CN101993151B - Loop control method for biochemical sewage treatment process - Google Patents

Abstract

The invention relates to a loop control method for the biochemical sewage treatment process, which comprises the following steps of: performing cascade control on the organic load of inflowing water, and controlling 'labor force' of the biochemical sewage treatment, namely the number of 'foods' of microbes so as to reduce the energy consumption in a sewage lift link; performing cascade control on dissolved oxygen concentration in an aeration tank to realize air supply of the aeration tank according to the need so as to reduce the energy consumption in a fan aeration link; and performing feedforward-cascade control on the sludge concentration in the aeration tank, and controlling the number of microbes in a sewage treatment system so as to reduce the energy consumption in a sludge return link. The loop control method effectively controls the biochemical sewage treatment process with three main energy consumption links, namely sewage lift, fan aeration and sludge return, obviously reduces the energy consumption in the biochemical sewage treatment process on the premise of guaranteeing that effluent steadily meets the standard, and solves the problem that the traditional PID single loop control is difficult to achieve ideal control effect by implementing loop control cascade control and a human intelligent control algorithm,.

Description

Biochemical processing procedure of sewage circuit controls method

Technical field

The present invention relates to a kind of sewage disposal process control technology, a kind of specifically biochemical processing procedure of sewage circuit controls method.

Background technology

Effluent quality is unstable, energy consumption is too high is the bottleneck factor that restricts the wastewater treatment in China development at present.Can guarantee effectively to reduce the sewage disposal system energy consumption under the stable water outlet prerequisite up to standard through control device.Good according to statistics process control can be saved 6% wastewater treatment operating cost.

A/O (anoxic/oxic, anoxic/aerobic) technology was developed in early 1980s, was a kind of sewage treatment process that present municipal sewage plant extensively adopts, and its technological process is as shown in Figure 1.A/O technology is made up of anoxic pond, aeration tank and three unit of second pond, wherein mainly passes through biochemical reaction in anoxic pond and unit, aeration tank, utilizes the microbiological oxidation of ARTIFICIAL CULTURE to decompose the organism in the sewage; , remove the solid particle in the sewage, thereby sewage is purified through the further Separation of Solid and Liquid of physical sedimentation effect in the second pond unit.

The sewage disposal plant effluent water-quality guideline mainly comprises chemical oxygen demand COD (Chemical OxygenDemanded) and ammonia nitrogen; Usually as long as the ammonia nitrogen index is up to standard; Then water outlet COD just can be up to standard; Therefore the process control of many wastewater treatments all with ammonia nitrogen concentration as the effluent quality index, but in actual application, COD on-line measurement instrument is used more extensive than ammonia nitrogen on-line monitoring appearance.

In the prior art, be the control of single object or single link to the municipal sewage treatment process, controlling level is very low, adopts manually control or simple single loop control mostly, and the control effect is relatively poor, and energy consumption is higher.Be controlled to be example with dissolved oxygen DO, dissolved oxygen DO control is to use control loop the most widely, but current dissolved oxygen DO setting value mostly is static, can not make real-time adjustment to environmental factor and system change, can't realize " air feed as required ".When the dissolved oxygen DO setting value was too small, the dissolved oxygen concentration of aeration tank was low excessively, influenced effluent quality and treatment effeciency; When the dissolved oxygen DO setting value is excessive, cause part of oxygen directly to escape into the air from sewage, caused unnecessary energy dissipation.

The main energy consumption equipment of the dissolved oxygen DO sewage disposal process of domestic sewage treatment plant comprises lift pump, fan blower and return sludge pump; Respectively corresponding sewage lifting, blower fan aeration and three links of mud backflow, these three power consumption links account for more than 80% of full factory of sewage treatment plant power consumption.

Summary of the invention

To the above-mentioned weak point that exists in the prior art; The technical matters that the present invention will solve provides a kind of biochemical processing procedure of sewage circuit controls methods towards sewage lifting, blower fan aeration and three main power consumption links of mud backflow; This method is guaranteeing to reduce the biochemical processing procedure of sewage energy consumption under the stable water outlet prerequisite up to standard.

For solving the problems of the technologies described above, the technical scheme that the present invention adopts is:

A kind of biochemical processing procedure of sewage circuit controls of the present invention method may further comprise the steps:

The entry organic loading is carried out tandem control, and " labour " of the biochemical treatment of control sewage is " food " quantity of microorganism, to reduce the energy consumption that sewage promotes link;

Dissolved oxygen concentration in the aeration tank is implemented tandem control, realize the air feed as required of aeration tank, to reduce the energy consumption of blower fan aeration link;

Sludge concentration in the aeration tank is implemented the control of feedforward-tandem, and the micro organism quantity in the control sewage disposal system is to reduce the energy consumption of mud backflow link.

Saidly the entry organic loading is carried out tandem control comprise external loop discharge setting value control loop and inner looping discharge control loop, wherein:

Discharge setting value control loop: according to the variation of entry water quality and effluent quality, the feedback through water outlet COD realizes the control of entry organic loading setting value, obtains the setting value of discharge through divider according to entry organic loading setting value;

Discharge control loop: as input value, the actuator lift pump in the sewage disposal system is applied control with the discharge setting value, regulate discharge and follow the discharge set point change.

Said entry organic loading setting value control algolithm is:

${\mathrm{Food}}_{\mathrm{sp}}\left(k\right)=\left\{\begin{array}{cc}{\mathrm{Food}}_{\mathrm{sp}}^{\max },& {\mathrm{Food}}_{\mathrm{sp}}(k-1)+{\mathrm{\&Delta;Food}}_{\mathrm{sp}}\left(k\right)>{\mathrm{Food}}_{\mathrm{sp}}^{\max }\\ {\mathrm{Food}}_{\mathrm{sp}}(k-1)+{\mathrm{\&Delta;Food}}_{\mathrm{sp}},& {\mathrm{Food}}_{\mathrm{sp}}^{\min }\&le;{\mathrm{Food}}_{\mathrm{sp}}(k-1)+{\mathrm{\&Delta;Food}}_{\mathrm{sp}}\left(k\right)\&le;{\mathrm{Food}}_{\mathrm{sp}}^{\max }\\ {\mathrm{Food}}_{\mathrm{sp}}^{\min },& {\mathrm{Food}}_{\mathrm{sp}}(k-1)+{\mathrm{\&Delta;Food}}_{\mathrm{sp}}\left(k\right)<{\mathrm{Food}}_{\mathrm{sp}}^{\min }\end{array}\right.---\left(1\right)$

Wherein:

ΔFood _sp(k)＝K _{P_Food}·Δe _COD(k)+K _{I_Food}·e _COD(k) (2)

e _COD(k)＝COD _out，sp(k)-COD _out(k) (3)

Δe _COD(k)＝e _COD(k)-e _COD(k-1) (4)

Food in the formula _Sp(k) be the k time sampling entry organic loading setting value, Food _Sp ^MaxWith Food _Sp ^MinBe the maximal value and the minimum value of sewage treatment plant's entry organic loading design, Δ Food _Sp(k) be the k time sampling entry load setting value added value, k is the sampling number sequence number, COD _{Out, sp}(k) be the k time sampling effluent COD concentration setting value, COD _Out(k) be the k time sampling effluent COD concentration measured value, e _COD(k) be the error of the k time sample effluent COD concentration setting value and effluent COD concentration measured value, e _COD(k-1) be the error of (k-1) inferior sampling effluent COD concentration setting value and effluent COD concentration measured value, Δ e _COD(k) be the error rate of the k time sampling effluent COD concentration, K _{P_Food}And K _{I_Food}Be respectively PI controller ratio and integral parameter.

Said divider is:

${\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="H2009100134414E00031.png"\; state="\{\{state\}\}">Q</figure-callout>}}_{0,\mathrm{sp}}\left(k\right)=\frac{{\mathrm{Food}}_{\mathrm{sp}}\left(k\right)}{{\mathrm{COD}}_m\left(k\right)}---\left(5\right)$

Q in the formula _{0, sp}(k) be the k time sampling discharge setting value, COD _m(k) be the k time sampling entry COD concentration measurement.

