずいてΔU、を、残存塩素量にもとずいてΔU
3を夫々ファジィ推論により求めて、これらの加算値を
ΔUとすることによって、推論の自動化及び操業ノウハ
ウからのモデル化を容易にしたものである。
B0 Summary of the Invention The present invention provides a method in which the added value of the chlorine demand Ul, the consumption amount U by activated carbon, and the correction amount ΔU is used as an inference value when inferring the amount of chlorine to be injected into a water treatment plant. ΔUl is based on the residence time of water, ΔU is based on the water temperature, and ΔU is based on the amount of residual chlorine.
3 by fuzzy inference, and the sum of these values is set as ΔU, thereby facilitating automation of inference and modeling based on operational know-how.

C1従来の技術 浄水場においては、第8図に示すように原水が着水井A
、に貯水され、次いで沈澱池Atで固液分離された後濾
過池A3、浄水池A4を通って配水される。この場合原
水中の有機物の分解、臭気や色の除去、鉄やマンガン等
の重金属の除去及び藻類の除去を目的として塩素が注入
される。この注入は例えば沈澱池の流入側、流出側、及
び濾過池の流出側で行われ、夫々前塩素注入、中塩素注
入及び後塩素注入と呼ばれている。塩素を注入するにつ
いては、水質変動、気象変動による塩素飛散量の変化、
更に薬品注入制御に共通する無駄時間が多いことから自
動化が困難であった。従って現在は多くの浄水場でベテ
ラン運転員のカンと経験により手動注入量(率)設定が
行われている。
C1 Conventional technology In a water treatment plant, raw water flows through the landing well A as shown in Figure 8.
, and then separated into solid and liquid in a sedimentation tank At, and then distributed through a filtration tank A3 and a water purification tank A4. In this case, chlorine is injected for the purpose of decomposing organic matter in the raw water, removing odor and color, removing heavy metals such as iron and manganese, and removing algae. This injection is carried out, for example, on the inlet side and outlet side of the sedimentation basin, and on the outlet side of the filtration basin, and is called pre-chlorine injection, intermediate chlorine injection and post-chlorine injection, respectively. When injecting chlorine, consider changes in water quality, changes in the amount of chlorine scattered due to weather changes,
Furthermore, automation is difficult because there is a lot of wasted time common to chemical injection control. Therefore, in many water treatment plants, the injection amount (rate) is now set manually based on the knowledge and experience of veteran operators.

こうした手動注入を行う場合、例えば前塩素の注入量は
、原水水質による塩素要求量(U、)と粉末活性炭によ
る消費量1t)と補正量(ΔU)とを加算したUI+U
t+ΔUの値として決定される。
When performing such manual injection, for example, the pre-chlorine injection amount is UI + U, which is the sum of the chlorine demand (U,) depending on the raw water quality, the consumption (1 t) of powdered activated carbon, and the correction amount (ΔU).
It is determined as the value of t+ΔU.

ここで粉末活性炭は、脱臭を行うために例えば20〜5
0mg/12注入され、活性炭1mgに対して塩素が0
.25mg/(2程度消費されることから、Utは活性
炭の注入量に対して0.25を乗じた値となる。また補
正量ΔUは日射量、沈澱池の滞留時間、水温及び藻類の
発生状況等を考慮して決定され、これら因子とΔUとの
関係は■〜■のようになる。
Here, powdered activated carbon is used for deodorizing, for example, 20 to 5
0mg/12 injected, 0 chlorine per 1mg of activated carbon
.. Since approximately 25mg/(2) is consumed, Ut is the value obtained by multiplying the amount of activated carbon injected by 0.25.The correction amount ΔU is based on the amount of solar radiation, residence time in the sedimentation tank, water temperature, and algae occurrence status. The relationship between these factors and ΔU is as shown in (■) to (■).

■日射量が多いときには塩素の飛散量が多いことがらΔ
Uを大きくする。
■When the amount of solar radiation is high, the amount of chlorine scattered is large, so Δ
Increase U.

