CN104156012B - A kind of real-time control method for improving the mixed concentration stable state accuracy of medicament-mixing device - Google Patents

Abstract

A kind of real-time control method for improving the mixed concentration stable state accuracy of medicament-mixing device of the present invention, the method 4 steps specific as follows：Step one：The one order inertia for obtaining medicinal liquid flow and water-carrying capacity transmission function postpones expression formula；Step 2：Derive the relational expression of controlled device steady oscillation amplitude under PI (proportional integral) controller action；Step 3：Determine PI controller parameters Kc and τ_iTo the Influencing Mechanism of steady oscillation amplitude a；Step 4：Regulation PI controller parameters, to control system controlled volume --- the steady oscillation amplitude of medicinal liquid flow and water-carrying capacity is adjusted.It is present inventive concept science, ingenious, by experimental verification, self adaptation and real-time, method is improved for the accurate control process for mixing concentration provides a kind of stable state accuracy easy to operate, quick, inexpensive, and accurate for actual industrial control system controls to provide a new approach.

Description

A kind of real-time control method for improving the mixed concentration stable state accuracy of medicament-mixing device

(1) technical field

The present invention relates to a kind of real-time control method for improving the mixed concentration stable state accuracy of medicament-mixing device.The method can be realized The steady state controling precision of mixed concentration is improved by dynamic adjustment controller parameter.Belong to intelligent equipment and control field.

(2) background technology

It is online that medicine-chest and water tank are separated the characteristics of mixed medicine, by mixing medicine during work compared with traditional premix prescription formula Device mixes medicine online.Online medicament-mixing device can realize that pesticide concentration changes on demand according to the order of severity of pest and disease damage, in disease pest Liquor strength is improved in evil critical regions, and liquor strength is reduced in the lighter region of pest and disease damage, reaches and is mixed on demand according to pest and disease damage degree Medicine.

Online mixed medicine is currently to improve one of agricultural chemicals utilization rate, the most effective means for reducing environmental pollution, is also to improve agriculture One of most effective means of medicine safety in utilization.The dynamic detection and stability contorting of liquid medicine mixing ratio are that online medicament-mixing device ensures The uniform and stable key of spray concentration, it is contemplated that liquid medicine mixing ratio is general 1 during spraying unit operation:300~1:500 it Between, therefore, the low discharge high precision test to agricultural chemicals is that online medicament-mixing device is applied to during actual job and must solve One key problem, especially in the market there is not yet can be to the volume flowmeter of agricultural chemicals low discharge detection, in some sections Though profit mass flowmenter difficult to understand can detect that high-precision sensor price is very high, therefore must seek at present to agricultural chemicals flow Seek low cost, the high-precision measuring means of agricultural chemicals.

In the prior art, applicant does not have found that the stable state for improving mixed concentration by dynamic adjustment controller parameter is controlled The medicament-mixing device of precision, if can not increase any hardware cost and do not changed hardware by dynamic adjustment controller parameter Steady state controling precision is improved under structure situation, with certain realistic meaning.The existing patent of analogous technical is retrieved Shi Faxian, the Chinese patent 1 of Application No. 200510041334.4《A kind of method that humidifier medicine is separated from water》Disclose , be stored in graduated container respectively for agricultural chemicals and water in advance in the method by the method that spraying machine Pesticides and water are separate In, accurate measurement is carried out to agricultural chemicals and water.Patent 2 (Application No. 200810027898) discloses one kind《Variable rate spray is automatic Medicament-mixing device》, the device one or more agricultural chemicals of consideration is individually or simultaneously sprayed, the situation of spray concentration real-time change；Patent 3 (Application No. 201010132393.3) is disclosed《A kind of automatic medicine-mixing device》, the device include stream portion and spire Point, it is characterized in：Agricultural chemicals and water separate individually storage, and realize the automatic fast of agricultural chemicals and water using negative pressure and turbulent closure scheme principle Speed is uniform to be mixed, but the device cannot be realized to medicine and water accurate measurement, it is impossible to realize the precise control to liquid medicine mixing ratio, and Liquid medicine mixing ratio is directly connected to the important indicator once to prevention and control of plant diseases, pest control effect in actual spray process, is directly connected to The realization of spray effect, when actual liquid medicine mixing ratio is less than required mixing ratio, liquor strength is too small, does not reach and prevents and treats disease pest Evil purpose, conversely, liquor strength is excessive, not only causes to damage to crop, can also cause environmental pollution.(the Application No. of patent 4 201210585002.2) disclose《A kind of vehicular medicament-mixing device》, it is main to consider that liquid and water are stored separately and realize online Uniform quick mixing, but still liquid medicine mixing ratio can not be controlled；Patent 5 (Application No. 99228118.0) discloses a kind of utilization Malleation and principle of negative pressure carry out liquid medicine mixing《Medicament-mixing device》, the device is mainly including be filled with water bottle, medicinal fluid bag and mixed liquid group Into, water injection hole and imbibing hole are provided with medicine mixer, the agricultural chemicals in medicinal fluid bag is contained in during work by necking down on medicine mixer Negative pressure is produced during part, water injection hole produces malleation, cause that agricultural chemicals enters medicament-mixing device from medicinal fluid bag under positive/negative-pressure collective effect Realize medicine, water mixing.Patent 6 (Application No. 201310670806.7) discloses one kind《Equipment for plant protection pesticide spraying system dynamically compares Example medicament-mixing device》, it is considered to using the mixed prescription formula of the secondary mixing of liquid medicine, will medicine and water mixed for the first time in the first case that mixes, so First mixed liquid and water are sucked to spray pipeline and carry out secondary mixing and (are difficult to accurately control liquid medicine to mix by the mixed Teat pipette of passing ratio afterwards Than).Patent 7 (Application No. 200920202638.8) discloses one kind《Mechanical creepage-type chemical mixing device used for atomizing machine》, the dress Put and drive plunger pump crankcase using same line shaft with the pressure roller of the roller shape being arranged on cam, make plunger displacement pump and wriggling The flow of pipe synchronously changes, it is intended to realize fixed medicine more mixed than dynamic, but the device may be only available for fixing the situation of liquid medicine mixing ratio, Liquid medicine mixing ratio ratio can not be changed.Water-carrying capacity and pressure are controlled by overflow valve mostly in above-mentioned patent, are led to Cross high-precision flow sensing device and high-precision control valve carries out observing and controlling to medicinal liquid flow, by controlling water-carrying capacity and medicinal liquid flow Mixed concentration is controlled, applicant does not have found that the stable state for improving mixed concentration by dynamic adjustment controller parameter controls essence The medicament-mixing device of degree.