Said discharge control loop algorithm is:

$f_Q\left(k\right)=\left\{\begin{array}{cc}{f}_Q^{\max },& f_Q(k-1)+{\mathrm{\&Delta;f}}_Q\left(k\right)>f_Q^{\max }\\ f_Q(k-1)+{\mathrm{\&Delta;f}}_Q,& f_Q^{\min }\&le;f_Q(k-1)+{\mathrm{\&Delta;f}}_Q\left(k\right)\&le;f_Q^{\max }\\ f_Q^{\min },& f_Q(k-1)+{\mathrm{\&Delta;f}}_Q\left(k\right)<f_Q^{\min }\end{array}\right.---\left(6\right)$

Wherein:

${\mathrm{\&Delta;f}}_Q\left(k\right)=K_{P\_\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="H2009100134414E00031.png"\; state="\{\{state\}\}">Q</figure-callout>}0}\&CenterDot;{\mathrm{\&Delta;e}}_{Q_0}\left(k\right)+K_{I\_\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="H2009100134414E00031.png"\; state="\{\{state\}\}">Q</figure-callout>}0}\&CenterDot;e_{Q_0}\left(k\right)---\left(7\right)$

$e_{Q_0}\left(k\right)={\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="H2009100134414E00031.png"\; state="\{\{state\}\}">Q</figure-callout>}}_{0,\mathrm{sp}}-Q_0\left(k\right)---\left(8\right)$

${\mathrm{\&Delta;e}}_{Q_0}\left(k\right)=e_{Q_0}\left(k\right)-e_{Q_0}(k-1)---\left(9\right)$

F in the formula _Q(k) be the k time sampling lift pump frequency converter frequency, f _Q(k-1) be (k-1) inferior sampling lift pump frequency converter frequency, f _Q ^MaxWith f _Q ^MinBe respectively the maximum and the minimum output power of frequency converter, k is the sampling number sequence number, K _{P_Q0}And K _{I_Q0}Be respectively PI controller control ratio and integral parameter,

Be the error of the k time sample discharge setting value and discharge measured value, Be the error of (k-1) inferior sampling discharge setting value and discharge measured value,

Be the discharge error rate of the k time sampling, Q _{0, sp}(k) be the k time sampling discharge setting value, Q ₀(k) be the k time sampling discharge measured value.

Said entry organic loading is the index that is used for characterizing the organism quantity in the entry, representes with following formula:

Food＝COD _inQ ₀ (2)

Food is the entry organic loading in the formula, COD _InBe the COD of entry, Q ₀Be discharge.

It is said that control comprises external loop dissolved oxygen DO setting value control loop and inner looping dissolved oxygen DO Human Simulating Intelligent Control loop to the enforcement of the dissolved oxygen concentration in aeration tank tandem, wherein:

Dissolved oxygen DO setting value control loop: as feedback signal, adopt the PI control algolithm with water outlet COD;

Dissolved oxygen DO Human Simulating Intelligent Control loop: as feedback signal, regulate the dissolved oxygen concentration in the aeration tank through the frequency conversion output frequency of control aeration link actuator with water outlet COD; The algorithm use Human Simulating Intelligent Control Algorithm, this algorithm use hierarchical control mechanism, simulation has the expert's of control experience control behavior, identification current working state on the upper strata; Adopt conventional PID control method at bottom,, realize multi-modal control or decision-making the corresponding pid parameter of the state configuration that picks out.

Said PI control algolithm is:

$S_{O,\mathrm{sp}}\left(k\right)=\left\{\begin{array}{cc}{S}_{O,\mathrm{sp}}^{\max },& S_{O,\mathrm{sp}}(k-1)+{\mathrm{\&Delta;S}}_{O,\mathrm{sp}}\left(k\right)>S_{O,\mathrm{sp}}^{\max }\\ S_{O,\mathrm{sp}}(k-1)+{\mathrm{\&Delta;S}}_{O,\mathrm{sp}}\left(k\right),& S_{O,\mathrm{sp}}^{\min }\&le;S_{O,\mathrm{sp}}(k-1)+{\mathrm{\&Delta;S}}_{O,\mathrm{sp}}\left(k\right)\&le;S_{O,\mathrm{sp}}^{\max }\\ S_{O,\mathrm{sp}}^{\min },& S_{O,\mathrm{sp}}(k-1)+{\mathrm{\&Delta;S}}_{O,\mathrm{sp}}\left(k\right)<S_{O,\mathrm{sp}}^{\min }\end{array}\right.---\left(10\right)$

Wherein

ΔS _O，sp(k)＝K _{P_SO}·e _COD(k)+K _{I_SO}·Δe _COD(k) (11)

e _COD(k)＝COD _out，sp-COD _out(k) (12)

Δe _COD(k)＝e _COD(k)-e _COD(k-1) (13)

S in the formula _{O, sp}(k) be the k time sampling dissolved oxygen concentration setting value, S _{O, sp}(k-1) be (k-1) inferior sampling dissolved oxygen concentration setting value, Δ S _{O, sp}(k) be the k time sampling dissolved oxygen concentration setting value added value, k is the sampling number sequence number, S _{O, sp} ^MaxWith S _{O, sp} ^MinBe respectively the maximum and the minimum value of dissolved oxygen DO setting value, K _{P_SO}And K _{I_SO}Expression PI controller control ratio and integral parameter, e _COD(k) be the error of the k time sample effluent COD concentration setting value and effluent COD concentration measured value, e _COD(k-1) be the error of (k-1) inferior sampling effluent COD concentration setting value and effluent COD concentration measured value, Δ e _COD(k) be the k time sampling effluent COD concentration error rate, COD _Out(k) be the k time sampling effluent COD concentration measured value, COD _{Out, sp}Be the effluent COD concentration setting value;

Said Human Simulating Intelligent Control Algorithm adopts production rule that expertise is described, and this rule list is shown:

IF(condition) THEN(action)；

Response according to the big young pathbreaker system of the error e (k) of dissolved oxygen concentration setting value and measured value is divided into four intervals, carries out staging treating, and the Different control strategy is taked in different intervals;

The positive and negative variation tendency of differentiating error current according to product e (k) the Δ e (k) of the error rate Δ e (k) of the error e (k) of dissolved oxygen concentration setting value and measured value and dissolved oxygen concentration setting value and measured value; Select P, I, D parameter for use and amplify system or rejection coefficient, calculate the frequency conversion output frequency of aeration link actuator at last.

Human Simulating Intelligent Control Algorithm is:

(1) when the Error Absolute Value of dissolved oxygen concentration setting value and measured value | e (k) | more than or equal to error upper limit e2, promptly | and e (k) | during >=e2, the illustrative system error is excessive, and the output valve of controller is by its maximum value output;

(2) when the Error Absolute Value of dissolved oxygen concentration setting value and measured value | e (k) | between error upper limit e2 and error lower limit e1, promptly e1＜| e (k) | during＜e2, consider reduction or do not use integral action that concrete control strategy also should be with reference to error change trend:

(21) if e (k) Δ e (k) >=0, illustrative system output positive deviation setting value, error just change towards the direction that absolute value increases, or error is a certain constant value, remain unchanged, and implement the control of the variation tendency of rapid torsional error absolute value, the controller output valve is:

Δu(k)＝k ₁{K _P1Δe(k)+K _I1e(k)+K _D1[e(k)-2e(k-1)+e(k-2)]} (14)

Δ u (k) is the k time sampling system output increment in the formula, k ₁Be amplification coefficient, K _P1, K _I1And K _D1Be respectively PID controller ratio, integration and differential parameter;

(22) if e (k) Δ e (k)＜0, setting value just is being partial in illustrative system output, error just changes towards the direction that absolute value is reducing, and implements to let system under the assistance of inertia, get back to the control of stable state, the controller output valve is:

Δu(k)＝k ₂{K _P1Δe(k)+K _I1e(k)+K _D1[e(k)-2e(k-1)+e(k-2)]} (15)

In the formula, k ₂The expression rejection coefficient;

(3) when the Error Absolute Value of dissolved oxygen concentration setting value and measured value | e (k) | between error dead band d and error lower limit e1; Be d＜| e (k) | during≤e1, systematic error is less, adds integration; Reducing steady-state error, and the reference error variation tendency is confirmed concrete control strategy:

(31) if e (k) Δ e (k) >=0, the controller output valve is:

Δu(k)＝k ₁{K _P2Δe(k)+K _I2e(k)+K _D2[e(k)-2e(k-1)+e(k-2)]} (16)

K in the formula _P2, K _I2And K _D2Be respectively PID controller ratio, integration and differential parameter;

(32) if e (k) Δ e (k)＜0, the controller output valve is:

Δu(k)＝k ₂{K _P2Δe(k)+K _I2e(k)+K _D2[e(k)-2e(k-1)+e(k-2)]} (17)

(4) when the Error Absolute Value of dissolved oxygen concentration setting value and measured value | e (k) | less than error skip distance d, promptly during e|≤d, ignore systematic error, the controller output valve is 0.

It is said that control comprises control of sludge concentration setting value feedforward and Feedback and sludge concentration control loop to the enforcement of the sludge concentration in aeration tank feedforward-tandem, wherein:

The control of sludge concentration setting value feedforward and Feedback: as feed-forward signal, as feedback signal, feedforward control adopts proportional control with water outlet COD with the entry organic loading, and FEEDBACK CONTROL adopts PI control;

Sludge concentration control loop: as feedback signal, realize through control mud capacity of returns with sludge concentration measured value in the aeration tank.

Said feedforward control adopts proportional control, obtains through following formula:

MLSS _sp，ff(k)＝k _{ff_MLSS}·Food _avg(k) (18)

MLSS in the formula _{Sp, ff}(k) be the k time sampling sludge concentration setting value feedforward value, K _{Ff_MLSS}The static feedforward of expression scale-up factor, Food _Avg(k) be 2 hours sliding averages of the k time sampling entry organic loading;

Feedback controller adopts PI control, shown in (22).

Wherein

ΔMLSS _sp(k)＝K _{P_MLSS}·e _{COD_avg}(k)+K _{I_MLSS}·Δe _{COD_avg}(k) (20)

e _COD(k)＝COD _out，sp-COD _out(k) (21)

Δe _COD(k)＝e _COD(k)-e _COD(k-1) (22)

The final setting value of sludge concentration is suc as formula shown in (26):

MLSS _SP(k)＝MLSS _{SP_ff}(k)+MLSS _{SP_fb}(k) (23)

MLSS in the formula _{Sp, fb}(k) be the k time sampling sludge concentration setting value value of feedback, MLSS _{Sp, fb}(k-1) be (k-1) inferior sampling sludge concentration setting value value of feedback, Δ MLSS _{Sp, fb}(k) be the k time sampling sludge concentration setting value value of feedback added value, MLSS _Sp ^MaxWith MLSS _Sp ^MinBe respectively the maximum and the minimum value of sludge concentration setting value, K _{P_MLSS}And K _{I_MLSS}Be respectively PI controller control ratio and integral parameter, e _COD(k) be the error of the k time sample effluent COD concentration setting value and effluent COD concentration measured value, e _COD(k-1) be the error of (k-1) inferior sampling effluent COD concentration setting value and effluent COD concentration measured value, Δ e _COD(k) be the k time sampling effluent COD concentration error rate, COD _Out(k) be the k time sampling effluent COD concentration measured value, COD _{Out, sp}Be effluent COD concentration setting value, MLSS _Sp(k) be the k time sampling sludge concentration setting value;

Said sludge concentration control loop algorithm is:

Wherein:

ΔQ _r(k)＝K _{P_Qr}·e _MLSS(k)+K _{I_Qr}·Δe _MLSS(k) (28)

e _MLSS(k)＝MLSS _sp-MLSS(k) (29)

Δe _MLSS(k)＝e _MLSS(k)-e _MLSS(k-1) (30)

Q in the formula _r(k) be the k time sampling mud capacity of returns measured value, Δ Q _r(k) be the k time sampling mud capacity of returns measured value added value, Q _r ^MaxWith Q _r ^MinMaximum and the minimum value of representing the mud capacity of returns respectively, Q rule of thumb usually _r ^MaxBe 3-5 discharge doubly, Q _r ^MinBe 0.5-1.5 discharge doubly; K _{P_Qr}And K _{I_Qr}Expression PI controller control ratio and integral parameter, e _MLSS(k) be the error of the k time sample sludge concentration setting value and sludge concentration measured value, e _MLSS(k-1) be the error of (k-1) inferior sampling sludge concentration setting value and sludge concentration measured value, Δ e _MLSS(k) be the k time sampling sludge concentration error rate, MLSS (k) is the k time sampling sludge concentration measured value, MLSS _SpBe the sludge concentration setting value.

The present invention has following beneficial effect and advantage:

1. reduce the biochemical processing procedure of sewage energy consumption.The inventive method adopts the circuit controls mode; Characteristics to different controlling object adopt the Different control algorithm; As the entry organic loading being adopted tandem control, dissolved oxygen concentration is adopted Human Simulating Intelligent Control Algorithm, sludge concentration is adopted the control of feedforward-tandem; Realization is implemented effective control to the biochemical processing procedure of sewage of sewage lifting, blower fan aeration and three main power consumption links of mud backflow, is guaranteeing obviously to reduce the biochemical processing procedure of sewage energy consumption under the stable water outlet prerequisite up to standard.

2. up to standard as controlled target with effluent quality; Entry organic loading, dissolved oxygen concentration and three controlling object of sludge concentration are implemented circuit controls; The controlled target unification of three objects on the effluent quality index, has been realized nutrition-entry organic loading, oxygen supply-dissolved oxygen concentration and the microorganism-sludge concentration three's of biochemical processing procedure of sewage balance.

3. non-linear with characteristics such as timely changes to entry organic loading, dissolved oxygen concentration and three controlling object of sludge concentration; Implement control of circuit controls tandem and Human Simulating Intelligent Control Algorithm, solved the control of traditional PI D single loop and be difficult to obtain the desirable problem of controlling effect.

Description of drawings

Fig. 1 is an A/O process flow diagram in the prior art;

Fig. 2 is discharge setting value loop control principle figure in the inventive method;

Fig. 3 is discharge loop control principle figure in the inventive method;

Fig. 4 (a) is dissolved oxygen DO setting value loop control principle figure in the inventive method;

Fig. 4 (b) is a dissolved oxygen DO setting value loop control principle block diagram in the inventive method;

Fig. 5 is dissolved oxygen DO loop control principle figure in the inventive method;

Fig. 6 is artificial intelligent PID control principle figure in the inventive method;

Fig. 7 is second-order system dynamic response curve figure in the inventive method;

Fig. 8 (a) is sludge concentration setting value control principle figure in the inventive method;

Fig. 8 (b) is a sludge concentration setting value control principle block diagram in the inventive method;

Fig. 9 is sludge concentration loop control principle figure in the inventive method;

Figure 10 (a) is the entry situation pictures taken on the spot of sewage treatment plant;

Figure 10 (b) is for using the water outlet situation pictures taken on the spot after the inventive method is handled.

Embodiment

A kind of biochemical processing procedure of sewage circuit controls of the present invention method may further comprise the steps:

The entry organic loading is carried out tandem control, and " labour " of the biochemical treatment of control sewage is " food " quantity of microorganism, to reduce the energy consumption that sewage promotes link;

Dissolved oxygen concentration in the aeration tank is implemented tandem control, realize the air feed as required of aeration tank, to reduce the energy consumption of blower fan aeration link;

Sludge concentration in the aeration tank is implemented the control of feedforward-tandem, and the micro organism quantity in the control sewage disposal system is to reduce the energy consumption of mud backflow link.