■沈澱池の滞留時間が長いときにはΔUを大きくする。■If the residence time in the sedimentation tank is long, increase ΔU.

■水温が高いと反応速度が大きいためΔUを大きくする
■If the water temperature is high, the reaction rate is high, so increase ΔU.

■藻類が発生している場合にはΔUを大きくする。■If algae are growing, increase ΔU.

D0発明が解決しようとする課題 上述のように運転員のカンと経験により塩素注入量を決
定する方法は、 (1)1日に数千回も設定変更を行うため運転員の負担
が大きい。
Problems to be Solved by the D0 Invention As mentioned above, the method of determining the amount of chlorine injection based on the operator's intuition and experience is: (1) Settings must be changed several thousand times a day, which places a heavy burden on the operator.

(2)操業ノウハウからのモデル化が困難であり、制御
アルゴリズムが構築できない。
(2) Modeling based on operational know-how is difficult, and control algorithms cannot be constructed.

などの欠点があった。There were drawbacks such as.

本発明の目的はこうした欠点を解消することにある。The object of the present invention is to eliminate these drawbacks.

61課題を解決するための手段及び作用本発明は原水の
水質にもとずいて計測した塩素要求量と活性炭による塩
素消費量と補正量との加算値を塩素注入量の推論値とし
て求める装置において、 日射量、塩素により処理される槽の流入水滞留時間、水
温、前記槽の残存塩素量及び当該残存塩素量の変化分と
第1〜第3の補正量ΔU8、ΔU2、ΔU3とについて
与えられたメンバシップ関数と、日射量及び流入水滞留
時間の各々について選ばれたメンバシップ関数の組み合
わせを条件部とし、第1の補正量について選ばれたメン
バシップ関数を結論部とする第1のルール群と、水温に
ついて選ばれたメンバシップ関数を条件部とし、第2の
補正量について選ばれたメンバシップ関数を結論部とす
る第2のルール群と、残存塩素量及びその変化分につい
て選ばれたメンバシップ関数の組み合わせを条件部とし
、第3の補正量について選ばれたメンバシップ関数を結
論部とする第3のルール群とを備えた知識データベース
を用意し、第1〜第3のルール群にもとすいてファジィ
推論を実行して夫々第1〜第3の補正量ΔU2、ΔUt
%ΔU、を求めるファジィ推論機構と、これら補正量Δ
U1%ΔU2、ΔU3と前記塩素要求量と前記塩素消費
量とを加算する手段とを設けたことを特徴とする。
61 Means and Effects for Solving the Problems The present invention provides an apparatus for determining the added value of the chlorine demand amount measured based on the quality of raw water, the chlorine consumption amount by activated carbon, and the correction amount as an inferred value of the chlorine injection amount. , given for the amount of solar radiation, the residence time of inflow water in the tank treated with chlorine, the water temperature, the amount of residual chlorine in the tank, the change in the amount of residual chlorine, and the first to third correction amounts ΔU8, ΔU2, ΔU3. A first rule in which the condition part is a combination of the membership function selected for each of the solar radiation amount and the inflow water residence time, and the conclusion part is the membership function selected for the first correction amount. a second rule group whose condition part is the membership function selected for the water temperature and whose conclusion part is the membership function selected about the second correction amount; A knowledge database is prepared that includes a third rule group whose condition part is a combination of membership functions selected for the third correction amount and whose conclusion part is a membership function selected for the third correction amount. The first to third correction amounts ΔU2 and ΔUt are calculated by performing fuzzy inference on each group.
A fuzzy inference mechanism that calculates %ΔU and these correction amounts Δ
The present invention is characterized by providing means for adding U1% ΔU2, ΔU3, the chlorine demand amount, and the chlorine consumption amount.

