CN104697497A - Digital tilt sensor and temperature nonlinear compensation method thereof - Google Patents

Abstract

The invention discloses a digital tilt sensor and a temperature nonlinear compensation method thereof. The digital tilt sensor comprises an MEMS (Micro-electromechanical Systems) sensor, a filter module, a microprocessor, a reference power supply module, a communication module and a voltage-stabilized power supply module, wherein the microprocessor contains an A/D conversion module, a temperature compensation module and a nonlinear compensation module. The MEMS sensor outputs an analog voltage signal containing tilt angle information, the analog voltage signal is output to the A/D conversion module by the filter module, and meanwhile, the MEMS sensor also outputs a temperature digital signal to the temperature compensation module. In the microprocessor, the outputs of the A/D conversion module and the temperature compensation module are integrated by the nonlinear compensation module, and then a tilt angle measured value digital signal is output. The temperature nonlinear compensation technology adopted by the temperature compensation module and the nonlinear compensation module makes up the defects that in the existing similar technologies, the temperature compensation error is large and the nonlinear error is obvious. The digital tilt sensor disclosed by the invention has the advantages of low cost, high measurement accuracy, vibration shock resistance, and applicability to wide-temperature harsh environments.

Description

A kind of digital obliquity sensor and nonlinear temperature compensation method thereof

Technical field

The present invention relates to obliquity sensor, more specifically, relate to a kind of digital obliquity sensor and nonlinear temperature compensation method thereof.

Background technology

Obliquity sensor is usually used in the occasion of the bad environments such as operational motion is frequent, outdoor, therefore requires very high in response speed, reliability, volume, stability, serviceable life and cost.Traditional mechanical type and electromagnetic type obliquity sensor equipment volume is large, precision is low, measure time delay long, affect greatly by environment temperature etc.Obliquity sensor based on MEMS acceleration transducer adopts little, low in energy consumption, the fast response time of volume and highly reliable sensing element, is applicable to engineering machinery field.

But the easy temperature influence of MEMS sensor, seen in existing document, temperature compensation model adopts an order polynomial, though the method calculates simple, accurate not; Adopt inverse sine method when existing document calculates measurement of dip angle value simultaneously, the method have ignored quadratic nonlinearity error, also accurate not to cubic non-linearity compensation of error.

Summary of the invention

For above defect or the Improvement requirement of prior art, the invention provides a kind of digital obliquity sensor and nonlinear temperature compensation method thereof, be intended to solve existing obliquity sensor to the coarse technical matters of nonlinear temperature error compensation.

The invention provides a kind of digital obliquity sensor, comprise MEMS sensor, filtration module, microprocessor, communication module, reference power supply module and power module of voltage regulation, microprocessor comprises temperature compensation module, A/D conversion module and nonlinearity compensation module, the power input of the power input of the power input of MEMS sensor, the power input of filtration module, microprocessor, reference power supply module and described communication module all connects power module of voltage regulation, the input end of temperature compensation module is connected to the first output terminal of MEMS sensor, the signal input part of filtration module is connected to the second output terminal of MEMS sensor, the first input end of A/D conversion module is connected to the output terminal of filtration module, second input end of A/D conversion module is connected to the output terminal of reference power supply module, the first input end of nonlinearity compensation module is connected to the output terminal of temperature compensation module, second input end of nonlinearity compensation module is connected to the output terminal of A/D conversion module, the input end of communication module is connected to the output terminal of nonlinearity compensation module.

Further, during work, the temperature digital signal that the first output terminal of described MEMS sensor exports carries out temperature compensation calculating to described temperature compensation module, calculates gained and is obtained by nonlinearity compensation module; The analog voltage signal containing obliquity information that second output terminal of MEMS sensor exports is transferred to described A/D conversion module after described filtration module filtering, and conversion gained is obtained by nonlinearity compensation module containing the voltage digital signal of obliquity information; The temperature compensation parameter that described nonlinearity compensation module utilizes described temperature compensation module to calculate carries out nonlinear compensation to containing the voltage digital signal of obliquity information, and the measurement of dip angle value digital signal after compensating is exported to communication module exports to user by described communication module.

