CN1333460A - Scanning tunnel microscope feedback controller adopting multi-modal fuzzy control algorithm - Google Patents

Abstract

The scanning tunnel microscope feedback control equipment adopting multimode fuzzy control algorithm possesses sensor, scanning control circuit, control computer, X,Y,Z scanning driving circuit and X,Y,Z scanner. It uses the sensor to amplify tunnel current signal, transfer it into scanning control circuit, the scanning control circuit receives the reference voltage PEF-H and REF-L of control computer, at the same time delivers the ampified tunnel current signal Im and control signal Ready, and when the tunnel current signal Im is less than REF-H and greater than REF-L, the control signal Ready is high potential, otherwise it is low potential. The control computer can receive the control signal Ready delivered by scanning control circuit, when it is high potential, the computer can output scanning driving signal of X and Y direction to control scanner and make it scan in X and Y direction, and when it is low potential, the computer can output the scanning driving signal of Z direction to control scanner and make it scan in Z direction.

Description

Adopt the scanning tunnel microscope feedback controller of multi-modal fuzzy control algorithm

Technical field

The present invention relates to a kind of scanning tunnel microscope, particularly a kind of scanning tunnel microscope feedback controller that adopts multi-modal fuzzy control algorithm

Background technology

Scanning tunnel microscope can be divided into detection of permanent supreme people's court and constant flow method and detect according to the difference of detection mode.Permanent supreme people's court keeps needle point to do scanning motion on the surface level of a constant altitude above the sample, detects the variation of tunnel current, detects pattern and electrical characteristics that data set has been represented sample surfaces.Can make the feature image of sample surfaces with this data set.Permanent supreme people's court is because height that needn't mobile needle point, and therefore, test speed is very fast; But it is only applicable to level and smooth relatively sample surfaces.Constant flow method is a height of adjusting needle point with the method for feedback, keeps constant tunnel current, and the movement locus of needle point height has reflected the variation of sample surfaces pattern.Constant flow method can be measured irregular surface accurately; But owing to need the ceaselessly height of mobile needle point, so test speed is slower.Because the unknown of sample surfaces pattern, so constant flow method is the main method of using in present this quasi-instrument.

Because being applied between the electric displacement of voltage and piezoelectric ceramics on the piezoelectric ceramics is nonlinear relationship, therefore can not provide the corresponding relation of determining between voltage and the displacement, can only adopt the control of the way realization displacement of FEEDBACK CONTROL, and adopt which kind of control algolithm, just become the direct key factor that influences control performance.The control algolithm of STM employing at present is PID (or PI) algorithm, because piezoelectric ceramics voltage---the electric displacement curve has severe nonlinear, causes adjusting of pid parameter very difficult.Particularly the variation of factors such as environment, condition, time makes the parameter of initially adjusting no longer suitable, the control characteristic variation, and degree of regulation reduces, and governing speed is slack-off.

There are Digital Instruments Co. (USA), Park InstrumentsCo. (USA) in current scanline tunnel microscope manufacturer, the performance that China produces a little less than, market is captured by above-mentioned two families basically.DI company is assurance X, Y, and the accuracy of three direction displacements of Z adds capacitance displacement sensor respectively in three directions, and the shift value of piezoelectric ceramics is recorded by capacitive transducer.Shortcoming is that capacitive transducer need be demarcated, otherwise data distortion.

Park company is for improving measuring speed, and (2Hz on basis 10Hz), has added the pre-differentiation function of sample topography in two kinds of sweep velocitys of original system.At the sample surfaces of relatively flat, use sweep velocity (10Hz) faster, at uneven relatively sample surfaces, the sweep velocity (2Hz) with slower is improved sweep velocity.Shortcoming or do not throw off time-based scan method can produce data distortion equally.

Summary of the invention

The object of the present invention is to provide a kind of above-mentioned deficiency that overcomes, guarantee the speed and the accuracy of scanning, the scanning tunnel microscope feedback controller of the employing multi-modal fuzzy control algorithm that data can distortion.