F、実施例 第1図においてIは前塩素注入制御装置であり、原水水
質にもとずいて塩素要求量を計測する塩素要求量計、活
性炭注入装置、原水流入量を測定する流入量測定部、日
射量計、水温計及び残存塩素を測定する残存塩素測定計
等のプロセスデータを取り込むための手段と、前塩素注
入量の推定値に応じた量だけ前塩素を沈澱池の流入側に
注入する手段とを有している。2はプロセスデータ人力
部、3は入力データ処理部であり、塩素要求量や活性炭
注入装置よりの活性炭注入量等のプロセスデータをファ
ジィ推論部4で用いる言語に変換する処理等を行う。フ
ァジィ推論部4は、メンバシップ関数や推論ルールを格
納した知識データベース4、と、プロセスデータ及び知
識データベース4.内の知識を用いて推論を行うファジ
ィ推論機構4゜とを有している。5はマンマシンインタ
ーフェイスであり、推論規則の変更やファジィ推論値及
びプロセスデータの表示等を行う。6は推論結果出力部
である。
F. Embodiment In Fig. 1, I is a pre-chlorine injection control device, which includes a chlorine demand meter that measures the chlorine demand based on the quality of raw water, an activated carbon injection device, and an inflow measuring section that measures the inflow of raw water. , a means for capturing process data such as a solar radiation meter, a water temperature meter, and a residual chlorine measuring meter for measuring residual chlorine, and injecting pre-chlorine into the inlet side of the settling tank in an amount corresponding to the estimated amount of pre-chlorine injection. and the means to do so. Reference numeral 2 denotes a process data manual section, and 3 an input data processing section, which performs processes such as converting process data such as the required amount of chlorine and the amount of activated carbon injected from an activated carbon injection device into a language used by the fuzzy inference section 4. The fuzzy inference unit 4 includes a knowledge database 4 storing membership functions and inference rules, and a process data and knowledge database 4. It has a fuzzy inference mechanism 4° that performs inference using knowledge within the system. Reference numeral 5 denotes a man-machine interface for changing inference rules and displaying fuzzy inference values and process data. 6 is an inference result output unit.

第2図はハードウェアの構成例を示す図であり、プロセ
スデータ入力部2及び入力データ処理部3をプロセスコ
ントローラ100により構成すると共に、ファジィ推論
部4をファジィコントローラ200により構成し、マン
マシンインターフェイスを工業用パーソナルコンピュー
タ300及びCRT等により構成している。
FIG. 2 is a diagram showing an example of the hardware configuration, in which the process data input unit 2 and input data processing unit 3 are configured by a process controller 100, the fuzzy inference unit 4 is configured by a fuzzy controller 200, and a man-machine interface is provided. It is composed of an industrial personal computer 300, a CRT, and the like.

次に前塩素流入量の推論のフローを第3図を参照しなが
ら述べる。塩素要求量計で計測した塩素要求ff1u、
と活性炭による消費量U、と補正量ΔUとを加算して前
塩素注入量推論値とする。U、は活性炭流入量Y AC
に0.25を乗じた値とし、この演算は入力データ処理
部3で行われる。ΔUはファジィ推論部4で推論したΔ
U IsΔU2、ΔU3を加算した値とし、ΔUIは沈
澱池における流入水の滞留時間Tと日射量とにもとずく
第1の補正量、ΔU2は水温による第2の補正量、ΔU
3は沈澱池出口の残存塩素量及びその変化分にもとずく
第3の補正量である。前記滞留時間Tは、原水流入量を
Q、沈澱池容積をQoとするとQ。/Qとして与えられ
る。
Next, the flow for inferring the amount of pre-chlorine inflow will be described with reference to FIG. Chlorine demand ff1u measured with a chlorine demand meter,
, the consumption amount U due to activated carbon, and the correction amount ΔU are added to obtain the pre-chlorine injection amount inference value. U, is activated carbon inflow amount Y AC
is multiplied by 0.25, and this calculation is performed in the input data processing section 3. ΔU is Δ inferred by fuzzy inference unit 4
U is the sum of ΔU2 and ΔU3, ΔUI is the first correction amount based on the residence time T of the inflow water in the sedimentation basin and the amount of solar radiation, ΔU2 is the second correction amount based on the water temperature, ΔU
3 is a third correction amount based on the amount of residual chlorine at the outlet of the sedimentation tank and its change. The residence time T is Q, where Q is the inflow of raw water and Qo is the sedimentation tank volume. /Q.