Water-carrying capacity and the accurate control of medicinal liquid flow are needed exist for by corresponding high-precision sensor and held in medicament-mixing device Row mechanism, high-precision sensor and executing agency's one side price are high, on the other hand no matter how high-precision sensor and execution Mechanism, because actual control system is influenceed so that each controlled volume is reaching by executing agency and sensor dead band and resolution ratio Can not accurately stablize in setting value during stable state, but be fluctuated with certain amplitude some cycles near equilbrium position, each The steady oscillation amplitude of controlled volume, cycle of oscillation and this 3 parameters of oscillation phase are simultaneously differed, this 3 parameter fluctuation situations Difference can produce different steady state controling precisions, if making the steady oscillation of each controlled volume by dynamic adjustment controller parameter Amplitude reduction, so that steady state controling precision higher is obtained, it is this not increase any hardware cost and do not changing hardware configuration In the case of make the method that steady state controling precision is improved, with certain realistic meaning.By mixed concentration computing formula：

In formula (1), D is mixed concentration；Q_pAnd Q_wRespectively medicinal liquid flow and water-carrying capacity, mL/min.

From formula (1), the control for mixing concentration is its precision by the control realization to medicinal liquid flow and water-carrying capacity Depending on Q_pAnd Q_wControl accuracy.However, due to the limitation of executing agency and measurement sensor resolution ratio, causing reach steady During state poised state, Q_pAnd Q_wFluctuated with some cycles in equilbrium position, if one of them is in positive half period in two flows, and it is another One is in negative half-cycle, so as to cause mixed concentration error to increase.If reach homeostasis, gas temperature and pressure with Same cycle, the form of same-phase vibrate, then Q_p/Q_wChange smaller, i.e. Q_pAnd Q_wSteady-sxtate wave motion is smaller on mixed concentration influence, such as Fruit can be such that medicinal liquid flow and water-carrying capacity is vibrated according to the form of same cycle, same-phase by certain method, so can make to mix The steady state controling precision of concentration is improved.

In order to make up the deficiencies in the prior art, the present invention proposes a kind of reality for improving the mixed concentration stable state accuracy of medicament-mixing device Period control method, it is intended to when mixed concentration control system reaches homeostasis, made by the Self-tuning System to controller parameter by Control amount Q_pAnd Q_wVibrated in the form of same cycle, same-phase, so as to improve the control accuracy of mixed concentration.One kind improves mixed medicine dress The characteristics of real-time control method for putting mixed concentration stable state accuracy is：When system reaches homeostasis, by controller Parameter carries out on-line tuning, so as to mix the control accuracy of concentration in improving control system.

(3) content of the invention

1st, purpose：Limited so that each quilt by executing agency and sensor dead band and resolution ratio for actual control system Control amount can not accurately be stablized in setting value when stable state is reached, but with certain amplitude some cycles near equilbrium position Fluctuation, the invention provides a kind of real-time control method for improving the mixed concentration stable state accuracy of medicament-mixing device, it appoints not increasing What hardware cost and not change and make controlled volume steady oscillation width by the Self-tuning System to controller parameter in the case of hardware configuration Degree reduces, so that the steady state controling precision of mixed concentration is improved, to need the field of the mixed concentration measurement of high accuracy and control Close and provide low cost, effective autocontrol method.

2nd, technical scheme：

A kind of real-time control method for improving the mixed concentration stable state accuracy of medicament-mixing device, the method is comprised the following steps that：

Step one：It is input quantity to obtain respectively to be input to the voltage or current signal of medicinal liquid flow regulating valve, to enter The medicinal liquid flow of spraying machine main pipeline is the medicinal liquid flow control system and voltage or electricity to be input to water overflow valve of output quantity Stream signal is input quantity, is the one of the water flow control system transmission function of output quantity with the water-carrying capacity for entering spraying machine main pipeline Rank inertial delay expression formula.

Water and liquid are stored in water tank and medicine-chest respectively in advance, mixed concentration and spray amount are set by controller, Under tractor motive power output shaft effect, water is pumped into the water pump being connected with tractor motive power output shaft the main pipeline of spraying machine, In the timing of pump rotary speed one, the overflow of water overflow valve is controlled to the voltage signal of water overflow valve by controller control input Amount, so as to realize the control of the water-carrying capacity to entering spraying machine main pipeline；Meanwhile, start chemicals feed pump and treat its stabilization one After determining rotating speed, precise control is carried out to the medical abortion amount for entering main pipeline by the aperture for changing medicinal liquid flow regulating valve, into spray The water and liquid of mist owner's pipeline after main pipeline uniformly mixes, by the high-speed switch electromagnetic valve and jet head sets in each spray boom area Close and spray.It is respectively input quantity to be input to the voltage or current signal of water overflow valve, to enter the current of spraying machine main pipeline It is output quantity to measure, and sets up the Mathematical Modeling of water flow control system；Respectively being input to the voltage or electricity of medicinal liquid flow regulating valve Stream signal is input quantity, is output quantity with the medicinal liquid flow for entering spraying machine main pipeline, sets up the number of medicinal liquid flow control system Learn model.

