CN102914945A - Distributed exposure dose control system and method - Google Patents

Abstract

The invention discloses a distributed exposure dose control system, which comprises a laser unit, a variable attenuator, a light path transmission unit, a light path guide unit and an energy sensor, wherein the laser unit generates a laser pulse, and then the energy sensor measures actually received laser pulse after the laser pulse passes through the variable attenuator, the light path transmission unit and the light path guide unit. The distributed exposure dose control system is characterized in that the laser unit, the variable attenuator, the light path transmission unit, the light path guide unit, and the energy sensor respectively adopt independent correction modules to obtain compensation factors of different units, obtain a pulse energy value after carrying out feed-forward compensation on the actually received laser pulse and related parameter according to calibrating factors of the different units, and carry out feedback compensation to obtain the optimal exposure dose after the pulse energy value is output by a uniform compensation control unit.

Description

A kind of distributed exposure dose control system and method

Technical field

The present invention relates to integrated circuit equipment manufacturing field, relate in particular to a kind of distributed exposure dose control system and method for lithographic equipment.

Background technology

Dosage control is the pith in the photoetching technique.Exposure dose refers to the luminous energy of the specific wavelength that photoresist absorbs on the silicon chip unit area in the exposure process, and namely certain a bit locates the integration of exposure light intensity to the time shutter on the silicon chip face: Wherein D is exposure dose; T is the time shutter; I is the function of time t for the exposure light intensity.Exposure dose directly affects the final performance index of litho machine, comprises critical size, and critical dimension uniformity, wafer finally produce into quality and production efficiency etc.So improve the performance of litho machine, will do meticulous control to the exposure dose of litho machine with regard to requiring.

Exposure dose control method before mainly contains following several:

The first, based on the heterogeneity characteristics of pulse, one by one pulse controlled method has been proposed.The method mainly is according to the pulse value that feeds back, and one by one compensation is carried out in each pulse.The dose fluctuations of last pulse of this algorithm can't be passed through this algorithm optimization.The second, the method for increase equipment and control loop is controlled the exposure dose precision in light path.The method mainly adopts two or more laser equipment, with a cover or a few cover light source control loop main exposure dosage control loop is compensated exposure, and this compensation is mainly carried out between the recurrent interval of main exposure.This method devices in system increases, and cost and complexity all increase greatly.Three, based on the exposure characteristics of pulse, introduce segmentation increment type PID control model, introduce slit umber of pulse N and moving average M and be used for smoothing processing, and control ratio, integration, differential coefficient, according to each impulsive measurement result in the single exposure, next pulse is compensated calculating.

The factor that exposure system can have influence on final exposure dose result has laser element, the optic path unit, and lighting module, light-beam position is measured (BST) unit, exposure energy sensor, variable attenuator module etc.And present exposure method is done closed loop compensation from the pulse situation that records at last merely.This compensation way is overly dependent upon the accuracy of sensor, and compensates from measurement result merely, compensates also not accurate enough, such as purpose be the compensation 3mJ, the actual effect of regulating out at last may only have 2.8mJ, and in addition, this compensation way also exists the shortcoming that relatively lags behind on the time.

Summary of the invention

For overcoming the defective that exists in the prior art, the invention provides a kind of distributed exposure dose control system and method, each subelement is carried out distributed control.

In order to realize the foregoing invention purpose, the present invention discloses a kind of distributed exposure dose control system, comprise laser element, variable attenuator, the optic path unit, light path guidance unit and energy sensor, this laser element produces a laser pulse through this variable attenuator, measure the actual laser pulse that receives by this energy sensor behind optic path unit and the light path guidance unit, it is characterized in that, this laser element, variable attenuator, the optic path unit, light path guidance unit and energy sensor adopt respectively independent calibration model to obtain the compensating factor of different units, and the laser pulse of this actual receptions and the parameter group calibration factor according to this different units is carried out obtaining a pulse energy value behind the feedforward compensation, and this pulse energy value is carried out feedback compensation with the acquisition exposure dose after by a unified compensation control module output.

Further, this independent calibration model comprises signal collection module, module parameter logging modle and output factor computing module.

Further, this signal collection module acquisition parameter and remove noise after be handed down to the module parameter logging modle.

Further, this signal collection module acquisition parameter and the step of removing noise comprise: acquisition parameter S1, and it is deposited among the corresponding buffer memory array ValN, after each the collection, V parameter ali-2 to the first two cycle in the array checks, when this parameter meets the following conditions simultaneously: (| Vali-2-(Vali-3+Vali-4)/2|＞| Vali-2|25%) ∩ (| Vali-2-(Vali-1+Vali)/2|＞| Vali-2|25%), think that then V parameter ali-2 is noise, in buffer memory, remove this parameter.

Further, this module parameter logging modle adopts buffer circle, the structure that each unit in this buffer circle forms for meeting this module parameter.

Further, this output factor computing module obtains operational parameter and weighing factor parameter from this module parameter logging modle, according to output parameter and the calibration factor after this operational parameter and weighing factor calculation of parameter and the input calibration, this operational parameter comprises absolute variable parameter, the remote effect result parameter directly affects result parameter.This absolute variable parameter eliminated measure to get the offset error of value and standard value.Set up the calibration curve of this remote effect result parameter, and try to achieve revised measurement result by tabling look-up, to eliminate the nonlinearity erron of measured value.On this directly affect result parameter by forward with oppositely connect respectively an identical survey sensor, and the measured value that the two draws is subtracted each other.

Further, this parameter group comprises: pulse energy Ep, time shutter Texp, optical maser wavelength WL, laser element operating voltage Vls,, laser element net cycle time Lls, laser element cell temperature Tls,, laser element trigger voltage HV, variable attenuator unit lens position Pva, variable attenuator unit net cycle time Lva, variable attenuator cell temperature Tva, transmission optical path unit slit width Wslit, transmission optical path unit objective aperture Dpo, transmission optical path unit eyeglass displacement Dza, transmission optical path unit net cycle time Ltran, beam direction unit light beam X-direction position Px, beam direction unit light beam Y-direction position Py, beam direction unit beam angle Pa, beam direction unit net cycle time Lbst, energy sensor operating voltage Vsr, energy sensor sample rate current Isr, energy sensor net cycle time Lsr.

Further, between this optimum exposure dosage and this laser pulse following formula: E is arranged _Set=(E _NomK _Laser+ B _Laser) K _VAK _TransferK _BSTK _Env, wherein Eset is laser pulse, Enom is optimum exposure dosage, K _LaserBe the laser gain factor, Kva is the variable attenuator gain factor, and Ktransfer is the transmission light path module gain factor, and Blaser is the drift of laser instrument dark current, and Kbst is the beam direction module gain factor, and Kenv is the energy sensor gain factor.