以下にΔU1、ΔUt%ΔU3のファジィ推論について
述べる。説明の便宜上、日射量をS v、滞留時間をT
1原水水温をWT、残存塩素量をReC12、残存塩素
量の変化分をdReCQと定義する。
Fuzzy inference of ΔU1 and ΔUt%ΔU3 will be described below. For convenience of explanation, the amount of solar radiation is S v and the residence time is T
1 Define raw water temperature as WT, residual chlorine amount as ReC12, and change in residual chlorine amount as dReCQ.

Sw、Tについては各々第4図(a)、(b)に示すよ
うに3段階のメンバシップ関数を規定し、そのファジィ
ラベル(メンバシップ関数名)をLlM、Sとする。こ
のり、M、Sの意味は次の通りである。
For Sw and T, three-stage membership functions are defined as shown in FIGS. 4(a) and 4(b), respectively, and their fuzzy labels (membership function names) are LIM, S. The meanings of this, M, and S are as follows.

L:多い(長い) M:中くらい S:少ない(短い) Sv、Tにもとずいて求められるΔUlについては第4
図(C)に示すように5段階のメンバシップ関数を規定
し、そのファジィラベルをSS、MS、MD、ML、L
Lとする。これらファジィラベルの意味は次の通りであ
る。
L: Many (long) M: Medium S: Few (short) Regarding ΔUl, which is calculated based on Sv and T, see the fourth section.
As shown in Figure (C), a five-stage membership function is defined, and its fuzzy labels are SS, MS, MD, ML, and L.
Let it be L. The meanings of these fuzzy labels are as follows.

LL:かなり大きい(長い) ML二大きい(長い) MD:中くらい MS:小さい(短い) SS:かなり小さい(短い) WT、ΔU、については各々第5図(a)、(b)に示
すように5段階のメンバシップ関数を規定し、そのファ
ジィラベルをΔUIと同様に定める。
LL: quite large (long) ML2 large (long) MD: medium MS: small (short) SS: quite small (short) WT and ΔU are as shown in Figures 5 (a) and (b), respectively. A five-stage membership function is defined for , and its fuzzy label is defined in the same way as ΔUI.

ReCQ、dRecQについては各々第6図(a)、(
b)に示すように3段階のメンバシップ関数を規定し、
ReCQのファジィラベルはSw、Tと同様であり、d
ReCQのファジィラベルはN、ZE、Pとする。この
N、ZE、Pの意味は次の通りである。
For ReCQ and dRecQ, Fig. 6(a) and (
Define a three-stage membership function as shown in b),
The fuzzy label of ReCQ is the same as Sw, T, and d
Let the fuzzy labels of ReCQ be N, ZE, and P. The meanings of N, ZE, and P are as follows.

N:負である ZE:変化しない P:正である R e CQ 、 d Re CQにもつずいて求めら
れるΔU3については第6図(C)に示すように5段階
のメンバシップ関数を規定し、そのファジィラベルをP
B、PS、ZE、NS、NBとする。これらファジィラ
ベルの意味は次の通りである。
N: Negative ZE: No change P: Positive Regarding ΔU3, which is subsequently obtained for Re CQ and d Re CQ, a five-stage membership function is defined as shown in FIG. 6(C), P the fuzzy label
B, PS, ZE, NS, and NB. The meanings of these fuzzy labels are as follows.

PB:急上昇または正で大きい PS:上昇または正で小さい ZE:変化しないまたはほとんどゼロ NS:低下または負で小さい NB:急低下または負で大きい また推論ルールとしては、次に記述するようにΔUI〜
ΔU、を夫々求めるための第1のルール群(ルール1〜
9)、第2のルール群(ルールlO〜14)及び第3の
ルール群(ルール15〜23)が用いられる。
PB: Rapid increase or positive and large PS: Increase or positive and small ZE: No change or almost zero NS: Decrease or negative and small NB: Sudden decrease or negative large
The first set of rules (Rule 1 to
9), the second rule group (rules 10 to 14) and the third rule group (rules 15 to 23) are used.