Medicinal liquid flow and water flow control system can be expressed as one order inertia delay system or by model order reducing method letter One order inertia delay system is turned to, single order time delay Inertial Processing transmission function is represented by：

Wherein, k is multiplication factor, and τ is time constant, and θ is system delay.

Tested by step response, it may be determined that 3 values of parameter of k, τ and θ in (2), that is, obtain the single order of controlled system Inertial delay system transter expression formula.

Step 2：Derive the relational expression of controlled device steady oscillation amplitude under PI controller actions.

(1) derivation of the expression formula of steady oscillation amplitude

It is restricted or the feedback control of low input resolution ratio can causes limit cycle.A simple generation to being input into resolution ratio Table is exactly using an input being quantized, as shown in Figure 2.Quantizer is output as

u_q=qround (u/q) (3)

Wherein, u and u_qThe respectively input and output of quantizer；Q is quantization step, and quantizer represents limited holding here Row mechanism resolution ratio；Round is bracket function.

Fig. 3 is the reponse system with quantizer, and wherein G (s) is control object or control process transmission function, and K (s) is Controller, y and r are respectively the output of process and reference input, and u is controlled volume.Because actuator low resolution result in staged Input so that control object export y with amplitude a equilbrium position shake, wherein, a is the total width from trough to crest Degree (such as Fig. 4).

For the reponse system containing quantizer as shown in Figure 3.If with the presence of integral action in controller, limit cycle It is inevitable.

During stable state, the average value for exporting y is equal to reference input r, i.e. y_ss=r,Corresponding input

Wherein G (0) is the steady-state gain of process, due to the presence of measurement noise, generally, u_ssCan not possibly just etc. In quantizer rank q_i, then quantizer output u_qIt is inevitable at least to be shaken between two quantizer levels.

It is assumed that process periodicity as shown in Figure 5 is persistently input into the excitation of u (t) signals.The signal is by without hysteresis loop Relay is produced.Wherein q₁、q₂It is limiting value, t₁It is u_qKeep q₁Time, T is cycle of oscillation, and the value of T is T=t₁+t₂, should Signal is represented by a series of time delays on frequency domain.Without loss of generality, it is assumed that q₂=0, q₁=q, then：

The process that this signal function to formula (2) is represented, output signal occurs concussion.The maximum or minimum value of concussion are deposited It is to gather, minimum or maximum are present in set。 In the range of maximum be

Time domain is transformed into obtain

So, maximum or minimum value is

Therefore, maximum or minimum value is extended to

I.e.

As n → ∞, (e^-T/τ)ⁿ→ 0, the finite sum of formula (9) is

Then

Similarly, can derive in θ+t₁Maximum in the range of+T ＜ t ＜ θ+2T：

Shock range a=y_ext1-y_ext2, i.e.,

A depends on t in formula (14)₁And T, their value is must determine for this.

(2)t₁With the derivation of T

From the figure 3, it may be seen that

U (s)=K (s) [r (s)-y (s)] (15)

WhereinR (s)=r₀/ s, y (s)=G (s) u_q(s),r₀It is reference input r Amplitude.

Wherein, t₀It is the transient process time.

Consider PI controllers, formula (16) is substituted into (15), and be converted into forms of time and space, then have

Work as θ<t<t₀During+θ,

In interval t₀+θ<t<t₀+t₁+ θ,

Similarly, for interval t₀+t₁+θ<t<t₀+t₁+t₂+ θ,

In above formula,

Wherein

Wherein round is bracket function.

By t=t₀, t=t₀+t₁, t=t₀+t₁+t₂Formula (18) is substituted into respectively, and (19), (20) obtain：

Due to u (t₀)=u (t₀+t₁)=u (t₀+t₁+t₂), convolution (22)~(24) obtain：

Especially, as τ=τ_IWhen, formula (18)~(20) can be changed into respectively：

Due to u (t₀)=u (t₀+t₁)=u (t₀+t₁+t₂), by t=t₀, t=t₀+t₁+t₂, t=t₀+t₁Formula is substituted into respectively (27), (28), (29) simultaneously solve：

As τ ≠ τ_IWhen, numerical solution is carried out to formula (22)~(24), can obtain t₁, t₂And T, then substitute into formula (14) Try to achieve controlled system steady oscillation amplitude a.

Step 3：Determine PI controller parameters K_cAnd τ_ITo the Influencing Mechanism of steady oscillation amplitude a.

Determine K by testing and emulating_cAnd τ_IIt is by appropriate in steady-state process to the affecting laws of steady oscillation amplitude a Regulation K_cAnd τ_ISteady oscillation amplitude a is reduced, steady state controling precision is improved and is laid the foundation.

Step 4：Regulation PI controller parameters, are adjusted to control system controlled volume steady oscillation amplitude, and acquisition makes steady State is shaken control accuracy highest optimal control parameter.