Further, this laser gain factor K _Laser, variable attenuator gain factor Kva, transmission light path module gain factor K transfer, laser instrument dark current drift Blaser, beam direction module gain factor K bst, energy sensor gain factor Kenv to obtain formula as follows:

Kenv∝fes(Ep，WL，Texp)，

Klaser∝flaser(Vls，Lls，Tls，HV)，

Kva∝fva(Pva，Lva，Tva，WL)，

Ktransfer∝ftran(Wslit，Dpo，Dza，Ltran，WL)，

Kbst∝fbst(Px，Py，Pa，Pbst)，

Ksensor∝fsensor(Vsr，Ist，Lst)，

Blaser∝Glaser(Vls，Lls，Tls)。

Further, this unified compensation control module is proofreaied and correct this pulse energy value through PI adjuster and overshoot adjuster after, after process HVEp computing module calculates the high-voltage value HV that need to issue laser instrument, this laser instrument triggers a laser pulse, this laser pulse feeds back to this PI adjuster and this overshoot adjuster after recording by this energy sensor module.

Further, the computing formula of the adjusted value of this PI adjuster is as follows: ${\mathrm{Factor}}_{\mathrm{PI}}=K_p\·\underset{k=i-N+1}{\overset{i-1}{\mathrm{\Σ}}}(\frac{E_{\mathrm{act}}\left[k\right]}{K_{\mathrm{transfer}}\left[k\right]}-E_{\mathrm{nom}}),$ Wherein, Enom is optimum exposure dosage, and Kp is the gain factor of this PI adjuster, and Eact is the actual output energy value of this laser instrument, and Ktransfer is the transmission light path module gain factor.

Further, the computing formula of this overshoot adjusted value is as follows: ${\mathrm{Factor}}_{\mathrm{Overshoot}}=\frac{K_f}{M}\·\underset{l=i-M}{\overset{i-1}{\mathrm{\Σ}}}(\frac{E_{\mathrm{act}}\left[k\right]}{K_{\mathrm{transfer}}\left[k\right]}-E_{\mathrm{set}}\left(l\right)),$ Wherein, Kf is the gain factor of this overshoot adjuster, and Eact is the actual output energy value of this laser instrument, and Ktransfer is the transmission light path module gain factor.

Further, this variable attenuator comprises a life-span timing module and eyeglass deviation post sensor.

Further, but the calibration model of described attenuator unit be:

Kva＝(K _pva*Pva+K _tva*Tva+K _wl*WL+B _pva+B _tva+B _wl)*Flife(Lva)，

Kva is attenuation rate in the formula, WL optical maser wavelength, Pva is variable attenuator unit lens position, Tva is the variable attenuator cell temperature, but Bpva is the position compensation datum offset factor of attenuator, but Btva is the temperature compensation datum offset factor of attenuator, but Bwl is the wavelength compensation datum offset factor of attenuator, and Lva is variable attenuator unit net cycle time; Wherein Flife is the life-span decay factor of variable attenuator, and the acquisition formula of Flife is that (Lva/Lrva), Lrva is the nominal operation life-span of attenuator to Flife=exp.

The present invention discloses a kind of distributed exposure dose control method simultaneously, comprise: laser element produces a laser pulse through variable attenuator, measure the actual laser pulse that receives by energy sensor behind optic path unit and the light path guidance unit, this laser element, variable attenuator, the optic path unit, light path guidance unit and energy sensor adopt respectively independent calibration model to obtain the compensating factor of different units, the laser pulse of this actual reception and the correlation parameter calibration factor according to this different units is carried out obtaining a pulse energy value behind the feedforward compensation, this pulse energy value is carried out feedback compensation to obtain optimum exposure dosage after by a unified compensation control module output.

Further, this independent calibration model comprises signal collection module, module parameter logging modle and output factor computing module.

Further, this signal collection module acquisition parameter and remove noise after be handed down to the module parameter logging modle.

Further, this signal collection module acquisition parameter and the step of removing noise comprise: acquisition parameter S1, and it is deposited among the corresponding buffer memory array ValN, after each the collection, V parameter ali-2 to the first two cycle in the array checks, when this parameter meets the following conditions simultaneously: (| Vali-2-(Vali-3+Vali-4)/2|＞| Vali-2|25%) ∩ (| Vali-2-(Vali-1+Vali)/2|＞| Vali-2|25%), think that then V parameter ali-2 is noise, in buffer memory, remove this parameter.

Further, this module parameter logging modle adopts buffer circle, the structure that each unit in this buffer circle forms for meeting this module parameter.

Further, this output factor computing module obtains operational parameter and weighing factor parameter from this module parameter logging modle, according to output parameter and the calibration factor after this operational parameter and weighing factor calculation of parameter and the input calibration, this operational parameter comprises absolute variable parameter, the remote effect result parameter directly affects result parameter.This absolute variable parameter eliminated measure to get the offset error of value and standard value.Set up the calibration curve of this remote effect result parameter, and try to achieve revised measurement result by tabling look-up, to eliminate the nonlinearity erron of measured value.On this directly affect result parameter by forward with oppositely connect respectively an identical survey sensor, and the measured value that the two draws is subtracted each other.

Further, this correlation parameter comprises: pulse energy Ep, time shutter Texp, optical maser wavelength WL, laser element operating voltage Vls,, laser element net cycle time Lls, laser element cell temperature Tls,, laser element trigger voltage HV, variable attenuator unit lens position Pva, variable attenuator unit net cycle time Lva, variable attenuator cell temperature Tva, transmission optical path unit slit width Wslit, transmission optical path unit objective aperture Dpo, transmission optical path unit eyeglass displacement Dza, transmission optical path unit net cycle time Ltran, beam direction unit light beam X-direction position Px, beam direction unit light beam Y-direction position Py, beam direction unit beam angle Pa, beam direction unit net cycle time Lbst, energy sensor operating voltage Vsr, energy sensor sample rate current Isr, energy sensor net cycle time Lsr.

Further, between this optimum exposure dosage and this laser pulse following formula: E is arranged _Set=(E _NomK _Laser+ B _Laser) K _VAK _TransferK _BSTK _Env, wherein Eset is laser pulse, Enom is optimum exposure dosage, K _LaserBe the laser gain factor, Kva is the variable attenuator gain factor, and Ktransfer is the transmission light path module gain factor, and Blaser is the drift of laser instrument dark current, and Kbst is the beam direction module gain factor, and Kenv is the energy sensor gain factor.

Further, this laser gain factor K _Laser, variable attenuator gain factor Kva, transmission light path module gain factor K transfer, laser instrument dark current drift Blaser, beam direction module gain factor K bst, energy sensor gain factor Kenv to obtain formula as follows:

Kenv∝fes(Ep，WL，Texp)，

Klaser∝flaser(Vls，Lls，Tls，HV)，

Kva∝fva(Pva，Lva，Tva，WL)，

Ktransfer∝ftran(Wslit，Dpo，Dza，Ltran，WL)，

Kbst∝fbst(Px，Py，Pa，Pbst)，

Ksensor∝fsensor(Vsr，Ist，Lst)，

Blaser∝Glaser(Vls，Lls，Tls)。

Further, this unified compensation control module is proofreaied and correct this pulse energy value through PI adjuster and overshoot adjuster after, after process HVEp computing module calculates the high-voltage value HV that need to issue laser instrument, this laser instrument triggers a laser pulse, this laser pulse feeds back to this PI adjuster and this overshoot adjuster after recording by this energy sensor module.