(第1のルール群) 1 ifSw=L and T=L thenΔU、=
LL2 if Sw=L and T=M thenΔ
U、=ML3  if  Sw=L  and  T=
S4  if  Sw=M  and  T=L5  
if  Sw=M  and  T=M6  ir S
w=M  and  T=S7  if  S w= 
S  and  T = L8  if  Sw=S 
 and  T=M9  rr  S w= S  a
nd  T = S(第2のルール群) 10 if WT = L L  thenll if
 WT =ML  then12 if WT=MD 
 then 13irWT=Ms  then 14 if WT = S S  then(第3のル
ール群) 15 if ReCQ=L and dRecf2he
n then then then then then then ΔU、=MD ΔU、=ML ΔU、=MD ΔU、=MS ΔU、=MD ΔU、=MS ΔU、=SS Δut=t、t。
(First rule group) 1 ifSw=L and T=L thenΔU,=
LL2 if Sw=L and T=M thenΔ
U,=ML3 if Sw=L and T=
S4 if Sw=M and T=L5
if Sw=M and T=M6 ir S
w=M and T=S7 if S w=
S and T = L8 if Sw=S
and T=M9 rr S w= S a
nd T = S (second rule group) 10 if WT = L L thenll if
WT=ML then12 if WT=MD
then 13irWT=Ms then 14 if WT=S S then (third rule group) 15 if ReCQ=L and dRecf2he
n then then then then then then then ΔU,=MD ΔU,=ML ΔU,=MD ΔU,=MS ΔU,=MD ΔU,=MS ΔU,=SS Δut=t,t.

ΔUt=ML ΔUt=MD ΔU、二MS ΔU、=SS =P  then  ΔU3=NB 16  if ReC12=L and dRec& 
=ZE ther+ ΔU、=NS17 if ReC
& ;L and dRelJ =N  then A
U、=ZE18 if ReCff =M and d
Re(J! −P  then ΔU、=NS19 i
f ReCσ=M and dReCQ =ZE th
en AUs=ZE20 if ReCI2 =M a
nd dRe(44then AU、−PS21  i
f ReC12=S and dReCQ =P  t
hen ΔU、−ZE22 1fReC(2=S an
d dRec12 =ZE then ΔU3=PS2
3 if ReCff =S and dRe(44t
hen AU3=PBこれらルールのifの部分は条件
部、thenの部分は結論部であり、その具体的意味は
、例えばルール1の場合「SlがL(多い)、かっTが
L (長い)ならばΔU、はLL(かなり大きい)であ
る。」ということであり、こうしたルール群を用いた推
論法としてはマムダ二法を適用する。
ΔUt=ML ΔUt=MD ΔU, 2MS ΔU,=SS=P then ΔU3=NB 16 if ReC12=L and dRec&
=ZE ther+ ΔU, =NS17 if ReC
&; L and dRelJ = N then A
U,=ZE18 if ReCff=M and d
Re(J! −P then ΔU, = NS19 i
fReCσ=M and dReCQ=ZE th
en AUs=ZE20 if ReCI2=M a
nd dRe(44then AU, -PS21 i
fReC12=S and dReCQ=Pt
hen ΔU, -ZE22 1fReC (2=S an
d dRec12 =ZE then ΔU3=PS2
3 if ReCff =S and dRe(44t
hen AU3=PB The if part of these rules is the condition part, and the then part is the conclusion part. For example, in the case of rule 1, "If Sl is L (many) and Kak T is L (long), then ΔU is LL (considerably large).'' Therefore, the Mamdani method is applied as an inference method using such a rule group.