3rd, advantage and effect：The present invention is had an advantageous effect in that：

(1) for actual industrial control system mixed concentration precision control process provide one kind it is easy to operate, quick, low The stable state accuracy of cost improves method, in the case of hardware cost is not increased, greatly improves controlled device control accuracy.

(2) it is real-time.The stable state accuracy adjusting method that the present invention is provided is real-time online adjustment, with very hard real time Property.

(3) adaptive ability is strong.The method be applicable under PI controller actions any one order inertia delay system (or The process of one order inertia delay system can be reduced to), do not rely on specific control object parameter.

(4) low cost.The raising of steady state controling precision does not have executing agency and sensing by changing higher resolution Device (high-resolution executing agency and sensor, price are high), therefore the present invention has functions that low cost.

(4) illustrate

Fig. 1 spraying machine medicament-mixing device structural representations；

The smooth signal quantization schematic diagrames of Fig. 2；

Fig. 3 contains the reponse system schematic diagram of quantizer；

Fig. 4 contains the reponse system response curve of quantizer；

Fig. 5 input signal curve maps；

A is with K for Fig. 6 oscillation amplitudes_cChange schematic diagram；

A is with τ for Fig. 7 oscillation amplitudes_IChange schematic diagram；

The symbol description of each figure is as follows：

1- water tanks；2- water pumps；3- water overflow valves；4- water flow sensors；5- atomisation pressure sensors；6- sprays flow biography Sensor；7- high-speed switch electromagnetic valves；8- shower nozzles；9- filters；10- check valves；11- differential pressure pick-ups；12- medicinal liquid flows are adjusted Valve；13- liquid proportional pressure control valves；14- chemicals feed pumps；15- medicine-chests；16- controllers；17- tractor motive power output shafts；18- Water pump input shaft speed probe；19- unit pace Hall sensors；20- car-mounted computers；

u：Quantizer is input into (i.e. controller output)；u_q：Quantizer is exported；r：Reference input；y：Controlled volume is exported；K：Control Device transmission function processed；G：Controlled device transmission function；a：Controlled device steady oscillation amplitude；T：Cycle of oscillation；q₁And q₂To quantify Two adjacent quantization levels of device；t₁：The u within a cycle of oscillation_qIt is maintained at q₂Time；t₂：The u within a cycle of oscillation_q It is maintained at q₁Time；t₀：The transient process time；K_c：PI controller gain coefficients；τ_I：PI controller integration time constants.

(5) specific embodiment

Step one：It is input quantity to obtain respectively to be input to the voltage or current signal of medicinal liquid flow regulating valve, to enter The medicinal liquid flow of spraying machine main pipeline is the medicinal liquid flow control system and voltage or electricity to be input to water overflow valve of output quantity Stream signal is input quantity, is the one of the water flow control system transmission function of output quantity with the water-carrying capacity for entering spraying machine main pipeline Rank inertial delay expression formula.

Water and liquid are stored in water tank (1) and medicine-chest (15) respectively in advance, it is dense to set mixed medicine by controller (16) Degree and spray amount, in the presence of tractor motive power output shaft (17), the water pump (2) being connected with tractor motive power output shaft is by water The main pipeline of spraying machine is pumped into, in the timing of pump rotary speed one, by the voltage signal of controller control input to water overflow valve (3) To control the spillway discharge of water overflow valve, so as to realize the control of the water-carrying capacity to entering spraying machine main pipeline；Meanwhile, start liquid Injection pump (14) and after its stabilization in certain rotating speed after, by changing the aperture of medicinal liquid flow regulating valve (12) to entering main pipeline Medical abortion amount carry out precise control, the water and liquid into spraying machine main pipeline after main pipeline uniformly mixes, liquid medicine mixed liquor Sprayed by the combination of the high-speed switch electromagnetic valve (7) and shower nozzle (8) in each spray boom area.Respectively being input to water overflow valve (3) Voltage or current signal be input quantity, with enter spraying machine main pipeline water-carrying capacity be output quantity, set up water-carrying capacity control system The Mathematical Modeling of system；It is respectively input quantity to be input to the voltage or current signal of medicinal liquid flow regulating valve (12), is sprayed with entering The medicinal liquid flow of mist owner's pipeline is output quantity, sets up the Mathematical Modeling of medicinal liquid flow control system.

Medicinal liquid flow and water flow control system can be expressed as one order inertia delay system or by model order reducing method One order inertia delay system is reduced to, single order time delay Inertial Processing transmission function is represented by：

Wherein, k is multiplication factor, and τ is time constant, and θ is system delay.

Tested by step response, it may be determined that 3 values of parameter of k, τ and θ in (2), that is, obtain the single order of controlled system Inertial delay system transter expression formula.

Step 2：Derive the relational expression of controlled device steady oscillation amplitude under PI controller actions.

(1) derivation of the expression formula of steady oscillation amplitude

It is restricted or the feedback control of low input resolution ratio can causes limit cycle.A simple generation to being input into resolution ratio Table is exactly using an input being quantized, as shown in Figure 2.Quantizer is output as

u_q=qround (u/q) (3)

Wherein, u and u_qThe respectively input and output of quantizer；Q is quantization step, and quantizer represents limited holding here Row mechanism resolution ratio；Round is bracket function.

Fig. 3 is the reponse system with quantizer, and wherein G (s) is control object or control process transmission function, and K (s) is Controller, y and r are respectively the output of process and reference input, and u is controlled volume.Because actuator low resolution result in staged Input so that control object export y with amplitude a equilbrium position shake, wherein, a is the total width from trough to crest Degree (such as Fig. 4).

For the reponse system containing quantizer as shown in Figure 3.If with the presence of integral action in controller, limit cycle It is inevitable.