Further, microprocessor adopts model to be the single-chip microcomputer of ADuC845.

Present invention also offers a kind of nonlinear temperature compensation method based on above-mentioned digital obliquity sensor, comprise the steps:

S1: evenly choose several test temperature points T within the scope of full temperature to be measured _i, adjacent temperature spot interval should be not more than 10 DEG C, at each test temperature point T _iunder, temperature control turntable the temperature inside the box should be incubated the sufficiently long time to make inclination angle sensor device internal and external temperature in a basic balance;

S2: at each test temperature point T _iunder, within the scope of-8 ° ~ 8 ° inclination angles, evenly choose several measurement of dip angle points θ _j, adjacent measurement of dip angle point angle intervals should be not more than 1 °;

S3: at each test temperature point T _iunder, after the temperature control turntable case inside holding long enough time, control turntable and rotate to each measurement of dip angle point θ _j, now gather the voltage digital signal V (T containing obliquity information that in inclination angle sensor device, A/D conversion module exports _i, θ _j) and temperature digital signal TEMP (T _i);

S4: at each temperature spot T _i, adopt least square fitting to gather digital voltage V (T _i, θ _j) and inclination angle theta _jcubic polynomial, obtain each temperature spot T _iunder zero measurement value V ₀(T _i) and sensitivity measure value S _c(T _i), and according to described zero measurement value V ₀(T _i) and described sensitivity measure value S _c(T _i) obtain acceleration calculation value under each temperature spot, each demarcation angle $a(T_i,{\mathrm{\θ}}_j)=\frac{V(T_i,{\mathrm{\θ}}_j)-V_0\left(T_i\right)}{g\·S_c\left(T_i\right)};$

S5: adopt each temperature spot T of least square fitting _iunder inclination value θ _jwith acceleration calculation value a (T _i, θ _j) three relations, obtain each temperature spot T _iunder quadratic nonlinearity co-efficient measurements C ₂(T _i), cubic non-linearity co-efficient measurements C ₃(T _i);

S6: adopt least square fitting zero-bit V ₀(T _i), sensitivity S _c(T _i), quadratic nonlinearity coefficient C ₂(T _i), cubic non-linearity coefficient C ₃(T _i) and described digital temperature signal value TEMP (T _i) polynomial function relation, obtain penalty coefficient p _{0, V0}, p _{1, V0}, p _{2, V0}, p _{3, V0}, p _{0, Sc}, p _{1, Sc}, p _{2, Sc}, p _{3, Sc}, p _{0, C2}, p _{1, C2}, p _{0, C3}, p _{1, C3};

Wherein p _{0, V0}, p _{0, Sc}, p _{0, C2}, p _{0, C3}represent the zero-bit of benchmark, sensitivity, quadratic nonlinearity coefficient, cubic non-linearity coefficient respectively; p _{1, V0}, p _{1, Sc}, p _{1, C2}, p _{1, C3}represent the temperature variation coefficient of first order of zero-bit, sensitivity, quadratic nonlinearity coefficient, cubic non-linearity coefficient respectively; p _{2, V0}, p _{2, Sc}represent the temperature variation quadratic coefficients of zero-bit, sensitivity; p _{3, V0}, p _{3, Sc}represent temperature variation three ordered coefficients of zero-bit, sensitivity;

S7: the zero-bit V obtaining digital obliquity sensor according to above-mentioned penalty coefficient ₀, sensitivity S _c, quadratic nonlinearity coefficient C ₂with cubic non-linearity coefficient C ₃temperature compensation value.

Further, in step s 4 which according to formula V (T _i, θ _j)=V ₀(T _i)+B ₁(T _i) θ _j+ B ₂(T _i) θ _j ²+ B ₃(T _i) θ _j ³+ ε _vcarry out least square fitting; Wherein, V ₀(T _i) be temperature spot T _iunder zero-bit, B ₁(T _i), B ₂(T _i), B ₃(T _i) be respectively temperature spot T _iunder angle once, secondary and cubic term coefficient, ε _vfor voltage regression criterion.

By S _c(T _i)=B ₁(T _i) 180/ π (unit: V/g, π are circular constant) can calculate temperature spot T _iunder sensitivity S _c(T _i).