従って上記のルールを適用すると、変数S。によりI7
におけるメンバシップ値を求めると共に変数TによりL
におけるメンバシップ値を求め、それら値の小さい方の
値で結論部であるLLのメンバシップ関数をカットする
ことになる。そしてこのようにカットしたメンバシップ
関数を各ルール毎に求め、これらを合成する。合成にあ
たっては、各メンパンツブ関数の最大値の部分を選択す
る。
Therefore, applying the above rule, the variable S. By I7
Find the membership value in L by the variable T.
The membership function of LL, which is the conclusion part, is cut using the smaller of these values. Then, the membership functions cut in this way are obtained for each rule, and these are synthesized. For synthesis, select the part with the maximum value of each member subfunction.

そしてその金成分の重心を求めてその値を夫々ΔU11
ΔU2、AU3とする。
Then, find the center of gravity of the gold component and calculate its value as ΔU11.
Let them be ΔU2 and AU3.

推論例として、日射量0 、7 K W H/ m ”
、滞留時間3hr、水温20℃、出口残存塩素濃度02
mg/Q及び残存塩素濃度の変化量を+0.1mg/ρ
10.5hrの場合を挙げると、ΔU11ΔtJ 、、
AU3は夫々第7図(a)〜(c)のようにして求まる
。この結果ΔtJ+= I 、42mg/Q 。
As an example of inference, solar radiation is 0, 7 K W H/m”
, residence time 3hr, water temperature 20℃, outlet residual chlorine concentration 02
mg/Q and the amount of change in residual chlorine concentration +0.1mg/ρ
In the case of 10.5 hr, ΔU11ΔtJ,,
AU3 is determined as shown in FIGS. 7(a) to 7(c), respectively. As a result, ΔtJ+=I, 42 mg/Q.

ΔU t = 0 、70 m g / Q 、  Δ
U3= 0.75 mg/eとなりΔUはこれらの合計
値2.87mg/(!となる。このとき塩素要求量計の
指示値が5.Om g / 12であり、活性炭の注入
を行っていないときには全体の推論結果は U + ” U t÷ΔU=5.0+0+2.87=7
.87mr、IQとなる。
ΔU t = 0, 70 mg/Q, Δ
U3 = 0.75 mg/e, and ΔU is the total value of these 2.87 mg/(! Sometimes the overall inference result is U + ” U t ÷ ΔU = 5.0 + 0 + 2.87 = 7
.. 87 mr, IQ.

なお本発明は前塩素注入量の推論のみでなく、中塩素注
入量の推論についても適用することができる。
Note that the present invention can be applied not only to the inference of the pre-chlorine injection amount but also to the inference of the intermediate chlorine injection amount.

また日射量、水温については、季節、天候、時刻等の手
入力データにもとずいて求める方法もあるが、実施例の
ようにセンサーにより直接計測したプロセスデータを使
用すれば、操作員の負担を軽減できると共に、手入力も
れによる推論ミスを防止できる。
There is also a method to obtain solar radiation and water temperature based on manually input data such as season, weather, time, etc., but if process data directly measured by a sensor is used as in the example, the burden on the operator will be reduced. It is possible to reduce errors and prevent inference errors due to omissions in manual input.

G1発明の効果 本発明によれば、流入水の滞留時間や水温等にもとすい
てファジィ推論を行って補正量ΔUを求め、このΔUと
塩素要求量の計測値と活性炭による消費量とを加算して
いるため、次のような効果がある。
G1 Effect of the Invention According to the present invention, the correction amount ΔU is obtained by performing fuzzy inference based on the residence time of inflow water, water temperature, etc., and this ΔU, the measured value of the chlorine demand amount, and the consumption amount by activated carbon are calculated. The addition has the following effects.

(1)塩素注入率を自動で演算でき、運転員の負担が軽
減される。
(1) The chlorine injection rate can be calculated automatically, reducing the burden on the operator.

(2)ベテラン運転員の知識と経験をルールとメンバー
シップ関数の形で表現することができ、ノウハウの共用
化が図れる。
(2) The knowledge and experience of veteran operators can be expressed in the form of rules and membership functions, making it possible to share know-how.

(3)操業ノウハウのモデル化が容易であり、制御アル
ゴリズムを構築することができ、ルールと制御アルゴリ
ズムの改良を行うことができる。
(3) Operation know-how can be easily modeled, control algorithms can be constructed, and rules and control algorithms can be improved.