During stable state, the average value for exporting y is equal to reference input r, i.e. y_ss=r,Corresponding input

Wherein G (0) is the steady-state gain of process, due to the presence of measurement noise, generally, u_ssCan not possibly just etc. In quantizer rank q_i, then quantizer output u_qIt is inevitable at least to be shaken between two quantizer levels.

It is assumed that process periodicity as shown in Figure 5 is persistently input into the excitation of u (t) signals.The signal is by without hysteresis loop Relay is produced.Wherein q₁、q₂It is limiting value, t₁It is u_qKeep q₁Time, T is cycle of oscillation, and the value of T is T=t₁+t₂, should Signal is represented by a series of time delays on frequency domain.Without loss of generality, it is assumed that q₂=0, q₁=q, then：

The process that this signal function to formula (2) is represented, output signal occurs concussion.The maximum or minimum value of concussion are deposited It is to gatherMinimum or maximum is present in set。 In the range of maximum be

Time domain is transformed into obtain

So, maximum or minimum value is

Therefore, maximum or minimum value is extended to

I.e.

As n → ∞, (e^-T/τ)ⁿ→ 0, the finite sum of formula (9) is

Then

Similarly, can derive in θ+t₁Maximum in the range of+T ＜ t ＜ θ+2T：

Shock range a=y_ext1-y_ext2, i.e.,

A depends on t in formula (14)₁And T, their value is must determine for this.

(2)t₁With the derivation of T

From the figure 3, it may be seen that

U (s)=K (s) [r (s)-y (s)] (15)

WhereinR (s)=r₀/ s, y (s)=G (s) u_q(s),r₀It is reference input r Amplitude.

Wherein, t₀It is the transient process time.

Consider PI controllers, formula (16) is substituted into (15), and be converted into forms of time and space, then have

Work as θ<t<t₀During+θ,

In interval t₀+θ<t<t₀+t₁+ θ,

Similarly, for interval t₀+t₁+θ<t<t₀+t₁+t₂+ θ,

In above formula,

Wherein

Wherein round is bracket function.

By t=t₀, t=t₀+t₁, t=t₀+t₁+t₂Formula (18) is substituted into respectively, and (19), (20) obtain：

Due to u (t₀)=u (t₀+t₁)=u (t₀+t₁+t₂), convolution (22)~(24) obtain：

Especially, as τ=τ_IWhen, formula (18)~(20) can be changed into respectively：

Due to u (t₀)=u (t₀+t₁)=u (t₀+t₁+t₂), by t=t₀, t=t₀+t₁+t₂, t=t₀+t₁Formula is substituted into respectively (27), (28), (29) simultaneously solve：

As τ ≠ τ_IWhen, numerical solution is carried out to formula (22)~(24), can obtain t₁, t₂And T, then substitute into formula (14) Try to achieve controlled system steady oscillation amplitude a.

Step 3：Determine PI controller parameters K_cAnd τ_ITo the Influencing Mechanism of steady oscillation amplitude.

Shown by experiment and simulation result：1) such as Fig. 6, K_cAdjustment effect very little on oscillation amplitude a or without influence, but Its influence is by the transient performance of control system, therefore K_cIt is not used in regulation oscillation amplitude a；2) it is steady controlled system is met such as Fig. 7 Under the premise of qualitative, with τ_IIncrease, oscillation amplitude a reduces, but τ_IAfter increasing to a certain extent, with τ_IIncrease, vibrate width Degree a reduces amplitude and diminishes, and shows τ_IAdjustment effect to a is also limited.It can thus be concluded that going out following regulation rule：Increase τ_IMake Oscillation amplitude a reduces, and realizes the on-line tuning to a, works as τ_IWhen increase is not obvious to a, adjustment process terminates.

Step 4：PI controller initial parameters are determined using Stable Boundary, after object to be controlled reaches stable state, by adjusting Whole K_cAnd τ_ITo control system controlled volume --- the steady oscillation amplitude of medicinal liquid flow and water-carrying capacity is adjusted, so that stable state is shaken Swing amplitude minimum.

Facts have proved：A kind of real-time control method for improving the mixed concentration stable state accuracy of medicament-mixing device of the present invention is not changing For accurate control process provides a kind of stable state accuracy raising side easy to operate, quick, inexpensive in the case of any hardware Method, for the accurate control of actual industrial control system provides a new approach.

Claims (1)

1. It is 1. a kind of to improve the real-time control method that medicament-mixing device mixes concentration stable state accuracy, it is characterised in that：The method is specifically walked It is rapid as follows：

   Step one：It is input quantity to obtain respectively to be input to the voltage or current signal of medicinal liquid flow regulating valve, is sprayed with entering The medicinal liquid flow of owner's pipeline is believed for the medicinal liquid flow control system of output quantity and to be input to the voltage or electric current of water overflow valve Number be input quantity, with enter spraying machine main pipeline water-carrying capacity as the water flow control system transmission function of output quantity single order be used to Property postpone expression formula；

   It is input quantity with the voltage or current signal that are input to medicinal liquid flow regulating valve in actual control system, is sprayed with entering The medicinal liquid flow of owner's pipeline is believed for the medicinal liquid flow control system of output quantity and to be input to the voltage or electric current of water overflow valve Number be input quantity, with enter spraying machine main pipeline water-carrying capacity as the water flow control system of output quantity can be expressed as single order be used to Property delay system is reduced to one order inertia delay system by model order reducing method, and single order time delay Inertial Processing transmission function can It is expressed as：