The purpose of discharge control is in order to regulate organism quantity to be removed in the sewage treatment plant.In the sewage biochemical treatment system, organism is microorganism " food ", and organic quantity can directly influence reproduction speed and the effluent quality of microorganism in the control entry.Here define the entry organic loading and characterize the organism quantity in the entry, shown in (1).

Food＝BOD _5，inQ ₀ (25)

Food representes entry organic loading, BOD in the formula _{5, in}BOD BOD on the 5th of expression entry ₅(Biological Oxygen Demanded), Q ₀Expression discharge.

Because BOD ₅Measuring period is long, needs 5 days time, and therefore the on-line measurement difficulty in theoretical Research And Engineering practice, replaces BOD with chemical oxygen demand COD usually ₅COD generally is higher than BOD ₅, difference therebetween can approximately be expressed as and can not be the organism of microbial degradation.BOD in the sanitary sewage ₅Being roughly 0.4-0.8 with the ratio of COD, is relatively-stationary for this ratio of sewage of specific water quality.Therefore the entry organic loading can use formula (2) to calculate usually.

Food _COD＝COD _inQ ₀ (26)

Food in the formula _CODBe the entry organic loading that adopts entry COD to calculate, COD _InBe the COD of entry, Q ₀Be discharge.

Because entry COD is by the decision of entry water quality, sewage treatment plant is uncontrollable, and therefore the control for the entry organic loading is exactly in fact the control for discharge.

1. discharge control loop

In order to keep the stability of sewage disposal system water outlet; The present invention proposes the tandem control of discharge; As shown in Figure 2; Take all factors into consideration entry water quality and disturb the variation with effluent quality, realize the control of entry organic loading setting value, obtain the setting value of discharge according to entry organic loading setting value through divider through the feedback of water outlet COD.

Saidly the entry organic loading is carried out tandem control comprise external loop discharge setting value control loop and inner looping discharge control loop, wherein:

Discharge setting value control loop: according to the variation of entry water quality and effluent quality, the feedback through water outlet COD realizes the control of entry organic loading setting value, obtains the setting value of discharge through divider according to entry organic loading setting value;

Discharge control loop: as input value, the actuator lift pump in the sewage disposal system is applied control with the discharge setting value, regulate discharge and follow the discharge set point change.

1.1 discharge setting value control loop

The entry organic loading characterizes organic content in the entry; And organism is a microorganism (active sludge) " food "; Therefore the setting value of entry organic loading is mainly determined by microorganism " appetite "; Specifically by the decision of information such as the population of microorganism, activity, quantity, but these indexs mostly are difficult to on-line measurement, and whether the present invention adopts water outlet COD signal to come indirect reflection entry organic loading appropriate.Effluent COD concentration exceeds standard, and explains that the entry organic loading is excessive; Effluent COD concentration is too small, explains that the entry organic loading is not enough.

Water outlet COD controller adopts PI control, shown in (3).

${\mathrm{Food}}_{\mathrm{sp}}\left(k\right)=\left\{\begin{array}{cc}{\mathrm{Food}}_{\mathrm{sp}}^{\max },& {\mathrm{Food}}_{\mathrm{sp}}(k-1)+{\mathrm{\&Delta;Food}}_{\mathrm{sp}}\left(k\right)>{\mathrm{Food}}_{\mathrm{sp}}^{\max }\\ {\mathrm{Food}}_{\mathrm{sp}}(k-1)+{\mathrm{\&Delta;Food}}_{\mathrm{sp}},& {\mathrm{Food}}_{\mathrm{sp}}^{\min }\&le;{\mathrm{Food}}_{\mathrm{sp}}(k-1)+{\mathrm{\&Delta;Food}}_{\mathrm{sp}}\left(k\right)\&le;{\mathrm{Food}}_{\mathrm{sp}}^{\max }\\ {\mathrm{Food}}_{\mathrm{sp}}^{\min },& {\mathrm{Food}}_{\mathrm{sp}}(k-1)+{\mathrm{\&Delta;Food}}_{\mathrm{sp}}\left(k\right)<{\mathrm{Food}}_{\mathrm{sp}}^{\min }\end{array}\right.---\left(27\right)$

Wherein:

ΔFood _sp(k)＝K _{P_Food}·Δe _COD(k)+K _{I_Food}·e _COD(k) (28)

e _COD(k)＝COD _out，sp(k)-COD _out(k) (29)

Δe _COD(k)＝e _COD(k)-e _COD(k-1) (30)

Food in the formula _Sp(k) be the k time sampling entry organic loading setting value, Food _Sp ^MaxWith Food _Sp ^MinBe the maximal value and the minimum value of sewage treatment plant's entry organic loading design, relevant with the volume of sewage treatment plant biochemistry pool with the microbial species group structure, Δ Food _Sp(k) be the k time sampling entry load setting value added value, k is the sampling number sequence number, COD _{Out, sp}(k) be the k time sampling effluent COD concentration setting value, COD _Out(k) be the k time sampling effluent COD concentration measured value, e _COD(k) be the error of the k time sample effluent COD concentration setting value and effluent COD concentration measured value, e _COD(k-1) be the error of (k-1) inferior sampling effluent COD concentration setting value and effluent COD concentration measured value, Δ e _COD(k) be the error rate of the k time sampling effluent COD concentration, K _{P_Food}And K _{I_Food}Be respectively PI controller ratio and integral parameter.

After obtaining the setting value of entry organic loading, just can obtain the setting value of discharge through divider, shown in (7).

${\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="H2009100134414E00031.png"\; state="\{\{state\}\}">Q</figure-callout>}}_{0,\mathrm{sp}}\left(k\right)=\frac{{\mathrm{Food}}_{\mathrm{sp}}\left(k\right)}{{\mathrm{COD}}_m\left(k\right)}---\left(31\right)$

Q in the formula _{0, sp}(k) be the k time sampling discharge setting value, COD _In(k) be the k time sampling entry COD concentration measurement.

1.2 discharge control loop

In the practical application process, the discharge control loop finally is that the actuator lift pump is applied control action, and controller is realized the control to lift pump through the output frequency of control of conversion device, and discharge control loop structure is as shown in Figure 3.

The discharge controller adopts PI control, shown in (8).

$f_Q\left(k\right)=\left\{\begin{array}{cc}{f}_Q^{\max },& f_Q(k-1)+{\mathrm{\&Delta;f}}_Q\left(k\right)>f_Q^{\max }\\ f_Q(k-1)+{\mathrm{\&Delta;f}}_Q\left(k\right),& f_Q^{\min }\&le;f_Q(k-1)+{\mathrm{\&Delta;f}}_Q\left(k\right)\&le;f_Q^{\max }\\ f_Q^{\min },& f_Q(k-1)+{\mathrm{\&Delta;f}}_Q\left(k\right)<f_Q^{\min }\end{array}\right.---\left(32\right)$

Wherein:

${\mathrm{\&Delta;f}}_Q\left(k\right)=K_{P\_\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="H2009100134414E00031.png"\; state="\{\{state\}\}">Q</figure-callout>}0}\&CenterDot;{\mathrm{\&Delta;e}}_{Q_0}\left(k\right)+K_{I\_\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="H2009100134414E00031.png"\; state="\{\{state\}\}">Q</figure-callout>}0}\&CenterDot;e_{Q_0}\left(k\right)---\left(33\right)$

$e_{Q_0}\left(k\right)={\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="H2009100134414E00031.png"\; state="\{\{state\}\}">Q</figure-callout>}}_{0,\mathrm{sp}}\left(k\right)-Q_0\left(k\right)---\left(34\right)$

${\mathrm{\&Delta;e}}_{Q_0}\left(k\right)=e_{Q_0}\left(k\right)-e_{Q_0}(k-1)---\left(35\right)$

F in the formula _Q(k) be the k time sampling lift pump frequency converter frequency, f _Q(k-1) be (k-1) inferior sampling lift pump frequency converter frequency, f _Q ^MaxWith f _Q ^MinBe respectively the maximum and the minimum output power of frequency converter, usually f _Q ^MaxGet work frequency 50Hz, consider the resistance of ducting of oxygen transmission, f _Q ^MinUsually get 20-35Hz; Δ f _Q(k) be the k time sampling lift pump frequency converter frequency added value, k is the sampling number sequence number, K _{P_Q0}And K _{I_Q0}Be respectively PI controller control ratio and integral parameter, Be the error of the k time sample discharge setting value and discharge measured value,

Be the error of (k-1) inferior sampling discharge setting value and discharge measured value,

Be the discharge error rate of the k time sampling, Q _{0, sp}(k) be the k time sampling discharge setting value, Q ₀(k) be the k time sampling discharge measured value.