(4)マンマシン装置を組み合わせれば、推論結果を視
覚等により確認できる。
(4) If man-machine devices are combined, the inference results can be confirmed visually.

【図面の簡単な説明】[Brief explanation of drawings]

第1図は本発明の実施例のシステムを示す構成図、第2
図は同実施例のハードウェアを示す構成図、第3図は演
算のフローを示す説明図、第4図〜第6図は各メンバシ
ップ関数を示すグラフ、第7図は推論演算を示す説明図
、第8図は浄水場の処理フローを示す説明図である。 ■・・・前塩素注入制御装置、2・・・プロセスデータ
入力部、3・・・人力データ処理部、4・・・ファジィ
推論部、5・・・マンマシンインターフェイス、6・・
・推論結果出力処理部。 メンバーシップ関数の改良変更を行うことにより、実施
例のハードウェア構成図 (C) (b) 敗 ReC1(mg/l ) ReC1s変化率 (rng/l/QSH) ΔU:+(mg/ Z)
Fig. 1 is a configuration diagram showing a system according to an embodiment of the present invention;
The figure is a configuration diagram showing the hardware of the same embodiment, Fig. 3 is an explanatory diagram showing the flow of calculations, Figs. 4 to 6 are graphs showing each membership function, and Fig. 7 is an explanation showing inference calculation. 8 are explanatory diagrams showing the processing flow of the water purification plant. ■... Pre-chlorine injection control device, 2... Process data input unit, 3... Human data processing unit, 4... Fuzzy reasoning unit, 5... Man-machine interface, 6...
・Inference result output processing unit. By improving and changing the membership function, the hardware configuration diagram of the embodiment (C) (b) Loss ReC1 (mg/l) ReC1s change rate (rng/l/QSH) ΔU: + (mg/Z)

Claims (1)

【特許請求の範囲】[Claims] (1)原水の水質にもとずいて計測した塩素要求量と活
性炭による塩素消費量と補正量との加算値を塩素注入量
の推論値として求める装置において、日射量、塩素によ
り処理される槽の流入水滞留時間、水温、前記槽の残存
塩素量及び当該残存塩素量の変化分と第1〜第3の補正
量ΔU_1、ΔU_2、ΔU_3とについて与えられた
メンバシップ関数と、日射量及び流入水滞留時間の各々
について選ばれたメンバシップ関数の組み合わせを条件
部とし、第1の補正量について選ばれたメンバシップ関
数を結論部とする第1のルール群と、水温について選ば
れたメンバシップ関数を条件部とし、第2の補正量につ
いて選ばれたメンバシップ関数を結論部とする第2のル
ール群と、残存塩素量及びその変化分について選ばれた
メンバシップ関数の組み合わせを条件部とし、第3の補
正量について選ばれたメンバシップ関数を結論部とする
第3のルール群とを備えた知識データベースを用意し、
第1〜第3のルール群にもとずいてファジィ推論を実行
して夫々第1〜第3の補正量ΔU_1ΔU_2、ΔU_
3を求めるファジィ推論機構と、これら補正量ΔU_1
、ΔU_2、ΔU_3と前記塩素要求量と前記塩素消費
量とを加算する手段とを設けたことを特徴とする浄水場
の塩素注入量推論装置。
(1) In a device that calculates the added value of chlorine demand measured based on the quality of raw water, chlorine consumption and correction amount by activated carbon as an inferred value for chlorine injection amount, the amount of solar radiation and the tank treated with chlorine are used. Membership functions given for the inflow water residence time, water temperature, the amount of residual chlorine in the tank, the change in the amount of residual chlorine, and the first to third correction amounts ΔU_1, ΔU_2, ΔU_3, the amount of solar radiation, and the inflow. A first rule group whose condition part is a combination of membership functions selected for each water residence time, and whose conclusion part is a membership function selected for a first correction amount, and a membership function selected for water temperature. A second rule group having a function as a condition part and a membership function selected for the second correction amount as a conclusion part, and a combination of membership functions selected for the amount of residual chlorine and its change as a condition part. , a third rule group having a membership function selected for the third correction amount as a conclusion part;
Fuzzy inference is executed based on the first to third rule groups to obtain the first to third correction amounts ΔU_1ΔU_2, ΔU_, respectively.
Fuzzy inference mechanism for calculating 3 and these correction amounts ΔU_1
, ΔU_2, ΔU_3, the chlorine demand amount, and the chlorine consumption amount.
JP1103164A 1989-04-22 1989-04-22 Chlorine injection amount inference apparatus of water purifying plant Pending JPH02280892A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP1103164A JPH02280892A (en) 1989-04-22 1989-04-22 Chlorine injection amount inference apparatus of water purifying plant