The object of the present invention is achieved like this: the scanning tunnel microscope feedback controller that adopts multi-modal fuzzy control algorithm, include sensor, control computer, X, Y, the Z scan drive circuit, X-axis, Y-axis, Z axle scanner, it is characterized in that also being provided with scan control circuit, sensor is received the tunnel current signal, with its amplification, reach scan control circuit, scan control circuit receives reference voltage REF-H and the REF-L that control computer is come, tunnel current signal Im and control signal Ready after output is amplified simultaneously, control signal Ready is noble potential at tunnel current signal Im during less than REF-H and greater than REF-L, otherwise be electronegative potential, control computer receives the control signal Ready of scan control circuit output, when control signal Ready is noble potential, computing machine output X, the scanning drive signal of Y direction, and pass through X respectively, Y scan drive circuit control X-axis and Y-axis scanner are made X respectively, the scanning motion of Y direction; When control signal Ready was low-voltage, computing machine was only exported the scanning drive signal of Z direction, and did the scanning motion of Z direction by Z scan drive circuit control Z axle scanner.

The present invention is owing to adopt the foregoing circuit structure and the corresponding computer program, thus have scanning accurately, precision is high, the characteristics that sweep velocity is fast, do not have distortion.

Description of drawings

Fig. 1 is an electric structure frame chart of the present invention

Fig. 2 is the scan control circuit schematic diagram

Fig. 3 is the FUZZY ALGORITHMS FOR CONTROL process flow diagram

The specific embodiment

As shown in Figure 1, adopt the scanning tunnel microscope feedback controller of multi-modal fuzzy control algorithm, include sensing Device, scan control circuit, control computer, X, Y, Z scan drive circuit, X-axis, Y-axis, Z axis scanner, sensing Device is received the tunnel current signal, with its amplification, reaches scan control circuit, scan control circuit reception control computer Reference voltage REF-H and REF-L, tunnel current signal Im and control signal Ready after output is amplified simultaneously, control letter Number Ready is high potential at tunnel current signal Im during less than REF-H and greater than REF-L, otherwise is electronegative potential, control Computer receives the control signal Ready of scan control circuit output, when control signal Ready is high potential, and computer The scanning drive signal of output X, Y-direction, and respectively by X, Y scan drive circuit control X-axis and Y axis scanning device branch Do not do the scanning motion of X, Y-direction; When control signal Ready was low-voltage, computer was only exported the scanning of Z direction Drive signal, and do the scanning motion of Z direction by Z scan drive circuit control Z axis scanner.

As shown in Figure 2, scan control circuit is by dual-stage amplifier F1, F2, comparator B1, B2, XOR gate E, electricity Resistance R1-R8, adjustable resistance P1, P2, capacitor C 1-C6 consist of jointly, and the negative signal of sensor Sensor connects through resistance R 1 Connect the pin 3 of first order amplifier F1, the positive signal of sensor Sensor connects the pin of first order amplifier F1 through resistance R 2 The pin 2 of 2, F1 also passes through capacitor C 4 ground connection, passes through resistance R 4 pins 6, and pin 6 links to each other pin by capacitor C 2 with pin 5 5 connect-the 15V power supply together by capacitor C 1 and pin 4, and the pin 3 of F1 is also by the earth of resistance R 3 with capacitor C 3, Pin 8 links to each other with pin 1 by adjustable resistance P1, and adjustable end and the pin 7 of adjustable resistance P1 connect+the 15V power supply together, F1's Pin 6 is also by capacitor C 5 ground connection, and pin 6 is gone back contact resistance R5, the other end of resistance R 5 respectively by capacitor C 6 ground connection, Receive the pin 2 of second level amplifier F2 by resistance R 6; The pin 2 of second level amplifier F2 is by resistance R 7, adjustable electric Resistance P2 links to each other with pin 6, and pin 3 is by resistance R 8 ground connection, and pin 4 connects-the 15V power supply, and pin 7 connects+the 15V power supply, and pin 6 also The pin 3 that connects respectively comparator B1, B2; The pin 2 of comparator B1 connects the output reference voltage REF-H of control computer, The pin 2 of B2 connects the output reference voltage REF-L of control computer, and the output pin 7 of comparator B1 connects XOR gate E Input 1, the output pin 7 of B2 connects the input 2 of XOR gate E, the output 3 of XOR gate exports computer to, Be the Ready control signal, the pin 6 of amplifier F2 is also exported the current input terminal Im that a signal is connected to the control computer. The model of amplifier F1 is in this circuit: μ A725, and the model of F2 is: μ A741; The model of comparator B1, B2 is LM393; The model of XOR gate is 7486.