Further, the computing formula of the adjusted value of this PI adjuster is as follows: ${\mathrm{Factor}}_{\mathrm{PI}}=K_p\·\underset{k=i-N+1}{\overset{i-1}{\mathrm{\Σ}}}(\frac{E_{\mathrm{act}}\left[k\right]}{K_{\mathrm{transfer}}\left[k\right]}-E_{\mathrm{nom}}),$ Wherein, Enom is optimum exposure dosage, and Kp is the gain factor of this PI adjuster, and Eact is the actual output energy value of this laser instrument, and Ktransfer is the transmission light path module gain factor.

Further, the computing formula of this overshoot adjusted value is as follows: ${\mathrm{Factor}}_{\mathrm{Overshoot}}=\frac{K_f}{M}\·\underset{l=i-M}{\overset{i-1}{\mathrm{\Σ}}}(\frac{E_{\mathrm{act}}\left[k\right]}{K_{\mathrm{transfer}}\left[k\right]}-E_{\mathrm{set}}\left(l\right)),$

Wherein, Kf is the gain factor of this overshoot adjuster, and Eact is the actual output energy value of this laser instrument, and Ktransfer is the transmission light path module gain factor.

Further, this variable attenuator comprises a life-span timing module and eyeglass deviation post sensor.

Further, the calibration model of this variable attenuator is:

Kva=(K _Pva* Pva+K _Tva* Tva+K _Wl* WL+B _Pva+ B _Tva+ B _Wl) * Flife (Lva), wherein Flife is the life-span decay factor of variable attenuator, Kva is attenuation rate, WL optical maser wavelength, Pva is variable attenuator unit lens position, Tva is the variable attenuator cell temperature, but Bpva is the position compensation datum offset factor of attenuator, but Btva is the temperature compensation datum offset factor of attenuator, but Bwl is the wavelength compensation datum offset factor of attenuator, and Lva is variable attenuator unit net cycle time.

Compared with prior art, technique effect of the present invention is embodied in:

The feedforward control of system has guaranteed that the compensation of single is more effective to efficient and the precision of dosage control.Integral body is controlled the model decoupling zero, by the distributed monitoring to each module, carrying the high-precision while and also guaranteed the real-time of system.The independent data model of each module has greatly made things convenient for the diagnostic analysis of system.Avoided violent effect of jitter once in a while to whole exposure process to the noise processed of individual pulse.It is convenient to realize, system does not need to increase extra special device and equipment, and can be used for the upgrading of new engine and second-hand equipment.Modularity is better, and native system has better upgrade maintenance.

Description of drawings

Can be by following detailed Description Of The Invention and appended graphic being further understood about the advantages and spirit of the present invention.

Fig. 1 is the structural representation of the first embodiment of the present invention;

Fig. 2 is exposure system control model synoptic diagram of the present invention;

Fig. 3 is unified compensation control block diagram of the present invention;

Fig. 4 is individual control module structural representation of the present invention;

Fig. 5 is parametric calibration curve map of the present invention;

But Fig. 6 is the control correction calculation illustraton of model of attenuation units of the present invention;

Fig. 7 is the structural representation of the second embodiment of the present invention;

Fig. 8 is the structural representation of the 3rd embodiment of the present invention.

Mainly be illustrated as follows:

Embodiment

Describe specific embodiments of the invention in detail below in conjunction with accompanying drawing.

The present invention is realization system of a kind of exposure dose control system and associated method, and the below describes the detailed case study on implementation that the present invention carries out dependency structure and step.

This method is based on the exposure dose control system, so at first briefly describe exposure dose control system.As shown in Figure 1, at first the pulse energy value of input is issued laser instrument 101, then laser instrument triggers laser pulse, can be through variable attenuator VA (102 and optic path unit 103 after laser pulse sends, arrive wafer 105, realize exposure, the actual value of exposure then can measure and inform system by the energy sensor 106 in the light path.Based on aforementioned exposure process, after system receives the required exposure dose that reaches, behind exposure dose and the unified compensation of correlation parameter input control module 207, the factor that draws during according to each model initialization, draw through this actual pulse energy value that should send behind the feedforward compensation, and this pulse is handed down to laser instrument triggers corresponding laser, after this trigger action, laser control model 201, variable attenuator model 202, optic path model of element 203, beam director model 204 is independent its corresponding administration module parameter information that gathers separately, and calculate the factor that next pulse energy triggers the needs compensation according to model, at last after energy sensor is measured this actual pulse energy, energy sensor control model 206 also needs the relevant parameter of energy sensor is gathered, and calculates the actual compensating factor of current pulse energy.All relevant factors are issued unified compensation control module 207 the most at last, calculate the triggering controlling value of next pulse energy.Whole system has formed a closed-loop control system that feedforward+feedback combines in the mode such as figure.

Next the mathematical model of system described.As shown in Figure 2, exposure system 107 is the part of empty frame, and the pass that pulse energy Eset and actual pulse ENERGY E act are set is Eact=Eset*Fes.Wherein Fes is the exposure system attenuation coefficient.To the parameter factors that the exposure result impacts, integral body has pulse energy Ep, time shutter Texp, optical maser wavelength WL.For each subelement, laser element 101 has operating voltage Vls, net cycle time Lls, cell temperature Tls, trigger voltage HV; There is lens position Pva variable attenuator unit 102, net cycle time Lva, cell temperature Tva; For transmission optical path unit 103, this unit is the most complicated, and parameter is maximum, comprises slit width Wslit, objective aperture Dpo, eyeglass displacement Dza, light path net cycle time Ltran etc.; There is light beam X-direction position Px beam direction unit 204, light beam Y-direction position Py, beam angle Pa, and pilot unit net cycle time Lbst; Also have at last energy sensor 206, its factor has operating voltage Vsr, sample rate current Isr, sensor net cycle time Lsr.

For traditional exposure system, be the integral body of a coupling depending on exposure system, have

Fes＝f(Ep，Texp，WL，Vls，Lls，.....，Vsr，Isr，Lsr)。

Parameter is numerous, and is too complicated to real-time computation modeling and the survey school of this model, is difficult to realize the real time correction of exposure system.So in the process of dosage control, at first Fes is considered as the model of a static state, the value Emeas that then records by energy sensor at every turn.Obtain difference

ΔE＝Emes-Eset

And then arrange in the energy in the exposure of next time, increase modified value

Emodify＝Eset+ΔE/Fes

Like this, visible system too depends on the precision of sensor, and real-time is also bad.