2. dissolved oxygen DO control loop

It is said that control comprises external loop dissolved oxygen DO setting value control loop and inner looping dissolved oxygen DO Human Simulating Intelligent Control loop to the enforcement of the dissolved oxygen concentration in aeration tank tandem, wherein:

Dissolved oxygen DO setting value control loop: as feedback signal, adopt the PI control algolithm with water outlet COD;

Dissolved oxygen DO Human Simulating Intelligent Control loop: as feedback signal, regulate the dissolved oxygen concentration in the aeration tank through the frequency conversion output frequency of control aeration link actuator with water outlet COD; The algorithm use Human Simulating Intelligent Control Algorithm, this algorithm use hierarchical control mechanism, simulation has the expert's of control experience control behavior, identification current working state on the upper strata; Adopt conventional PID control method at bottom,, realize multi-modal control or decision-making the corresponding pid parameter of the state configuration that picks out.

2.1 dissolved oxygen DO setting value control loop

Because the dissolved oxygen concentration setting value directly influences effluent quality, so the dissolved oxygen DO setting value control loop structure of the present invention's proposition shown in Fig. 4 (a), 4 (b), as feedback signal, control algolithm adopts the PI control algolithm with water outlet COD.

Said PI control algolithm is:

$S_{O,\mathrm{sp}}\left(k\right)=\left\{\begin{array}{cc}{S}_{O,\mathrm{sp}}^{\max },& S_{O,\mathrm{sp}}(k-1)+{\mathrm{\&Delta;S}}_{O,\mathrm{sp}}\left(k\right)>S_{O,\mathrm{sp}}^{\max }\\ S_{O,\mathrm{sp}}(k-1)+{\mathrm{\&Delta;S}}_{O,\mathrm{sp}}\left(k\right),& S_{O,\mathrm{sp}}^{\min }\&le;S_{O,\mathrm{sp}}(k-1)+{\mathrm{\&Delta;S}}_{O,\mathrm{sp}}\left(k\right)\&le;S_{O,\mathrm{sp}}^{\max }\\ S_{O,\mathrm{sp}}^{\min },& S_{O,\mathrm{sp}}(k-1)+{\mathrm{\&Delta;S}}_{O,\mathrm{sp}}\left(k\right)<S_{O,\mathrm{sp}}^{\min }\end{array}\right.---\left(36\right)$

Wherein

ΔS _O，sp(k)＝K _{P_SO}·e _COD(k)+K _{I_SO}·Δe _COD(k) (37)

e _COD(k)＝COD _out，sp-COD _out(k) (38)

Δe _COD(k)＝e _COD(k)-e _COD(k-1) (39)

S in the formula _{O, sp}(k) be the k time sampling dissolved oxygen concentration setting value, S _{O, sp}(k-1) be (k-1) inferior sampling dissolved oxygen concentration setting value, Δ S _{O, sp}(k) be the k time sampling dissolved oxygen concentration setting value added value, k is the sampling number sequence number, S _{O, sp} ^MaxWith S _{O, sp} ^MinBe respectively the maximum and the minimum value of dissolved oxygen DO setting value, usually S _{O, sp} ^MaxValue 2.5-3.5mg/L, S _{O, sp} ^MinValue 1.0-2.5mg/L, K _{P_SO}And K _{I_SO}Expression PI controller control ratio and integral parameter, e _COD(k) be the error of the k time sample effluent COD concentration setting value and effluent COD concentration measured value, e _COD(k-1) be the error of (k-1) inferior sampling effluent COD concentration setting value and effluent COD concentration measured value, Δ e _COD(k) be the k time sampling effluent COD concentration error rate, COD _Out(k) be the k time sampling effluent COD concentration measured value, COD _{Out, sp}Be the effluent COD concentration setting value;

2.2 dissolved oxygen DO Human Simulating Intelligent Control

Dissolved oxygen DO control system structure is as shown in Figure 5, and DOsp and DO represent the setting value and the actual value of dissolved oxygen DO respectively among the figure.Controller is regulated the rotating speed of motor through the output of control of conversion device, thereby controls the operation of fan blower, finally realizes The Control of Dissolved Oxygen.

The present invention propose based on expertise the dissolved oxygen DO Human Simulating Intelligent Control (Human-simulationintelligent control, HSIC), the control behavior of simulating expert.

The present invention combines Human Simulating Intelligent Control and traditional PID control method; Artificial intelligent PID control method is proposed; Its basic thought is to adopt hierarchical control mechanism; Adopt intelligence control method on the upper strata, simulation has the operator's who enriches control experience control behavior, differentiates current working state to greatest extent through feature identification.Adopt conventional PID control method at bottom,, thereby realize multi-modal control or decision-making the corresponding pid parameter of the state configuration that picks out.The algorithm use production rule is described expertise, and rule list is shown:

IF(condition) THEN(action)

This rule-based symbolic Model is applicable to describes cause-effect relationship and non-qualitatively analytic relationship, is convenient to the intuition inference logic and the various fuzzy message qualitatively of expressing human, and the reasoning decision-making rapidly accurately.

Algorithm selects for use the error rate Δ e (k) of error e (k) and dissolved oxygen concentration setting value and measured value of dissolved oxygen concentration setting value and measured value as the input variable of controller; The behavioral characteristics of descriptive system; Characterize its residing duty; As shown in Figure 6, the output u (k) of system can represent with formula.

u(k)＝f(e(k)，Δe(k)) (40)

The bright specifically control principle of dynamic response of the second-order system that provides below in conjunction with Fig. 7, d is the error skip distance among the figure, u representes controller output, e (k)=SP-u (k), Δ e (k)=e (k)-e (k-1).

At first, be divided into I, II, III, four intervals of IV according to the response of the big young pathbreaker system of the error e (k) of dissolved oxygen concentration setting value and measured value, carry out staging treating, the Different control strategy is taked in different intervals;

Then; The positive and negative variation tendency of differentiating error current according to product e (k) the Δ e (k) of the error rate Δ e (k) of the error e (k) of dissolved oxygen concentration setting value and measured value and dissolved oxygen concentration setting value and measured value; Select P, I, D parameter for use and amplify system or rejection coefficient, calculate the frequency conversion output frequency of aeration link actuator at last.