After the output signal of control circuit entered computer, computer carried out computing according to fuzzy control program, the idiographic flow of its program as shown in Figure 3:

Program is called in internal memory with program from hard disk since 100, the CPU program that brings into operation; 101 executive systems initialize, CPU receives operator's tunnel current setting value by a keyboard entry; Traveling probe makes it near sample, and enters the tunnel State.

102 read the tunnel current detected value from A/D converter; 103 usefulness tunnel currents setting value subtracts the tunnel current detected value, Its difference is deviation, and deviation is assigned to deviation register e; Program changes 104 over to.

Whether 104 judgment bias register e are zero, if the result is True, then change 105 over to; If be False, then Change 108 over to.

108 carry out the operation of calculation deviation rate of change, and deviation variation rate Δ e is determined by following formula: $\mathrm{\Δe}=\frac{e}{\mathrm{\Δt}};$ 109 couples of e Carry out the obfuscation operation with Δ e, program changes 110 over to.

110 results according to obfuscation call the fuzzy control rule subprogram; 111 call the fuzzy inference rule subprogram; 112 carry out fuzzy judgment according to result of calculation operates, and draws output variable.

113 according to output variable generation output controlling increment; Program changes 114 over to.

114 store increment, and carry out the D/A conversion, the voltage of a 0～5V of output.

115 send this voltage into high pressure generator, and (some piezoelectric ceramic devices is with 0～150V) to produce 0 corresponding～300V Voltage; 116 are added to this voltage on the piezoelectric ceramics, make probe produce corresponding displacement; Program flow returns 102 then, continue Continuous repeatedly execution.

105 judge whether each point of inswept presumptive area, if be True, program changes 106 over to; If be False, Program changes 114 over to, continues to carry out.

106 pairs of data of storing are processed, and data are sent into image buffer memory, for showing figure at display Resemble and get ready. 107 FEEDBACK CONTROL EPs (end of program).

Claims (3)

1. scanning tunnel microscope feedback controller that adopts multi-modal fuzzy control algorithm, include sensor, control computer, X, Y, the Z scan drive circuit, X-axis, Y-axis, Z axle scanner, it is characterized in that also being provided with scan control circuit, sensor is received the tunnel current signal, with its amplification, reach scan control circuit, scan control circuit receives reference voltage REF-H and the REF-L that control computer is come, tunnel current signal Im and control signal Ready after output is amplified simultaneously, control signal Ready is noble potential at tunnel current signal Im during less than REF-H and greater than REF-L, otherwise be electronegative potential, control computer receives the control signal Ready of scan control circuit output, when control signal Ready is noble potential, computing machine output X, the scanning drive signal of Y direction, and pass through X respectively, X is made in Y scan drive circuit control X-axis and Y-axis scanning respectively, the scanning motion of Y direction; When control signal Ready was low-voltage, computing machine was only exported the scanning drive signal of Z direction, and did the scanning motion of Z direction by Z scan drive circuit control Z axle scanner.