   $G\left(s\right)=\frac{{\mathrm{ke}}^{-\mathrm{\&theta;}s}}{\mathrm{\&tau;}s+1}---\left(2\right)$

   Wherein, k is multiplication factor, and τ is time constant, and θ is system delay；

   Tested by step response, it may be determined that 3 values of parameter of k, τ and θ in (2), that is, obtain the one order inertia of controlled system The transmission function expression formula of delay system；

   Step 2：Derive the relational expression of controlled device steady oscillation amplitude under PI controller actions；

   (1) derivation of the expression formula of steady oscillation amplitude

   It is restricted or the feedback control of low input resolution ratio can causes limit cycle；A simple representative to being input into resolution ratio is just It is that, using an input being quantized, quantizer is output as

   u_q=qround (u/q) (3)

   Wherein, u and u_qThe respectively input and output of quantizer；Q is quantization step, and quantizer represents limited execution machine here Structure resolution ratio；Round is bracket function；

   Reponse system with quantizer, wherein G (s) are control object or control process transmission function, and K (s) is controller, y The output of process and reference input are respectively with r, u is controlled volume；Because actuator low resolution result in stepped input, from And cause that control object output y is shaken with amplitude a in equilbrium position, wherein, a is the total amplitude from trough to crest；

   For the reponse system containing quantizer；If with the presence of integral action in controller, limit cycle is inevitable；

   During stable state, the average value for exporting y is equal to reference input r, i.e. y_ss=r,Corresponding input

   $u_{ss}=\frac{y_{ss}}{G\left(0\right)}=\frac{r}{G\left(0\right)}---\left(4\right)$

   Wherein G (0) is the steady-state gain of process, due to the presence of measurement noise, generally, u_ssThe amount of may not be exactly equal to Change device rank q_i, then quantizer output u_qIt is inevitable at least to be shaken between two quantizer levels；

   It is assumed that the process is by periodically persistently input u (t) signal is encouraged, the signal is produced by the relay without hysteresis loop, its Middle q₁、q₂It is limiting value, t₁It is u_qKeep q₁Time, T is cycle of oscillation, and the value of T is T=t₁+t₂, the signal be represented by frequency A series of time delays on domain, without loss of generality, it is assumed that q₂=0, q₁=q, then：

   $u_q\left(s\right)=\frac{q}{s}(1-e^{-t_{1}s}+e^{-Ts}-e^{-(t_1+T)s}+e^{-2Ts}-e^{-(t_1+2T)s}+...)---\left(5\right)$

   The process that this signal function to formula (2) is represented, output signal occurs concussion；The maximum or minimum value of concussion are present in SetMinimum or maximum is present in setIn θ + T ＜ t ＜ θ+t₁Maximum is in+T range

   $y\left(s\right)=\frac{{\mathrm{ke}}^{-\mathrm{\&theta;}s}}{\mathrm{\&tau;}s+1}\frac{q}{s}(1-e^{-t_{1}s}+e^{-Ts})---\left(6\right)$

   Time domain is transformed into obtain

   Y (t)=kq (1-e^-(t-θ-T)/τ+e^{-(t-θ-t1)/τ}-e^-(t-θ)/τ) (7)

   So, maximum or minimum value is

   $y(t_1+T+\mathrm{\&theta;})=kq(1-e^{-t_1/\mathrm{\&tau;}}+e^{-T/\mathrm{\&tau;}}-e^{-(t_1+T)/\mathrm{\&tau;}})---\left(8\right)$

   Therefore, maximum or minimum value is extended to

   $y_{ext1}=kq(1-e^{-t_1/\mathrm{\&tau;}}+e^{-T/\mathrm{\&tau;}}-e^{-(t_1+T)/\mathrm{\&tau;}}+e^{-2T/\mathrm{\&tau;}}+...)---\left(9\right)$

   I.e.

   $y_{ext1}=kq(1-e^{-t_1/\mathrm{\&tau;}})(1+e^{-T/\mathrm{\&tau;}}+e^{-2T/\mathrm{\&tau;}}+e^{-3T/\mathrm{\&tau;}}...)---\left(10\right)$

   As n → ∞, (e^-T/τ)ⁿ→ 0, the finite sum of formula (9) is

   $\underset{n\&RightArrow;\mathrm{\&infin;}}{\lim }\underset{m=0}{\overset{n}{\&Sigma;}}{\left(e^{-T/\mathrm{\&tau;}}\right)}^m=\frac{1}{1-e^{-T/\mathrm{\&tau;}}}---\left(11\right)$

   Then

   $y_{ext1}=kq\frac{1-e^{-t_1/\mathrm{\&tau;}}}{1-e^{-T/\mathrm{\&tau;}}}---\left(12\right)$

   Similarly, can derive in θ+t₁Maximum in the range of+T ＜ t ＜ θ+2T：

   $y_{ext2}=-kq\frac{e^{-T/\mathrm{\&tau;}}(1-e^{-t_1/\mathrm{\&tau;}})}{1-e^{-T/\mathrm{\&tau;}}}---\left(13\right)$

   Shock range a=y_ext1-y_ext2, i.e.,

   $a=kq\frac{1-e^{-t_1/\mathrm{\&tau;}}+e^{-T/\mathrm{\&tau;}}-e^{-(T-t_1)/\mathrm{\&tau;}}}{1-e^{-T/\mathrm{\&tau;}}}---\left(14\right)$

   A depends on t in formula (14)₁And T, their value is must determine for this；

   (2)t₁With the derivation of T

   U (s)=K (s) [r (s)-y (s)] (15)

   WhereinR (s)=r₀/ s, y (s)=G (s) u_q(s),r₀It is the width of reference input r Value；