Said Human Simulating Intelligent Control Algorithm is:

(1) when the Error Absolute Value of dissolved oxygen concentration setting value and measured value | e (k) | more than or equal to error upper limit e2; Promptly | e (k) | during >=e2 (interval IV among Fig. 7); The illustrative system error is excessive, this moment no matter the error variation tendency how, the output valve of controller is by its maximum value output; With rapid adjustment error, Error Absolute Value is reduced with maximal rate, at this moment be equivalent to implement open loop control;

(2) when the Error Absolute Value of dissolved oxygen concentration setting value and measured value | e (k) | between error upper limit e2 and error lower limit e1; Be e1＜| e (k) | during＜e2 (interval III among Fig. 7); Systematic error is bigger; Consider reduction or do not use integral action that concrete control strategy also should be with reference to error change trend:

(21) if e (k) Δ e (k) >=0 (like AB section, CD section and EF section among Fig. 7), illustrative system output positive deviation setting value, error just change towards the direction that absolute value increases, or error is a certain constant value, remain unchanged.Can consider to implement stronger control action this moment, with the variation tendency of rapid torsional error absolute value, prevents that it from continuing to increase, and controller is exported suc as formula shown in (17):

Δu(k)＝k ₁{K _P1Δe(k)+K _I1e(k)+K _D1[e(k)-2e(k-1)+e(k-2)]} (41)

Δ u (k) is the k time sampling system output increment in the formula, k ₁Be amplification coefficient, e (k)＞0 o'clock, k1＞1, e (k)＜0 o'clock, k1＜-1, K _P1, K _I1And K _D1Be respectively PID controller ratio, integration and differential parameter;

(22) if e (k) Δ e (k)＜0 (like OA section, BC section, DE and FG section among Fig. 7); Setting value just is being partial in illustrative system output; Error just changes towards the direction that absolute value reduces, and can consider to implement more weak control action this moment, lets system under the assistance of inertia, get back to stable state.So both can subtract

Mini system overshoot does not influence the response speed of system again.The controller output valve is suc as formula shown in (18).

Δu(k)＝k ₂{K _P1Δe(k)+K _I1e(k)+K _D1[e(k)-2e(k-1)+e(k-2)]} (42)

K in the formula ₂The expression rejection coefficient, e (k)＞0 o'clock, 0＜k ₂＜1, e (k)＜0 o'clock ,-1＜k ₂＜0;

(3) when the Error Absolute Value of dissolved oxygen concentration setting value and measured value | e (k) | between error skip distance d and error lower limit e1; Be d＜| e (k) | during≤e1 (interval II among Fig. 7); Systematic error is less; Should add integration this moment, to reduce steady-state error, also wants the reference error variation tendency to confirm concrete control strategy:

(31) if e (k) Δ e (k) >=0 can implement stronger control action, the controller output valve is:

Δu(k)＝k ₁{K _P2Δe(k)+K _I2e(k)+K _D2[e(k)-2e(k-1)+e(k-2)]} (43)

K in the formula _P2, K _I2And K _D2Be respectively PID controller ratio, integration and differential parameter;

(32) if e (k) Δ e (k)＜0 can implement more weak control action, the controller output valve is:

Δu(k)＝k ₂{K _P2Δe(k)+K _I2e(k)+K _D2[e(k)-2e(k-1)+e(k-2)]} (44)

(4) when the Error Absolute Value of dissolved oxygen concentration setting value and measured value | e (k) | less than error skip distance d, promptly during e|≤d, ignore systematic error, the controller output valve is 0.

3. sludge concentration control loop

It is said that control comprises control of sludge concentration setting value feedforward and Feedback and sludge concentration control loop to the enforcement of the sludge concentration in aeration tank feedforward-tandem, wherein:

The control of sludge concentration setting value feedforward and Feedback: as feed-forward signal, as feedback signal, feedforward control adopts proportional control with water outlet COD with the entry organic loading, and FEEDBACK CONTROL adopts PI control;

Sludge concentration control loop: as feedback signal, realize through control mud capacity of returns with sludge concentration measured value in the aeration tank.

Sludge concentration is that (the amount MLSS with mixed liquor suspended solid, MLSS representes microbial numbers; The proportionate relationship that MixedLiquid Suspended Sludge) should keep relative stability with the entry organic loading; The present invention as feed-forward signal, proposes sludge concentration feedforward-tandem control as shown in Figure 8, because sludge concentration is slow speed per hour control variable with the entry organic loading; Time scale in hour; Therefore rule of thumb, here with 2 hours sliding averages of entry organic loading as feed-forward signal, tandem control with the measured value of water outlet COD and aeration tank sludge concentration as the sludge concentration feedback signal.

3.1 sludge concentration setting value control loop

Said feedforward control adopts proportional control, obtains through following formula:

MLSS _sp，ff(k)＝k _{ff_MLSS}·Food _avg(k) (45)

MLSS in the formula _{Sp, ff}(k) be the k time sampling sludge concentration setting value feedforward value, K _{Ff_MLSS}The static feedforward of expression scale-up factor, Food _Avg(k) be 2 hours sliding averages of the k time sampling entry organic loading;

Feedback controller adopts PI control, shown in (22).

Wherein

ΔMLSS _sp，fb(k)＝K _{P_MLSS}·e _COD(k)+K _{I_MLSS}·Δe _COD(k) (47)

e _COD(k)＝COD _out，sp-COD _out(k) (48)

Δe _COD(k)＝e _COD(k)-e _COD(k-1) (49)

The final setting value of sludge concentration is suc as formula shown in (26):

MLSS _sp(k)＝MLSS _sp，ff(k)+MLSS _sp，fb(k) (50)

MLSS in the formula _{Sp, fb}(k) be the k time sampling sludge concentration setting value value of feedback, MLSS _{Sp, fb}(k-1) be (k-1) inferior sampling sludge concentration setting value value of feedback, Δ MLSS _{Sp, fb}(k) be the k time sampling sludge concentration setting value value of feedback added value, MLSS _Sp ^MaxWith MLSS _Sp ^MinBe respectively the maximum and the minimum value of sludge concentration setting value, consider MLSS from energy consumption with preventing the angle that the sludge bulking unusual service condition takes place usually _Sp ^MaxValue 1500-4000mg/L considers MLSS from effluent quality _Sp ^MinValue 600-1000mg/L, K _{P_MLSS}And K _{I_MLSS}Be respectively PI controller control ratio and integral parameter, e _COD(k) be the error of the k time sample effluent COD concentration setting value and effluent COD concentration measured value, e _COD(k-1) be the error of (k-1) inferior sampling effluent COD concentration setting value and effluent COD concentration measured value, Δ e _COD(k) be the k time sampling effluent COD concentration error rate, COD _Out(k) be the k time sampling effluent COD concentration measured value, COD _{Out, sp}Be effluent COD concentration setting value, MLSS _Sp(k) be the k time sampling sludge concentration setting value;

3.2 sludge concentration circuit controls

The control of said sludge concentration realizes through control mud capacity of returns, and is as shown in Figure 9.The sludge concentration controller algorithm is:

Wherein:

ΔQ _r(k)＝K _{P_Qr}·e _MLSS(k)+K _{I_Qr}·Δe _MLSS(k) (28)

e _MLSS(k)＝MLSS _sp-MLSS(k) (29)

Δe _MLSS(k)＝e _MLSS(k)-e _MLSS(k-1) (30)

Q in the formula _r(k) be the k time sampling mud capacity of returns measured value, Δ Q _r(k) be the k time sampling mud capacity of returns measured value added value, Q _r ^MaxWith Q _r ^MinMaximum and the minimum value of representing the mud capacity of returns respectively, Q rule of thumb usually _r ^MaxBe 3-5 discharge doubly, Q _r ^MinBe 0.5-1.5 discharge doubly; K _{P_Qr}And K _{I_Qr}Expression PI controller control ratio and integral parameter, e _MLSS(k) be the error of the k time sample sludge concentration setting value and sludge concentration measured value, e _MLSS(k-1) be the error of (k-1) inferior sampling sludge concentration setting value and sludge concentration measured value, Δ e _MLSS(k) be the k time sampling sludge concentration error rate, MLSS (k) is the k time sampling sludge concentration measured value, MLSS _SpBe the sludge concentration setting value.

Shown in Figure 10 (a), 10 (b), certain sewage treatment plant has used the inventive method biochemical processing procedure of sewage has been carried out circuit controls, and stable effluent quality is up to standard, and simultaneity factor power consumption descends 10%.