Further, formula is pressed in step s 5 $\mathrm{\θ}=(a+C_2\left(T_i\right)\·a^2+C_3\left(T_i\right)\·a^3)\frac{180}{\mathrm{\π}}+{\mathrm{\ϵ}}_{\mathrm{\θ}}$ Carry out least square fitting; Wherein, C ₂(T _i), C ₃(T _i) represent each temperature spot T respectively _iunder quadratic nonlinearity co-efficient measurements and cubic non-linearity co-efficient measurements; ε _θrepresent angle calculation regression criterion.

Further, in step s 6, least square fitting is carried out according to following formula,

$V_0\left(T_i\right)=p_{0,V0}+p_{1,V0}\·\left[\mathrm{TEMP}\left(T_i\right)\right]+p_{2,V0}\·{\left[\mathrm{TEMP}\left(T_i\right)\right]}^2+p_{3,V0}\·{\left[\mathrm{TEMP}\left(T_i\right)\right]}^3+{\mathrm{\ϵ}}_{V_0};$

S _c(T _i)＝p _0,Sc+p _1,Sc·[TEMP(T _i)]+p _2,Sc·[TEMP(T _i)] ²+p _3,Sc·[TEMP(T _i)] ³+ε _Sc；

$C_2\left(T_i\right)=p_{0,C2}+p_{1,C2}\·\left[\mathrm{TEMP}\left(T_i\right)\right]+{\mathrm{\ϵ}}_{C_2};C_3\left(T_i\right)=p_{0,C3}+p_{1,C3}\·\left[\mathrm{TEMP}\left(T_i\right)\right]+{\mathrm{\ϵ}}_{C_3};$

Wherein p _{0, V0}, p _{0, Sc}, p _{0, C2}, p _{0, C3}represent benchmark (refer to 0 DEG C time) zero-bit, sensitivity, quadratic nonlinearity coefficient, cubic non-linearity coefficient respectively; p _{1, V0}, p _{1, Sc}, p _{1, C2}, p _{1, C3}represent the temperature variation coefficient of first order of zero-bit, sensitivity, quadratic nonlinearity coefficient, cubic non-linearity coefficient respectively; p _{2, V0}, p _{2, Sc}represent the temperature variation quadratic coefficients of zero-bit, sensitivity respectively; p _{3, V0}, p _{3, Sc}represent temperature variation three ordered coefficients of zero-bit, sensitivity respectively; for temperature compensation parameter fitting residual error, wherein and ε _scthe principal element affecting angle error after temperature compensation, affect temperature consistency, ε _scthen affect angle agreement.

In the present invention, the nonlinear temperature compensation technique that temperature compensation module and nonlinearity compensation module adopt compensate for that temperature compensation error in existing similar technique is large, the obvious shortcoming of nonlinearity erron; Because the present invention is not limited to the sensor sensing acceleration of sinusoidal form and the relation (i.e. a=gsin (θ π/180)) at pitch angle, analyze quadratic nonlinearity error and the cubic non-linearity error of remarkable existence wherein, consider the appreciable impact of temperature on MEMS simultaneously, and this is proposed to concrete exercisable nonlinear temperature compensation method, therefore the present invention is more accurate to nonlinear temperature error compensation.The present invention simultaneously reduces system composition to greatest extent under the prerequisite of assurance function.Obliquity sensor of the present invention has that cost is low, measuring accuracy is high, anti-vibrating and impact, be applicable to the advantage of wide temperature harsh and unforgiving environments, is particularly useful for the measurement of high speed railway track superelevation.

Accompanying drawing explanation

Fig. 1 is the comprising modules structural representation of digital tilt angle sensor provided by the invention;

The specific implementation process flow diagram of the temperature compensation that Fig. 2 adopts for digital tilt angle sensor provided by the invention;

Fig. 3 is the specific implementation process flow diagram of the non-linear compensation method that digital tilt angle sensor provided by the invention adopts.

Embodiment

In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.In addition, if below in described each embodiment of the present invention involved technical characteristic do not form conflict each other and just can mutually combine.