This method is considered in the exposure system between each subelement independence each other, and system is carried out decoupling zero, obtains fes (Ep, WL, Texp) ∩ flaser (Vls, Lls, Tls, HV) ∩ fva (Pva, Lva, Tva) ∩ ftran (Wslit, Dpo, Dza, Ltran) ∩ fbst (Px, Py, Pa, Pbst) ∩ fsensor (Vsr, Ist, Lst)

For each module, relate to parameter and greatly reduce, and the coupled relation of inside modules parameter much can be released by the principle of work of this module is theoretical, such as the parameter Px for the BST module, Py, Pa, also has the V parameter sr for sensor, Isr, Lsr.

In general each module comprises two parts to the factor of influence of exposure: to the gain factor of light beam and itself might produce side-play amount.Except sensor assembly, the gain factor of other modules is generally less than 1, and the gain factor of sensor assembly might be less than 1.And for the value of side-play amount, may be for just also may be for negative.And as the compensating parameter to exposure control, will get respectively the inverse of gain factor and the negative value of side-play amount.Thereby, by above-mentioned formula, through surveying the school, can obtain separately gain factor and the side-play amount of module.

Kenv∝fes(Ep，WL，Texp)

Klaser∝flaser(Vls，Lls，Tls，HV)

Kva∝fva(Pva，Lva，Tva，WL)

Ktransfer∝ftran(Wslit，Dpo，Dza，Ltran，WL)

Kbst∝fbst(Px，Py，Pa，Pbst)

Ksensor∝fsensor(Vsr，Ist，Lst)

In addition, for laser instrument and energy sensor, except the gain coefficient that will calculate them, also need to calculate their dark current drift Blaser and Bsensor.

Blaser∝Glaser(Vls，Lls，Tls)

Bsensor∝Gsensor(Vsr，Ist，Lst，Texp)

Wherein: Glaser is the laser instrument dark current computation model after the decoupling zero; Glaser is the sensor dark current computation model after the decoupling zero.

For unified compensation control module, the present invention specifies in conjunction with Fig. 3, requiring the standard energy of input is EpNom, this standard energy value need to draw the actual energy value Eact that will export by feedforward compensation algorithm (feedforward compensate algorithm), wherein in the feedforward arithmetic required usefulness operational factor F@laser, F@VA, F@BST, F@TransUnit, respectively by corresponding laser controlling model (Laser Control Model) 201, variable attenuator control model (VA control model) 202, beam director control model (BST control model) 204, optic path control model (Transfer Unit Model) 203 independently provides respectively, after this energy value is proofreaied and correct via PI adjuster (PI-regulator) 301 and overshoot adjuster (overshoot regulator) 302, draw the current value Eset that needs setting, calculate the high-voltage value HV that need to issue laser instrument via pulse energy and control voltage conversion model HVEp computing module 303 again, then be the laser pulse that laser instrument (laser trigger) 101 triggers corresponding energy, laser pulse sends by finally arriving wafer wafer105 by variable attenuator (VA) 206 and optic path unit (Laser transfer unit) 103 etc., pulse energy value Ep@sensor after recording and process by energy sensor/sensor module (energy sensor) 206 after pulse being sent feeds back to the pulse energy value that PI adjuster 301 and overshoot adjuster 302 be used for next step and calculates, and finally forms closed loop.

Therefore, for system shown in Figure 1, the target pulse energy that need to reach is Enom, and then actual input value is:

E _set＝(E _nom·K _laser+B _laser)·K _VA·K _transfer·K _BST·K _env

And the value that records is Em (i), then puts into the actual value Eact (i) that model calculates to be:

$E_{\mathrm{act}}\left[i\right]=\frac{(E_m\left(i\right)-B_{\mathrm{sensor}}[i-1\left]\right)}{K_{\mathrm{sensor}}[i-1]}$

Finally for the i time Laser emission pulse Eset[i among the buffer], calculated by following formula:

E _set[i]＝(E _nom·K _laser[i-1]+B _laser[i])·K _VA[i-1]·K _transfer[i-1]·K _BST[i-1]·K _env[i-1]-Fct _PI+Fct _Overshoot

PI adjusted value Fct wherein _PIBe calculated as:

${\mathrm{Factor}}_{\mathrm{PI}}=K_p\·\underset{k=i-N+1}{\overset{i-1}{\mathrm{\Σ}}}(\frac{E_{\mathrm{act}}\left[k\right]}{K_{\mathrm{transfer}}\left[k\right]}-E_{\mathrm{nom}})$

Overshoot adjusted value Fct wherein _OvershootBe calculated as:

${\mathrm{Factor}}_{\mathrm{Overshoot}}=\frac{K_f}{M}\·\underset{l=i-M}{\overset{i-1}{\mathrm{\Σ}}}(\frac{E_{\mathrm{act}}\left[k\right]}{K_{\mathrm{transfer}}\left[k\right]}-E_{\mathrm{set}}\left(l\right))$

For wherein each individual control module, such as the laser controlling model 201 among Fig. 1, variable attenuator control model 202, beam director control model 203, optic path control model 203, energy sensor module 106 etc., they have unified structure.Be described as follows as an example of Fig. 3 example: this standalone module is comprised of 3 submodules, signal collection module 401, module parameter logging modle 402 and output factor computing module 403.Wherein signal collection module 401 is responsible for all relevant parameters of acquisition module and is given module parameter logging modle 402, the critical function that signal collection module 401 also needs to finish is exactly the noise remove function, the below illustrates: this module can gather corresponding sampling parameter S1, and it is deposited among the corresponding buffer memory array ValN, after each the collection, to the V parameter al in the first two cycle in the array _I-2Check, when this parameter satisfies following 2 conditions simultaneously:

a.|Val _i-2-(Val _i-3+Val _i-4)/2|＞|Val _i-2|·25％

b.|Val _i-2-(Val _i-1+Val _i)/2|＞|Val _i-2|·25％

Think that then V parameter ali-2 is noise, in buffer memory, remove this parameter.Simultaneously the pointer of corresponding buffer memory is done corresponding movement.Then be module parameter logging modle 402, that this module is used is annular buffer, the structure that each unit among the buffer forms for meeting this module parameter.Output parameter computing module 403 at last, concrete model function according to this module, and from reference record module 402, obtain the related operation parameter, then also have the weighing factor parameter of this module paired pulses Dose Results etc., output parameter V1 (such as sensor assembly) and the calibration factor F1 (such as the VA module) after the calibration calculated and exported to this module 403 of comprehensive above parameter.

For each standalone module, environmental parameter E can obtain correlation from the temperature sensor of systematic unity such as temperature, time parameter T such as each time shutter, also should be issued by the exposure system unification and obtain.Then for each state parameter K, can according to the concrete condition of each model and the work requirements of exposure, need the measured value that suitably increases the correlated sampling sensor or adopt other relevant modules.For sampling parameter X, mainly refer to sensor assembly, their measured value is also put into model as input value simultaneously and is carried out analytical calculation.In addition, the sensor assembly that adopts all needs with the unified calibration of standard transducer to be delivered for use at limber up period and each maintenance phase.