2. the scanning tunnel microscope feedback controller of employing multi-modal fuzzy control algorithm according to claim 1, it is characterized in that, scan control circuit is by two-stage amplifier F1, F2, comparer B1, B2, XOR gate E, resistance R 1-R8, adjustable resistance P1, P2, capacitor C 1-C6 constitutes jointly, the negative signal of sensor Sensor connects the pin 3 of first order amplifier F1 through resistance R 1, the positive signal of sensor Sensor connects the pin 2 of first order amplifier F1 through resistance R 2, the pin 2 of F1 is also by capacitor C 4 ground connection, by resistance R 4 pins 6, pin 6 links to each other with pin 5 by capacitor C 2, pin 5 connects-the 15V power supply together by capacitor C 1 and pin 4, the pin 3 of F1 is also by the in parallel ground connection of resistance R 3 with capacitor C 3, pin 8 links to each other with pin 1 by adjustable resistance P1, adjustable end and the pin 7 of adjustable resistance P1 connect+the 15V power supply together, the pin 6 of F1 is also by capacitor C 5 ground connection, pin 6 also connects resistance R 5, and the other end of resistance R 5 is respectively by capacitor C 6 ground connection, receive the pin 2 of second level amplifier F2 by resistance R 6; The pin 2 of second level amplifier F2 links to each other with pin 6 by resistance R 7, adjustable resistance P2, and pin 3 is by resistance R 8 ground connection, and pin 4 connects-the 15V power supply, and pin 7 connects+the 15V power supply, and pin 6 also connects the pin 3 of comparer B1, B2 respectively; The pin 2 of comparer B1 connects the output reference voltage REF-H of control computer, the pin 2 of B2 connects the output reference voltage REF-L of control computer, the output terminal pin 7 of comparer B1 connects the input end 1 of XOR gate E, the output terminal pin 7 of B2 connects the input end 2 of XOR gate E, the output terminal 3 of XOR gate exports computing machine to, be the Ready control signal, the pin 6 of amplifier F2 is also exported the current input terminal Im that a signal is connected to control computer.

3. the scanning tunnel microscope feedback controller of employing multi-modal fuzzy control algorithm according to claim 1, after the output signal that it is characterized in that control circuit enters computing machine, computing machine carries out computing according to fuzzy control program, and the idiographic flow of its program is:

Program is called in internal memory with program from hard disk since 100, the CPU program that brings into operation; 101 executive system initialization, CPU receives operator's tunnel current setting value by a keyboard entry; Traveling probe makes it near sample, and enters tunnel state.

102 read the tunnel current detected value from A/D converter; 103 usefulness tunnel current setting values subtract the tunnel current detected value, and its difference is a deviation, and deviation is composed to deviation register e; Program changes 104 over to.

104 judge that whether deviation register e is zero, if the result is True, then changes 105 over to; If be False, then change 108 over to.

108 carry out the operation of calculation deviation rate of change, and deviation variation rate Δ e is determined by following formula: $\mathrm{\Δe}=\frac{e}{\mathrm{\Δt}};$ The 109 couples of e and Δ e carry out the obfuscation operation, and program changes 110 over to.

110 results according to obfuscation call the fuzzy control rule subroutine; 111 call the fuzzy inference rule subroutine; 112 carry out fuzzy judgment according to result of calculation operates, and draws output variable.

113 according to output variable generation output control increment; Program changes 114 over to.

114 store increment size, and carry out the D/A conversion, the voltage of a 0～5V of output.

115 send this voltage into high pressure generator, produce the voltage of corresponding 0～300V; 116 are added to this voltage on the piezoelectric ceramics, make probe produce corresponding displacement; Program flow returns 102 then, continue to carry out repeatedly.

105 judge whether each point of inswept presumptive area, if be True, program changes 106 over to; If be False, program changes 114 over to, continues to carry out.

106 pairs of data of being stored are handled, and data are sent into image buffer memory, for displayed image on display is got ready.107 FEEDBACK CONTROL EOP (end of program).

Images