   $u_q\left(s\right)=\frac{q_2}{s}+\frac{q_1-q_2}{s}(e^{-t_{0}s}-e^{-(t_0+t_1)s})---\left(16\right)$

   Wherein, t₀It is the transient process time；

   Consider PI controllers, formula (16) is substituted into (15), and be converted into forms of time and space, then have

   $\begin{array}{c}u\left(t\right)=\frac{K_c}{{\mathrm{\&tau;}}_I}\{r_0(t+{\mathrm{\&tau;}}_I)-{\mathrm{kq}}_1\&lsqb;({\mathrm{\&tau;}}_I-\mathrm{\&tau;})(1-e^{-(t-\mathrm{\&theta;})/\mathrm{\&tau;}})+t-\mathrm{\&theta;}\&rsqb;+k(q_1-q_2)\&lsqb;({\mathrm{\&tau;}}_I-\mathrm{\&tau;})(1-e^{-(t-t_0-\mathrm{\&theta;})/\mathrm{\&tau;}})+t-t_0-\mathrm{\&theta;}\&rsqb;-\\ k(q_1-q_2)\&lsqb;({\mathrm{\&tau;}}_I-\mathrm{\&tau;})(1-e^{-(t-t_0-t_1-\mathrm{\&theta;})/\mathrm{\&tau;}})+t-t_0-t_1-\mathrm{\&theta;}\&rsqb;\end{array}---\left(17\right)$

   Work as θ<t<t₀During+θ,

   $u\left(t\right)=\frac{K_c}{{\mathrm{\&tau;}}_I}\{r_0(t+{\mathrm{\&tau;}}_I)-{\mathrm{kq}}_1\&lsqb;({\mathrm{\&tau;}}_I-\mathrm{\&tau;})(1-e^{-(t-\mathrm{\&theta;})/\mathrm{\&tau;}})+t-\mathrm{\&theta;}\&rsqb;\}---\left(18\right)$

   In interval t₀+θ<t<t₀+t₁+ θ,

   $\begin{array}{c}u\left(t\right)=\frac{K_c}{{\mathrm{\&tau;}}_I}\{r_0(t+{\mathrm{\&tau;}}_I)-{\mathrm{kq}}_1\&lsqb;({\mathrm{\&tau;}}_I-\mathrm{\&tau;})(1-e^{-(t-\mathrm{\&theta;})/\mathrm{\&tau;}})+t-\mathrm{\&theta;}\&rsqb;+\\ k(q_1-q_2)\&lsqb;({\mathrm{\&tau;}}_I-\mathrm{\&tau;})(1-e^{-(t-t_0-\mathrm{\&theta;})/\mathrm{\&tau;}})+t-t_0-\mathrm{\&theta;}\&rsqb;\}\end{array}---\left(19\right)$

   Similarly, for interval t₀+t₁+θ<t<t₀+t₁+t₂+ θ,

   $\begin{array}{c}u\left(t\right)=\frac{K_c}{{\mathrm{\&tau;}}_I}\{r_0(t+{\mathrm{\&tau;}}_I)-{\mathrm{kq}}_1\&lsqb;({\mathrm{\&tau;}}_I-\mathrm{\&tau;})(1-e^{-(t-\mathrm{\&theta;})/\mathrm{\&tau;}})+t-\mathrm{\&theta;}\&rsqb;+k(q_1-q_2)\\ \&lsqb;({\mathrm{\&tau;}}_I-\mathrm{\&tau;})(1-e^{-(t-t_0-\mathrm{\&theta;})/\mathrm{\&tau;}})+t-t_0-\mathrm{\&theta;}\&rsqb;-k(q_1-q_2)\&lsqb;({\mathrm{\&tau;}}_I-\mathrm{\&tau;})(1-e^{-(t-t_0-t_1-\mathrm{\&theta;})/\mathrm{\&tau;}})+t-t_0-t_1-\mathrm{\&theta;}\&rsqb;\}\end{array}---\left(20\right)$

   In above formula,

   Wherein

   $n=round\left(\frac{u\left(t\right)}{q_1-q_2}\right)---\left(21\right)$

   Wherein round is bracket function；

   By t=t₀, t=t₀+t₁, t=t₀+t₁+t₂Formula (18) is substituted into respectively, and (19), (20) obtain：

   $u\left(t_0\right)=\frac{K_c}{{\mathrm{\&tau;}}_I}\{r_0(t_0+{\mathrm{\&tau;}}_I)-{\mathrm{kq}}_1\&lsqb;({\mathrm{\&tau;}}_I-\mathrm{\&tau;})(1-e^{-(t-\mathrm{\&theta;})/\mathrm{\&tau;}})+t_0-\mathrm{\&theta;}\&rsqb;\}---\left(22\right)$

   $\begin{array}{c}u(t_0+t_1)=\frac{K_c}{{\mathrm{\&tau;}}_I}\{r_0(t_0+t_1+{\mathrm{\&tau;}}_I)-{\mathrm{kq}}_1\&lsqb;({\mathrm{\&tau;}}_I-\mathrm{\&tau;})(1-e^{-(t_0+t_1-\mathrm{\&theta;})/\mathrm{\&tau;}})+t_0+t_1-\mathrm{\&theta;}\&rsqb;+\\ k(q_1-q_2)\&lsqb;({\mathrm{\&tau;}}_I-\mathrm{\&tau;})(1-e^{-(t_1-\mathrm{\&theta;})/\mathrm{\&tau;}})+t_1-\mathrm{\&theta;}\&rsqb;\}\end{array}---\left(23\right)$