For the survey school process of each design parameter, according to the concrete condition of parameter, it surveys school side's formula also can be different.Roughly can be divided into following several:

1) for some absolute variable parameters, such as time parameter, only need to be corrected to time standard time in the time of correction and be consistent, namely eliminate and measure to such an extent that the offset error of value and standard value is just passable.Be Tact=T@sensor-Toffset;

2) for the parameter of some remote effect to the exposure result, such as attenuator lens angle in the light path, PO lens location parameter (being not limited to above parameter), the correction of these parameters can be adopted the method for setting up calibration curve.Can be by the survey school flow process of this sensor being tried to achieve the calibration curve of measurement, then the data with each calibration point on the curve deposit in the correction card of storer, in the measuring process of normal exposure, try to achieve revised measurement result by tabling look-up.

The process that obtains calibration curve is: the input end at sensor adds calibration instrument one by one, input known quantity (such as voltage) x1, and x2 ..., xn ..., and obtain actual measured results y1, y2 ..., yn ....So can make calibration curve as shown below.These yn values that actual measurement is obtained deposit corresponding all xn values wherein in as content as an address in the storer, and this has just set up a correction card.Then, record a yn value when actual measurement, just make controller remove to access this address yn, read its content xn, this xn is measured through corrected value.For the y value between certain two calibration point yn and yn+1 the time, can go to search corresponding x value as end product by the most contiguous value yn or yn+1, this result will be with certain residual error so.

Calibration curve sees also the calibration curve section of Fig. 5 between any two calibration points, can regard approx one section straight-line segment as, if the slope of this section straight line is s=dx/dy, (note, timing y is independent variable, and x is functional value), the maximum slope of calibration curve is sl, by figure (b) as seen, the maximum residual error that may cause is

Δx＝sl*Δy

Δ y=yn+1-yn wherein

If consider to get two-way error, the absolute value of residual error can reduce half, is

±Δx＝±sl*Δy/2

If Y is the range of y, get permanent equally spaced N calibration point during calibration, namely

yn+1-yn＝Δy＝Y/N

So get Δ x=sl*Y/2N

The attenuator lens angle is surveyed the school as example in the light path, and the range when surveying the school is 90 degree, gets 19 spaced points at constant 18 intervals, is spaced apart 5 degree between the point.Set up a lens angle correction card, corresponding each absolute angle has a measurement value sensor after surveying the school, such as following table:

0	5	10	15	20	25	30	35	40	45	50	55	60	65	70	75	80	85	90
																			S0	S5	S10	S15	S20	S25	S30	S35	S40	S45	S50	S55	S60	S65	S70	S75	S80	S85	S90

Can guarantee that S0 is monotone increasing to the value of S90 after surveying the school.

When real-time exposure, if value such as Sval=S35 in the value that the records table, then directly the corresponding absolute lens angle value Dval=35degree of S35 being used for model calculates, if be the value between S0-S90 at random, then can get than the large minimum value of this value and the maximal value less than this value, do interpolation calculation, to draw corresponding absolute lens angle value.Draw the calibration calculations angle value and be as being positioned at [S50-S55] when the value that records in the exposure when interval, tabling look-up:

Dval＝(Sval-S50)*(55-50)/(S55-S50)+50

This survey calibration method is used for all remote effect to the parameter of measurement result, such as attenuator lens angle and PO lens location, but is not limited to above parameter.Certainly this value Dval also has the residual error that can't eliminate.This sentence can be removed.

3) for the parameter that directly has influence on measurement result, such as the pulse energy value that records.Method in employing situation 2 is with the nonlinearity erron of eliminating measured value.Can also adopt differential mode, adopt the variate mode, to measurement point by forward with oppositely connect respectively an identical survey sensor, and the measured value that the two draws is subtracted each other.Like this, zero-bit output and even nonlinearity item have been eliminated in this output, reduced non-linear, nearly one times of raising sensitivity, and offset common-mode error.

More than be the survey school process of parameter itself, survey the school out for how model independently being controlled on each subsystem ground, the below will describe as an example of model VA example

Fig. 5 is with a kind of typical VA module.Incident light is via VA, and part energy is used as the reflected light shunting.Emergent light is the value after decaying.For this module, ambient temperature value Tva can directly directly obtain from subsystem, lambda1-wavelength WL can be used as the machine constant and obtains, so this VA model is in order to realize correction calculation, life-span timing module 212 and an eyeglass deviation post sensor 211 of only needing increase itself, hardware system need not to change substantially.

The VA exposure control model that needs the survey school to go out is Kva ∝ fva (Pva, Lva, Tva, WL), at first analyze the VA model, energy be the pulse of Ein via this module after, its outgoing energy is Eout, and the ratio of Eout and Ein both had been Kva, and for the parameter factors lens position Pva of model, VA net cycle time Lva, VA environment temperature Tva, and laser wavelength WL are to model analysis Pva as can be known, Lva, Tva and Wl are independently on the impact of the module factor.So can test respectively separately factor coefficient.Except the COEFFICIENT K lva of working time Lva, other several parameter K pva, Ktva, Kwl can draw with following mode, take Kpva as example, keeps other factor inconvenience, and changing needs Kpva, the attenuation rate Kva of record VA module.Obtain each self-corresponding one group of data [Eout[i], Kpva[i]], the parameter that affects of this factor is done the first-order linear match, draw:

Kva＝K _pva*Pva+B _pva

Wherein for Apva, have

$K_{\mathrm{pva}}=\frac{\mathrm{n\Σ}{\mathrm{Kva}}_i\·{\mathrm{Pva}}_i-\mathrm{\Σ}{\mathrm{Kva}}_i\·\mathrm{\Σ}{\mathrm{Pva}}_i}{\mathrm{n\Σ}{\mathrm{Pva}}_i^2-{\left(\mathrm{\Σ}{\mathrm{Kva}}_i\right)}^2};$

$B_{\mathrm{pva}}=\frac{\mathrm{\Σ}{\mathrm{Kva}}_i\·\mathrm{Pv}{a}_i^2-\mathrm{\Σ}({\mathrm{Kva}}_i\·{\mathrm{Pva}}_i)\·\mathrm{\Σ}{\mathrm{Pva}}_i}{\mathrm{n\Σ}{\mathrm{Pva}}_i^2-{\left(\mathrm{\Σ}{\mathrm{Kva}}_i\right)}^2};$

In like manner draw K _Tva, B _Tva, L _Wl, B _Wl

For net cycle time parameter coefficient Klva, employing be that the experimental formula of SMEE draws:

Klva＝Flife＝exp(-Lva/Lrva)

At last, drawing real-time VA calibration model is:

Kva＝(K _pva*Pva+K _tva*Tva+K _wl*WL+B _pva+B _tva+B _wl)*Flife(Lva)

In the calculating of VA compensation, use Fval, just can realize more accurate compensation correction to VA, rather than only depend on energy sensor.