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP1103164A JPH02280892A (en) 1989-04-22 1989-04-22 Chlorine injection amount inference apparatus of water purifying plant

Publications (1)

Publication Number Publication Date
JPH02280892A true JPH02280892A (en) 1990-11-16

Family

ID=14346870

Family Applications (1)

Application Number Title Priority Date Filing Date
JP1103164A Pending JPH02280892A (en) 1989-04-22 1989-04-22 Chlorine injection amount inference apparatus of water purifying plant

Country Status (1)

Country Link
JP (1) JPH02280892A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107091534A (en) * 2017-04-27 2017-08-25 扬州大学 A kind of solar water heater fuzzy control device

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107091534A (en) * 2017-04-27 2017-08-25 扬州大学 A kind of solar water heater fuzzy control device

Similar Documents

Publication Publication Date Title
DE69032557T2 (en) METHOD AND SYSTEM FOR CONTROLLING A METHOD
US6808630B2 (en) System and method for AI controlling waste-water treatment by neural network and back-propagation algorithm
JP4468269B2 (en) Process monitoring apparatus and method
CN107908111A (en) A kind of computer control method of the sludge dewatering system based on BP neural network
Piotrowski et al. Mixed integer nonlinear optimization of biological processes in wastewater sequencing batch reactor
Tzoneva Optimal PID control of the dissolved oxygen concentration in the wastewater treatment plant
JPH02280892A (en) Chlorine injection amount inference apparatus of water purifying plant
WO2021210654A1 (en) Process monitoring device, process monitoring method, and program
DE102004025876A1 (en) Device and method for stack property estimation
Han et al. Efficient economic model predictive control of water treatment process with learning-based Koopman operator
Galan et al. Optimization strategies for the design and synthesis of distributed wastewater treatment networks
JPH1157780A (en) Sewage disposal plant, its measuring instrument and support equipment
JP3603182B2 (en) Wastewater treatment process simulator
WO2013136503A1 (en) System for operating water treatment plant and method for planning amount of water supply
Onat et al. Fuzzy plus integral control of the effluent turbidity in direct filtration
DE4140915A1 (en) Control system for sewage treatment plant with single-stage activation tank - using fuzzy logic computer to derive regulating value, e.g. for rate of oxygen addn. from inexact measured and input data
JP4088075B2 (en) Water treatment process hybrid water quality measuring device and water treatment system having the same
Zondervan et al. Dynamic optimization of chemical cleaning in dead-end ultra filtration
JPH06335694A (en) Controller of device for biologically treating waste water
Algoufily Wastewater Treatment Plant Optimization: Development of Membrane Bioreactor Fouling Monitoring Tool and Prediction of Transmembrane Pressure Using Artificial Neural Networks
JPS61263688A (en) Device for controlling use of water in water purification plant
Ferràndez et al. DEVELOPMENT OF FUZZY LOGIC SYSTEM FOR CONTROLLING OZONE DOSING RATE IN A DRINKING WATER TREATMENT PLANT
JPH078711A (en) Filtration pond cleaning time controlling apparatus
Bae et al. Knowledge-based control and case-based diagnosis based upon empirical knowledge and fuzzy logic for the SBR plant
Kumar et al. A Robust Fuzzy Logic Control of Two Tanks Liquid Level Process