Claims (1)

1. biochemical processing procedure of sewage circuit controls method is characterized in that may further comprise the steps:

The entry organic loading is carried out tandem control, and " labour " of the biochemical treatment of control sewage is " food " quantity of microorganism, to reduce the energy consumption that sewage promotes link;

Dissolved oxygen concentration in the aeration tank is implemented tandem control, realize the air feed as required of aeration tank, to reduce the energy consumption of blower fan aeration link;

Sludge concentration in the aeration tank is implemented the control of feedforward-tandem, and the micro organism quantity in the control sewage disposal system is to reduce the energy consumption of mud backflow link;

Saidly the entry organic loading is carried out tandem control comprise external loop discharge setting value control loop and inner looping discharge control loop, wherein:

Discharge setting value control loop: according to the variation of entry water quality and effluent quality, the feedback through water outlet COD realizes the control of entry organic loading setting value, obtains the setting value of discharge through divider according to entry organic loading setting value;

Discharge control loop: as input value, the actuator lift pump in the sewage disposal system is applied control with the discharge setting value, regulate discharge and follow the discharge set point change;

Said entry organic loading setting value control algolithm is:

${\mathrm{Food}}_{\mathrm{sp}}\left(k\right)=\left\{\begin{array}{cc}{\mathrm{Food}}_{\mathrm{sp}}^{\max },& {\mathrm{Food}}_{\mathrm{sp}}(k-1)+\mathrm{\&Delta;}{\mathrm{Food}}_{\mathrm{sp}}\left(k\right)>{\mathrm{Food}}_{\mathrm{sp}}^{\max }\\ {\mathrm{Food}}_{\mathrm{sp}}(k-1)+\mathrm{\&Delta;}{\mathrm{Food}}_{\mathrm{sp}}\left(k\right),& {\mathrm{Food}}_{\mathrm{sp}}^{\min }\&le;{\mathrm{Food}}_{\mathrm{sp}}(k-1)+\mathrm{\&Delta;}{\mathrm{Food}}_{\mathrm{sp}}\left(k\right)\&le;{\mathrm{Food}}_{\mathrm{sp}}^{\max }\\ {\mathrm{Food}}_{\mathrm{sp}}^{\min },& {\mathrm{Food}}_{\mathrm{sp}}(k-1)+\mathrm{\&Delta;}{\mathrm{Food}}_{\mathrm{sp}}\left(k\right)<{\mathrm{Food}}_{\mathrm{sp}}^{\min }\end{array}\right.$

(1)

Wherein:

ΔFood _sp(k)＝K _{P_Food}·Δe _COD(k)+K _{I_Food}·e _COD(k) (2)

e _COD(k)＝COD _out，sp(k)-COD _out(k) (3)

Δe _COD(k)＝e _COD(k)-e _COD(k-1) (4)

Food in the formula _Sp(k) be the k time sampling entry organic loading setting value,

With Be the maximal value and the minimum value of sewage treatment plant's entry organic loading design, Δ Food _Sp(k) be the k time sampling entry load setting value added value, k is the sampling number sequence number, COD _{Out, sp}(k) be the k time sampling effluent COD concentration setting value, COD _Out(k) be the k time sampling effluent COD concentration measured value, e _COD(k) be the error of the k time sample effluent COD concentration setting value and effluent COD concentration measured value, e _COD(k-1) be the error of (k-1) inferior sampling effluent COD concentration setting value and effluent COD concentration measured value, Δ e _COD(k) be the error rate of the k time sampling effluent COD concentration, K _{P_Food}And K _{I_Food}Be respectively PI controller ratio and integral parameter;

Said divider is:

$Q_{0,\mathrm{sp}}\left(k\right)=\frac{{\mathrm{Food}}_{\mathrm{sp}}\left(k\right)}{{\mathrm{COD}}_{\mathrm{in}}\left(k\right)}---\left(5\right)$

Q in the formula _{0, sp}(k) be the k time sampling discharge setting value, COD _In(k) be the k time sampling entry COD concentration measurement;

Said discharge control loop algorithm is:

$f_Q\left(k\right)=\left\{\begin{array}{cc}{f}_Q^{\max },& f_Q(k-1)+\mathrm{\&Delta;}{f}_Q\left(k\right)>f_Q^{\max }\\ f_Q(k-1)+\mathrm{\&Delta;}{f}_Q\left(k\right),& f_Q^{\min }\&le;f_Q(k-1)+\mathrm{\&Delta;}{f}_Q\left(k\right)\&le;f_Q^{\max }\\ f_Q^{\min },& f_Q(k-1)+\mathrm{\&Delta;}{f}_Q\left(k\right)<f_Q^{\min }\end{array}\right.---\left(6\right)$

Wherein:

$\mathrm{\&Delta;}{f}_Q\left(k\right)=K_{P\_Q0}\&CenterDot;\mathrm{\&Delta;}{e}_{Q_0}\left(k\right)+K_{I\_Q0}\&CenterDot;e_{Q_0}\left(k\right)---\left(7\right)$

$e_{Q_0}\left(k\right)=Q_{0,\mathrm{sp}}-Q_0\left(k\right)---\left(8\right)$

$\mathrm{\&Delta;}{e}_{Q_0}\left(k\right)=e_{Q_0}\left(k\right)-e_{Q_0}(k-1)---\left(9\right)$

F in the formula _Q(k) be the k time sampling lift pump frequency converter frequency, f _Q(k-1) be (k-1) inferior sampling lift pump frequency converter frequency,

With

Be respectively the maximum and the minimum output power of frequency converter, k is the sampling number sequence number, K _{P_Q0}And K _{I_Q0}Be respectively PI controller control ratio and integral parameter,

Be the error of the k time sample discharge setting value and discharge measured value,

Be the error of (k-1) inferior sampling discharge setting value and discharge measured value,

Be the discharge error rate of the k time sampling, Q _{0, sp}(k) be the k time sampling discharge setting value, Q ₀(k) be the k time sampling discharge measured value;

Said entry organic loading is the index that is used for characterizing the organism quantity in the entry, representes with following formula:

Food＝COD _inQ ₀

Food is the entry organic loading in the formula, COD _InBe the COD of entry, Q ₀Be discharge;

It is said that control comprises external loop dissolved oxygen DO setting value control loop and inner looping dissolved oxygen DO Human Simulating Intelligent Control loop to the enforcement of the dissolved oxygen concentration in aeration tank tandem, wherein:

Dissolved oxygen DO setting value control loop: as feedback signal, adopt the PI control algolithm with water outlet COD;

Dissolved oxygen DO Human Simulating Intelligent Control loop: as feedback signal, regulate the dissolved oxygen concentration in the aeration tank through the frequency conversion output frequency of control aeration link actuator with water outlet COD; The algorithm use Human Simulating Intelligent Control Algorithm, this algorithm use hierarchical control mechanism, simulation has the expert's of control experience control behavior, identification current working state on the upper strata; Adopt conventional PID control method at bottom,, realize multi-modal control or decision-making the corresponding pid parameter of the state configuration that picks out;

Said PI control algolithm is:

$S_{O,\mathrm{sp}}\left(k\right)=\left\{\begin{array}{cc}{S}_{O,\mathrm{sp}}^{\max },& S_{O,\mathrm{sp}}(k-1)+\mathrm{\&Delta;}{S}_{O,\mathrm{sp}}\left(k\right)>S_{O,\mathrm{sp}}^{\max }\\ S_{O,\mathrm{sp}}(k-1)+\mathrm{\&Delta;}{S}_{O,\mathrm{sp}}\left(k\right),& S_{O,\mathrm{sp}}^{\min }\&le;S_{O,\mathrm{sp}}(k-1)+\mathrm{\&Delta;}{S}_{O,\mathrm{sp}}\left(k\right)\&le;\\ S_{O,\mathrm{sp}}^{\min },& S_{O,\mathrm{sp}}(k-1)+\mathrm{\&Delta;}{S}_{O,\mathrm{sp}}\left(k\right)<S_{O,\mathrm{sp}}^{\min }\end{array},S_{O,\mathrm{sp}}^{\max }\right.---\left(10\right)$

Wherein

ΔS _O，sp(k)＝K _{P_SO}·e _COD(k)+K _{I_SO}·Δe _COD(k) (11)

e _COD(k)＝COD _out，sp-COD _out(k) (12)