Other modules according to the characteristic of itself and to system exposure result's influence mode, can be set up corresponding model in a similar manner.In addition, since known to the operating characteristic of module own, so the coupled relation of some inside modules factor is also known, therefore this relation can be joined in the standalone module calibration model.For the correction parameter Vx in the model, this is directly its parameter can be used for the situation that overall exposing is calculated for sensor assembly is this.Also can make corresponding correction.

For block mold and its layout, wherein the result has the greatest impact is laser instrument to exposure, can set up the model that it arranges emission energy value and actual transmission energy value for this module, and because laser instrument itself has very perfect control system, also can finish compensation to its emitted energy by laser instrument itself, then related data be given host computer and process; What next impact was very crucial is exactly sensor assembly, need to stand its accurate compensation sensor assembly, analyzes the various impact inputs on him, and sets up corresponding compensation calibration model.And, can also adopt a plurality of sensors are set in exposure light path, they are carried out outside separately the compensation correction, the sampling model of setting up again the coupling of an integral body calculates compensation.Then be the light path factor of influence, light path comprises a lot of parts: ZA, PO; BST sees through the unit, object lens etc.; can to these modules respectively or unified set up a compensation model, protect for the Patents of the existing THFFC of this module unified model part SMEE.

According to the different layouts of Exposure Control Module, configuration and conditions of demand can also have the various types of like example of other.Below enumerate again two other typical exposure applications example.

Fig. 7 is the structural representation of the second embodiment of the present invention.

As shown in Figure 7, with respect to the exposure dose control system in the embodiment one, the laser instrument in the present case, optic path unit, the state of light beam guide unit (BST) is controlled, thereby the factor of influence of its exposure dose all is used as the known input parameter of system.Then input parameter all confirm complete after, at first the pulse energy value of input is issued laser instrument 101, then laser instrument triggers laser pulse, can be through variable attenuator VA102 and optic path unit 103 after laser pulse sends, arrive wafer 105, realize exposure, the actual value of exposure then can measure and inform system by the energy sensor 106 in the light path.Based on aforementioned exposure process, after system receives the required exposure dose that reaches, behind exposure dose and the unified compensation of correlation parameter input control module 207, the factor that draws during according to each model initialization, draw through this actual pulse energy value that should send behind the feedforward compensation, and this pulse is handed down to laser instrument triggers corresponding laser, after this trigger action, variable attenuator model 202 independent its corresponding administration module parameter informations that gather, and calculate the factor that next pulse energy triggers the needs compensation according to model, at last after energy sensor is measured this actual pulse energy, energy sensor control model 206 also needs the relevant parameter of energy sensor is gathered, and calculates the actual compensating factor of current pulse energy.All relevant factors are issued unified compensation control module 207 the most at last, and control model 207 can calculate in conjunction with known input parameter the triggering controlling value of next pulse energy.Whole system has formed a closed-loop control system that feedforward+feedback combines in the mode such as figure.

Certainly to should example two, its mutually deserved unified compensation control module 207 also can adjust accordingly according to the variation of control system among Fig. 2, and the structure of each standalone module and fundamental mode and example 1 are similar.

Fig. 8 is the structural representation of the 3rd embodiment of the present invention.

As shown in Figure 7, with respect to the exposure dose control system in the case 2, respectively increased an energy sensor in optic path unit and wafer exposure position respectively in the present case.Two sensors of optic path unit are finished the closed loop compensation effect of pulse energy jointly.And the energy sensor of wafer exposure position mainly is demarcation and calibration for the first two sensor, and normal operation is that this module is not done sampling.System's control flow is as follows: input parameter all confirm complete after, at first the pulse energy value of input is issued laser instrument 101, then laser instrument triggers laser pulse, can be through variable attenuator VA102 and optic path unit 103 after laser pulse sends, arrive wafer 105, realize exposure.And after laser pulse triggered, sensor 1 and sensor 2 can record pulse energy value and after processing it be informed system at the diverse location of light path.Based on aforementioned exposure process, after system receives the required exposure dose that reaches, behind exposure dose and the unified compensation of correlation parameter input control module 207, the factor that draws during according to each model initialization, draw through this actual pulse energy value that should send behind the feedforward compensation, and this pulse is handed down to laser instrument triggers corresponding laser, after this trigger action, variable attenuator model 202 independent its corresponding administration module parameter informations that gather, and calculate this module factor that next pulse energy triggers the needs compensation according to model, then when energy sensor 1 and energy sensor 2 all finish measure this actual pulse energy after, energy sensor control model 1 also can gather the relevant parameter of energy sensor with energy sensor model 2, calculates the actual compensating factor of current pulse energy.All relevant factors are issued unified compensation control module 207 the most at last, and control model 207 can calculate in conjunction with known input parameter the triggering controlling value of next pulse energy.Whole system has formed a closed-loop control system that feedforward+feedback combines in the mode such as figure.

Certainly to should example 3, its mutually deserved unified compensation control module 207 also can adjust accordingly according to the variation of control system among Fig. 3, and the structure of each standalone module and fundamental mode and example 1 are similar.

Described in this instructions is preferred embodiment of the present invention, and above embodiment is only in order to illustrate technical scheme of the present invention but not limitation of the present invention.All those skilled in the art all should be within the scope of the present invention under this invention's idea by the available technical scheme of logical analysis, reasoning, or a limited experiment.

Claims (34)

1. distributed exposure dose control system, comprise laser element, variable attenuator, the optic path unit, light path guidance unit and energy sensor, described laser element produces a laser pulse through described variable attenuator, measure the actual laser pulse that receives by described energy sensor behind optic path unit and the light path guidance unit, it is characterized in that, described laser element, variable attenuator, the optic path unit, light path guidance unit and energy sensor adopt respectively independent calibration model to obtain the compensating factor of different units, the described actual laser pulse that receives and the parameter group calibration factor according to described different units is carried out obtaining a pulse energy value behind the feedforward compensation, and described pulse energy value is carried out feedback compensation to obtain exposure dose after by a unified compensation control module output.

2. distributed exposure dose control as claimed in claim 1 system is characterized in that described independent calibration model comprises signal collection module, module parameter logging modle and output factor computing module.

3. distributed exposure dose control as claimed in claim 2 system is characterized in that, described signal collection module acquisition parameter and remove noise after be handed down to the module parameter logging modle.