   $\begin{array}{c}u(t_0+t_1+t_2)=\frac{K_c}{{\mathrm{\&tau;}}_I}\{r_0(t+{\mathrm{\&tau;}}_I)-{\mathrm{kq}}_1\&lsqb;({\mathrm{\&tau;}}_I-\mathrm{\&tau;})(1-e^{-(t_0+t_1+t_2-\mathrm{\&theta;})/\mathrm{\&tau;}})+t_0+t_1+t_2-\mathrm{\&theta;}\&rsqb;+\\ k(q_1-q_2)\&lsqb;({\mathrm{\&tau;}}_I-\mathrm{\&tau;})(1-e^{-(t_1-\mathrm{\&theta;})/\mathrm{\&tau;}})+t_1-\mathrm{\&theta;}\&rsqb;-k(q_1-q_2)\&lsqb;({\mathrm{\&tau;}}_I-\mathrm{\&tau;})(1-e^{-(t_2-\mathrm{\&theta;})/\mathrm{\&tau;}})+t_2-\mathrm{\&theta;}\&rsqb;\}\end{array}---\left(24\right)$

   Due to u (t₀)=u (t₀+t₁)=u (t₀+t₁+t₂), convolution (22)~(24) obtain：

   $(r_0-{\mathrm{kq}}_2)t_1-k(q_1-q_2)\mathrm{\&theta;}-{\mathrm{kq}}_1({\mathrm{\&tau;}}_I-\mathrm{\&tau;})(e^{-(t_0-\mathrm{\&theta;})/\mathrm{\&tau;}}-e^{-(t_0+t_1-\mathrm{\&theta;})/\mathrm{\&tau;}})+k(q_1-q_2)({\mathrm{\&tau;}}_I-\mathrm{\&tau;})(1-e^{-(t_1-\mathrm{\&theta;})/\mathrm{\&tau;}})=0---\left(25\right)$

   $\begin{array}{c}(r_0-{\mathrm{kq}}_1)t_2+k(q_1-q_2)\mathrm{\&theta;}-{\mathrm{kq}}_1({\mathrm{\&tau;}}_I-\mathrm{\&tau;})(e^{-(t_0+t_1-\mathrm{\&theta;})/\mathrm{\&tau;}}-e^{-(t_0+t_1++t_2-\mathrm{\&theta;})/\mathrm{\&tau;}})+\\ k(q_1-q_2)({\mathrm{\&tau;}}_I-\mathrm{\&tau;})(e^{-(t_1-\mathrm{\&theta;})/\mathrm{\&tau;}}-e^{-(t_1+t_2-\mathrm{\&theta;})/\mathrm{\&tau;}})-k(q_1-q_2)({\mathrm{\&tau;}}_I-\mathrm{\&tau;})(1-e^{-(t_2-\mathrm{\&theta;})/\mathrm{\&tau;}})=0\end{array}---\left(26\right)$

   Especially, as τ=τ_IWhen, formula (18)~(20) can be changed into respectively：

   $u\left(t\right)=\frac{K_c}{{\mathrm{\&tau;}}_I}\{r_0(t+{\mathrm{\&tau;}}_I)-{\mathrm{kq}}_1(t-\mathrm{\&theta;})\}---\left(27\right)$

   $u\left(t\right)=\frac{K_c}{{\mathrm{\&tau;}}_I}\{n(t+{\mathrm{\&tau;}}_I)-{\mathrm{kq}}_1\&lsqb;(t-\mathrm{\&theta;})\&rsqb;+k(q_1-q_2)(t-t_0-\mathrm{\&theta;})\}---\left(28\right)$

   $u\left(t\right)=\frac{K_c}{{\mathrm{\&tau;}}_I}\{r_0(t+{\mathrm{\&tau;}}_I)-{\mathrm{kq}}_1(t-\mathrm{\&theta;})+k(q_1-q_2)(t-t_0-\mathrm{\&theta;})-k(q_1-q_2)(t-t_0-t_1-\mathrm{\&theta;})\}---\left(29\right)$

   Due to u (t₀)=u (t₀+t₁)=u (t₀+t₁+t₂), by t=t₀, t=t₀+t₁+t₂, t=t₀+t₁Formula (27) is substituted into respectively, (28), (29) and solve：

   $t_1=\frac{k(q_1-q_2)\mathrm{\&theta;}}{r_0-{\mathrm{kq}}_2}---\left(30\right)$

   $t_2=\frac{k(q_1-q_2)\mathrm{\&theta;}}{{\mathrm{kq}}_1-r_0}---\left(31\right)$

   $T=t_1+t_2=k(q_1-q_2)\mathrm{\&theta;}(\frac{1}{{\mathrm{kq}}_1-r_0}+\frac{1}{r_0-{\mathrm{kq}}_2})---\left(32\right)$

   As τ ≠ τ_IWhen, numerical solution is carried out to formula (22)~(24), can obtain t₁, t₂And T, then substituting into formula (14) can try to achieve Controlled system steady oscillation amplitude a；

   Step 3：Determine PI controller parameters K_cAnd τ_ITo the Influencing Mechanism of steady oscillation amplitude a；

   Determine K by testing and emulating_cAnd τ_IIt is by appropriate regulation in steady-state process to the affecting laws of steady oscillation amplitude a K_cAnd τ_ISteady oscillation amplitude a is reduced, steady state controling precision is improved and is laid the foundation；

   Step 4：Regulation PI controller parameters, to control system controlled volume --- the steady oscillation width of medicinal liquid flow and water-carrying capacity Degree is adjusted.