Δe _COD(k)＝e _COD(k)-e _COD(k-1) (13)

S in the formula _{O, sp}(k) be the k time sampling dissolved oxygen concentration setting value, S _{O, sp}(k-1) be (k-1) inferior sampling dissolved oxygen concentration setting value, Δ S _{O, sp}(k) be the k time sampling dissolved oxygen concentration setting value added value, k is the sampling number sequence number,

With

Be respectively the maximum and the minimum value of dissolved oxygen DO setting value, K _{P_SO}And K _{I_SO}Expression PI controller control ratio and integral parameter, e _COD(k) be the error of the k time sample effluent COD concentration setting value and effluent COD concentration measured value, e _COD(k-1) be the error of (k-1) inferior sampling effluent COD concentration setting value and effluent COD concentration measured value, Δ e _COD(k) be the k time sampling effluent COD concentration error rate, COD _Out(k) be the k time sampling effluent COD concentration measured value, COD _{Out, sp}Be the effluent COD concentration setting value;

Said Human Simulating Intelligent Control Algorithm adopts production rule that expertise is described, and this rule list is shown:

IF(condition) THEN(action)；

Response according to the big young pathbreaker system of the error e (k) of dissolved oxygen concentration setting value and measured value is divided into four intervals, carries out staging treating, and the Different control strategy is taked in different intervals;

The positive and negative variation tendency of differentiating error current according to product e (k) the Δ e (k) of the error rate Δ e (k) of the error e (k) of dissolved oxygen concentration setting value and measured value and dissolved oxygen concentration setting value and measured value; Select P, I, D parameter for use and amplify system or rejection coefficient, calculate the frequency conversion output frequency of aeration link actuator at last;

Human Simulating Intelligent Control Algorithm is:

(1) when the Error Absolute Value of dissolved oxygen concentration setting value and measured value | e (k) | more than or equal to error upper limit e2, promptly | and e (k) | during >=e2, the illustrative system error is excessive, and the output valve of controller is by its maximum value output;

(2) when the Error Absolute Value of dissolved oxygen concentration setting value and measured value | e (k) | between error upper limit e2 and error lower limit e1, promptly e1＜| e (k) | during＜e2, consider reduction or do not use integral action that concrete control strategy also should be with reference to error change trend:

(21) if e (k) Δ e (k) >=0, illustrative system output positive deviation setting value, error just change towards the direction that absolute value increases, or error is a certain constant value, remain unchanged, and implement the control of the variation tendency of rapid torsional error absolute value, the controller output valve is:

Δu(k)＝k ₁{K _P1Δe(k)+K _I1e(k)+K _D1[e(k)-2e(k-1)+e(k-2)]} (14)

Δ u (k) is the k time sampling system output increment in the formula, k ₁Be amplification coefficient, K _P1, K _I1And K _D1Be respectively PID controller ratio, integration and differential parameter;

(22) if e (k) Δ e (k)＜0, setting value just is being partial in illustrative system output, error just changes towards the direction that absolute value is reducing, and implements to let system under the assistance of inertia, get back to the control of stable state, the controller output valve is:

Δu(k)＝k ₂{K _P1Δe(k)+K _I1e(k)+K _D1[e(k)-2e(k-1)+e(k-2)]} (15)

In the formula, k ₂The expression rejection coefficient;

(3) when the Error Absolute Value of dissolved oxygen concentration setting value and measured value | e (k) | between error dead band d and error lower limit e1; Be d＜| e (k) | during≤e1, systematic error is less, adds integration; Reducing steady-state error, and the reference error variation tendency is confirmed concrete control strategy:

(31) if e (k) Δ e (k) >=0, the controller output valve is:

Δu(k)＝k ₁{K _P2Δe(k)+K _I2e(k)+K _D2[e(k)-2e(k-1)+e(k-2)]} (16)

K in the formula _P2, K _I2And K _D2Be respectively PID controller ratio, integration and differential parameter;

(32) if e (k) Δ e (k)＜0, the controller output valve is:

Δu(k)＝k ₂{K _P2Δe(k)+K _I2e(k)+K _D2[e(k)-2e(k-1)+e(k-2)]} (17)

(4) when the Error Absolute Value of dissolved oxygen concentration setting value and measured value | e (k) | less than error skip distance d, i.e. e | during≤d, ignore systematic error, the controller output valve is 0;

It is said that control comprises sludge concentration setting value feedforward-feedback control and sludge concentration control loop to the enforcement of the sludge concentration in aeration tank feedforward-tandem, wherein:

Sludge concentration setting value feedforward-feedback control: as feed-forward signal, as feedback signal, feedforward control adopts proportional control with water outlet COD with the entry organic loading, and FEEDBACK CONTROL adopts PI control;

Sludge concentration control loop: as feedback signal, realize through control mud capacity of returns with sludge concentration measured value in the aeration tank;

Said feedforward control adopts proportional control, obtains through following formula:

MLSS _sp，ff(k)＝k _{ff_MLSS}·Food _avg(k) (18)

MLSS in the formula _{Sp, ff}(k) be the k time sampling sludge concentration setting value feedforward value, K _{Ff_MLSS}The static feedforward of expression scale-up factor, Food _Avg(k) be 2 hours sliding averages of the k time sampling entry organic loading;

Feedback controller adopts PI control, shown in (19):

Wherein

ΔMLSS _sp(k)＝K _{P_MLSS}·e _{COD_avg}(k)+K _{I_MLSS}·Δe _{COD_avg}(k) (20)

e _COD(k)＝COD _out，sp-COD _out(k) (21)

Δe _COD(k)＝e _COD(k)-e _COD(k-1) (22)

The final setting value of sludge concentration is suc as formula shown in (23):

MLSS _SP(k)＝MLSS _{SP_ff}(k)+MLSS _{SP_fb}(k) (23)

MLSS in the formula _{Sp, fb}(k) be the k time sampling sludge concentration setting value value of feedback, MLSS _{Sp, fb}(k-1) be (k-1) inferior sampling sludge concentration setting value value of feedback, Δ MLSS _{Sp, fb}(k) be the k time sampling sludge concentration setting value value of feedback added value, With

Be respectively the maximum and the minimum value of sludge concentration setting value, K _{P_MLSS}And K _{I_MLSS}Be respectively PI controller control ratio and integral parameter, e _COD(k) be the error of the k time sample effluent COD concentration setting value and effluent COD concentration measured value, e _COD(k-1) be the error of (k-1) inferior sampling effluent COD concentration setting value and effluent COD concentration measured value, Δ e _COD(k) be the k time sampling effluent COD concentration error rate, COD _Out(k) be the k time sampling effluent COD concentration measured value, COD _{Out, sp}Be effluent COD concentration setting value, MLSS _Sp(k) be the k time sampling sludge concentration setting value;

Said sludge concentration control loop algorithm is:

Wherein:

ΔQ _r(k)＝K _{P_Qr}·e _MLSS(k)+K _{I_Qr}·Δe _MLSS(k) (28)

e _MLSS(k)＝MLSS _sp-MLSS(k) (29)

Δe _MLSS(k)＝e _MLSS(k)-e _MLSS(k-1) (30)

Q in the formula _r(k) be the k time sampling mud capacity of returns measured value, Δ Q _r(k) be the k time sampling mud capacity of returns measured value added value,

With Maximum and the minimum value of representing the mud capacity of returns respectively, usually rule of thumb

Be 3-5 discharge doubly, Be 0.5-1.5 discharge doubly; K _{P_Qr}And K _{I_Qr}Expression PI controller control ratio and integral parameter, e _MLSS(k) be the error of the k time sample sludge concentration setting value and sludge concentration measured value, e _MLSS(k-1) be the error of (k-1) inferior sampling sludge concentration setting value and sludge concentration measured value, Δ e _MLSS(k) be the k time sampling sludge concentration error rate, MLSS (k) is the k time sampling sludge concentration measured value, MLSS _SpBe the sludge concentration setting value.

Images