4. distributed exposure dose control as claimed in claim 3 system, it is characterized in that, described signal collection module acquisition parameter and the step of removing noise comprise: acquisition parameter S1, and it is deposited among the buffer memory array ValN, after each the collection, V parameter ali-2 to the first two cycle in the array checks, when this parameter meets the following conditions simultaneously: (| Vali-2-(Vali-3+Vali-4)/2|＞| Vali-2|25%) ∩ (| Vali-2-(Vali-1+Vali)/2|＞| Vali-2|25%), think that then V parameter ali-2 is noise, in buffer memory, remove this parameter.

5. distributed exposure dose control as claimed in claim 2 system is characterized in that described module parameter logging modle adopts buffer circle, the structure that each unit in the described buffer circle forms for meeting this module parameter.

6. distributed exposure dose control as claimed in claim 2 system, it is characterized in that, described output factor computing module obtains operational parameter and weighing factor parameter from described module parameter logging modle, according to output parameter and the calibration factor after described operational parameter and weighing factor calculation of parameter and the input calibration, described operational parameter comprises absolute variable parameter, the remote effect result parameter directly affects result parameter.

7. distributed exposure dose control as claimed in claim 6 system is characterized in that described computing method are: described absolute variable parameter is eliminated measured to get the offset error of value and standard value.

8. distributed exposure dose control as claimed in claim 6 system, it is characterized in that, described computing method are: set up the calibration curve of described remote effect result parameter, and try to achieve revised measurement result by tabling look-up, to eliminate the nonlinearity erron of measured value.

9. distributed exposure dose control as claimed in claim 6 system is characterized in that described computing method are: on described direct affect result parameter by forward with oppositely connect respectively an identical survey sensor, and the measured value that the two draws is subtracted each other.

10. distributed exposure dose control as claimed in claim 1 system, it is characterized in that, described parameter group comprises: pulse energy Ep, time shutter Texp, optical maser wavelength WL, laser element operating voltage Vls, laser element net cycle time Lls,, laser element cell temperature Tls, laser element trigger voltage HV, variable attenuator unit lens position Pva, variable attenuator unit net cycle time Lva, variable attenuator cell temperature Tva, transmission optical path unit slit width Wslit, transmission optical path unit objective aperture Dpo, transmission optical path unit eyeglass displacement Dza, transmission optical path unit net cycle time Ltran, beam direction unit light beam X-direction position Px, beam direction unit light beam Y-direction position Py, beam direction unit beam angle Pa, beam direction unit net cycle time Lbst, energy sensor operating voltage Vsr, energy sensor sample rate current Isr, energy sensor net cycle time Lsr.

11. distributed exposure dose control as claimed in claim 10 system is characterized in that between described optimum exposure dosage and the described laser pulse following formula: E being arranged _Set=(E _NomK _Laser+ B _Laser) K _VAK _TransferK _BSTK _Env, wherein Eset is laser pulse, Enom is optimum exposure dosage, K _LaserBe the laser gain factor, Kva is the variable attenuator gain factor, and Ktransfer is the transmission light path module gain factor, and Blaser is the drift of laser instrument dark current, and Kbst is the beam direction module gain factor, and Kenv is the energy sensor gain factor.

12. distributed exposure dose control as claimed in claim 11 system is characterized in that described laser gain factor K _Laser, variable attenuator gain factor Kva, transmission light path module gain factor K transfer, laser instrument dark current drift Blaser, beam direction module gain factor K bst, energy sensor gain factor Kenv to obtain formula as follows:

Kenv∝fes(Ep，WL，Texp)，

Klaser∝flaser(Vls，Lls，Tls，HV)，

Kva∝fva(Pva，Lva，Tva，WL)，

Ktransfer∝ftran(Wslit，Dpo，Dza，Ltran，WL)，

Kbst∝fbst(Px，Py，Pa，Pbst)，

Ksensor∝fsensor(Vsr，Ist，Lst)，

Blaser∝Glaser(Vls，Lls，Tls)。

13. distributed exposure dose control as claimed in claim 1 system, it is characterized in that, described unified compensation control module is proofreaied and correct described pulse energy value through PI adjuster and overshoot adjuster after, after process HVEp computing module calculates the high-voltage value HV that need to issue laser instrument, described laser instrument triggers a laser pulse, described laser pulse feeds back to described PI adjuster and described overshoot adjuster after recording by described energy sensor module.

14. distributed exposure dose control as claimed in claim 13 system is characterized in that the computing formula of the adjusted value of described PI adjuster is as follows: ${\mathrm{Factor}}_{\mathrm{PI}}=K_p\·\underset{k=i-N+1}{\overset{i-1}{\mathrm{\Σ}}}(\frac{E_{\mathrm{act}}\left[k\right]}{K_{\mathrm{transfer}}\left[k\right]}-E_{\mathrm{nom}}),$ Wherein, Enom is optimum exposure dosage,

Kp is the gain factor of described PI adjuster, and Eact is the actual output energy value of described laser instrument, and Ktransfer is the transmission light path module gain factor.

15. distributed exposure dose control as claimed in claim 13 system is characterized in that the computing formula of described overshoot adjusted value is as follows: ${\mathrm{Factor}}_{\mathrm{Overshoot}}=\frac{K_f}{M}\·\underset{l=i-M}{\overset{i-1}{\mathrm{\Σ}}}(\frac{E_{\mathrm{act}}\left[k\right]}{K_{\mathrm{transfer}}\left[k\right]}-E_{\mathrm{set}}\left(l\right)),$ Wherein, Kf is the gain factor of described overshoot adjuster, and Eact is the actual output energy value of described laser instrument, and Ktransfer is the transmission light path module gain factor.

16. distributed exposure dose control as claimed in claim 1 system is characterized in that described variable attenuator comprises a life-span timing module and eyeglass deviation post sensor.

17. distributed exposure dose control as claimed in claim 16 system it is characterized in that, but the calibration model of described attenuator unit is:

Kva＝(K _pva*Pva+K _tva*Tva+K _wl*WL+B _pva+B _tva+B _wl)*Flife(Lva)，

Kva is attenuation rate in the formula, WL optical maser wavelength, Pva is variable attenuator unit lens position, Tva is the variable attenuator cell temperature, but Bpva is the position compensation datum offset factor of attenuator, but Btva is the temperature compensation datum offset factor of attenuator, but Bwl is the wavelength compensation datum offset factor of attenuator, and Lva is variable attenuator unit net cycle time; Wherein Flife is the life-span decay factor of variable attenuator, and the acquisition formula of Flife is that (Lva/Lrva), Lrva is the nominal operation life-span of attenuator to Flife=exp.

18. distributed exposure dose control method, comprise: laser element produces a laser pulse through variable attenuator, measure the actual laser pulse that receives by energy sensor behind optic path unit and the light path guidance unit, described laser element, variable attenuator, the optic path unit, light path guidance unit and energy sensor adopt respectively independent calibration model to obtain the compensating factor of different units, the described actual laser pulse that receives and the correlation parameter calibration factor according to described different units is carried out obtaining a pulse energy value behind the feedforward compensation, described pulse energy value is carried out feedback compensation to obtain optimum exposure dosage after by a unified compensation control module output.

19. distributed exposure dose control method as claimed in claim 18 is characterized in that, described independent calibration model comprises signal collection module, module parameter logging modle and output factor computing module.

20. distributed exposure dose control method as claimed in claim 19 is characterized in that, described signal collection module acquisition parameter and remove noise after be handed down to the module parameter logging modle.

21. distributed exposure dose control method as claimed in claim 20, it is characterized in that, described signal collection module acquisition parameter and the step of removing noise comprise: acquisition parameter S1, and it is deposited among the corresponding buffer memory array ValN, after each the collection, V parameter ali-2 to the first two cycle in the array checks, when this parameter meets the following conditions simultaneously: (| Vali-2-(Vali-3+Vali-4)/2|＞| Vali-2|25%) ∩ (| Vali-2-(Vali-1+Vali)/2|＞| Vali-2|25%), think that then V parameter ali-2 is noise, in buffer memory, remove this parameter.

22. distributed exposure dose control method as claimed in claim 19 is characterized in that, described module parameter logging modle adopts buffer circle, the structure that each unit in the described buffer circle forms for meeting this module parameter.

23. distributed exposure dose control method as claimed in claim 19, it is characterized in that, described output factor computing module obtains operational parameter and weighing factor parameter from described module parameter logging modle, according to output parameter and the calibration factor after described operational parameter and weighing factor calculation of parameter and the input calibration, described operational parameter comprises absolute variable parameter, the remote effect result parameter directly affects result parameter.

24. distributed exposure dose control method as claimed in claim 23 is characterized in that described computing method are: described absolute variable parameter eliminated measure to get the offset error of value and standard value.

25. distributed exposure dose control method as claimed in claim 23, it is characterized in that, described computing method are: set up the calibration curve of described remote effect result parameter, and try to achieve revised measurement result by tabling look-up, to eliminate the nonlinearity erron of measured value.

26. distributed exposure dose control method as claimed in claim 23, it is characterized in that, described computing method are: on described direct affect result parameter by forward with oppositely connect respectively an identical survey sensor, and the measured value that the two draws is subtracted each other.

27. distributed exposure dose control method as claimed in claim 18, it is characterized in that, described correlation parameter comprises: pulse energy Ep, time shutter Texp, optical maser wavelength WL, laser element operating voltage Vls, laser element net cycle time Lls,, laser element cell temperature Tls, laser element trigger voltage HV, variable attenuator unit lens position Pva, variable attenuator unit net cycle time Lva, variable attenuator cell temperature Tva, transmission optical path unit slit width Wslit, transmission optical path unit objective aperture Dpo, transmission optical path unit eyeglass displacement Dza, transmission optical path unit net cycle time Ltran, beam direction unit light beam X-direction position Px, beam direction unit light beam Y-direction position Py, beam direction unit beam angle Pa, beam direction unit net cycle time Lbst, energy sensor operating voltage Vsr, energy sensor sample rate current Isr, energy sensor net cycle time Lsr.

28. distributed exposure dose control method as claimed in claim 19 is characterized in that, between described optimum exposure dosage and the described laser pulse following formula: E is arranged _Set=(E _NomK _Laser+ B _Laser) K _VAK _TransferK _BSTK _Env, wherein Eset is laser pulse, Enom is optimum exposure dosage, K _LaserBe the laser gain factor, Kva is the variable attenuator gain factor, and Ktransfer is the transmission light path module gain factor, and Blaser is the drift of laser instrument dark current, and Kbst is the beam direction module gain factor, and Kenv is the energy sensor gain factor.

29. distributed exposure dose control method as claimed in claim 28 is characterized in that, described laser gain factor K _Laser, variable attenuator gain factor Kva, transmission light path module gain factor K transfer, laser instrument dark current drift Blaser, beam direction module gain factor K bst, energy sensor gain factor Kenv to obtain formula as follows:

Kenv∝fes(Ep，WL，Texp)，

Klaser∝flaser(Vls，Lls，Tls，HV)，

Kva∝fva(Pva，Lva，Tva，WL)，

Ktransfer∝ftran(Wslit，Dpo，Dza，Ltran，WL)，

Kbst∝fbst(Px，Py，Pa，Pbst)，

Ksensor∝fsensor(Vsr，Ist，Lst)，

Blaser∝Glaser(Vls，Lls，Tls)。

30. distributed exposure dose control method as claimed in claim 18, it is characterized in that, described unified compensation control module is proofreaied and correct described pulse energy value through PI adjuster and overshoot adjuster after, after process HVEp computing module calculates the high-voltage value HV that need to issue laser instrument, described laser instrument triggers a laser pulse, described laser pulse feeds back to described PI adjuster and described overshoot adjuster after recording by described energy sensor module.

31. distributed exposure dose control method as claimed in claim 30 is characterized in that the computing formula of the adjusted value of described PI adjuster is as follows: ${\mathrm{Factor}}_{\mathrm{PI}}=K_p\·\underset{k=i-N+1}{\overset{i-1}{\mathrm{\Σ}}}(\frac{E_{\mathrm{act}}\left[k\right]}{K_{\mathrm{transfer}}\left[k\right]}-E_{\mathrm{nom}}),$ Wherein, Enom is optimum exposure dosage, and Kp is the gain factor of described PI adjuster, and Eact is the actual output energy value of described laser instrument, and Ktransfer is the transmission light path module gain factor.

32. distributed exposure dose control method as claimed in claim 30 is characterized in that the computing formula of described overshoot adjusted value is as follows: ${\mathrm{Factor}}_{\mathrm{Overshoot}}=\frac{K_f}{M}\·\underset{l=i-M}{\overset{i-1}{\mathrm{\Σ}}}(\frac{E_{\mathrm{act}}\left[k\right]}{K_{\mathrm{transfer}}\left[k\right]}-E_{\mathrm{set}}\left(l\right)),$ Wherein, Kf is the gain factor of described overshoot adjuster, and Eact is the actual output energy value of described laser instrument, and Ktransfer is the transmission light path module gain factor.

33. distributed exposure dose control method as claimed in claim 18 is characterized in that, described variable attenuator comprises a life-span timing module and eyeglass deviation post sensor.

34. distributed exposure dose control method as claimed in claim 33 it is characterized in that, but the calibration model of described attenuator unit is:

Kva＝(K _pva*Pva+K _tva*Tva+K _wl*WL+B _pva+B _tva+B _wl)*Flife(Lva)，

Wherein Flife is the life-span decay factor of variable attenuator, Kva is attenuation rate, WL optical maser wavelength, Pva is variable attenuator unit lens position, Tva is the variable attenuator cell temperature, but Bpva is the position compensation datum offset factor of attenuator, and Btva is the temperature compensation datum offset factor of variable attenuator, Bwl is the wavelength compensation datum offset factor of variable attenuator, and Lva is variable attenuator unit net cycle time.

Images