Obliquity sensor of the present invention has that cost is low, measuring accuracy is high, anti-vibrating and impact, be applicable to the advantage of wide temperature harsh and unforgiving environments, is particularly useful for the measurement of high speed railway track superelevation.As shown in Figure 1, a kind of high-precision digital obliquity sensor of nonlinear temperature compensation technique that adopts comprises power module of voltage regulation 101, MEMS sensor 102, filtration module 103, microprocessor 104, reference power supply module 105 and communication module 106, and wherein microprocessor includes temperature compensation module 104-1, A/D conversion module 104-2 and nonlinearity compensation module 104-3.Power module of voltage regulation 101 provides stable power supply for other modules 102 ~ 106 above-mentioned, and adapts to the external power input of wide region, the positive and negative reversal connection of protection power supply; The temperature digital signal that first output terminal of MEMS sensor 102 exports carries out temperature compensation calculating to the temperature compensation module 104-1 of microprocessor 104, calculates gained and obtains 104-3 by nonlinearity compensation module; The A/D conversion module 104-2 being transferred to microprocessor 104 after filtration module 103 filtering of the analog voltage signal containing obliquity information of the second output terminal output of MEMS sensor 102, conversion gained is obtained by nonlinearity compensation module 104-3 containing the voltage digital signal of obliquity information; The temperature compensation parameter that the nonlinearity compensation module 104-3 of microprocessor 104 utilizes temperature compensation module 104-1 to calculate carries out nonlinear compensation to the voltage digital signal containing obliquity information, and the measurement of dip angle value digital signal after compensating is exported to communication module 106; Communication module 106 provides format conversion and interface for measurement of dip angle value digital signal exports user to; Reference power supply module 105 provides accurate reference voltage for A/D conversion module 104-2 in microprocessor 104 carries out A/D conversion.

In the present invention, MEMS sensor 102 can adopt model to be the micro sensing chip of SCA103T.Remodeling micro sensing chip resolution is high, good stability.Microprocessor 104 can adopt model to be the single-chip microcomputer of ADuC845; This model single-chip microcomputer is integrated with 24 high resolving power sigma-delta type ADC with more corresponding functions in one, and greatly reduce the composition of metering circuit, be conducive to the volume reducing sensor, therefore this type single-chip microcomputer is particularly suitable for precision measurement.Reference power supply module 105 can adopt model to be high precision, the ultra-low noise reference voltage source of ADR4525, and its initial output voltage error is only maximum ± and 0.02%, output voltage noise is 1.25 μ Vp-p only, and temperature coefficient is less than 2ppm/ DEG C.Power module of voltage regulation 101 can adopt model to be the low-dropout regulator of LT1962.Communication module 106 can adopt model to be the RS-232 driving chip of MAX232, and the UART TTL signal that microprocessor 104 exports by this chip is converted to serial RS-232 formatted output.

In embodiments of the present invention, principle of work of the present invention is as follows: the power module of voltage regulation 101 being core by low noise low-dropout regulator provides stable power supply for other modules of digital tilt angle sensor, ensure that the best electrical specification of other modules such as MEMS sensor 102 grade under various power supply input condition; MEMS sensor 102 carries temperature sensor, reduces system composition, and ensures that the temperature of temperature sensor sensitivity presses close to measuring tempeature most; Adopt the microprocessor 104 being integrated with 24 high resolving power sigma-delta type A/D conversion module 104-2 to improve Measurement Resolution to greatest extent, reduce adjunct circuit noise; Reference power supply module 105 has high precision and ultra-low noise simultaneously, ensure that A/D conversion module 104-2 works accurately, stablizes.Communication module 106 provides format conversion and interface for measurement of dip angle value digital signal that microprocessor 104 exports exports user to.

The present invention is a kind of high accuracy number inclination angle sensor device adopting nonlinear temperature compensation technique; This device comprises MEMS sensor, filtration module, microprocessor, communication module, reference power supply module and power module of voltage regulation; Wherein microprocessor includes A/D conversion module, temperature compensation module and nonlinearity compensation module.

Power module of voltage regulation provides stable electricity for other modules above-mentioned; MEMS sensor exports the analog voltage signal and temperature digital signal that contain obliquity information, analog voltage signal is through the A/D conversion module of filtration module input microprocessor, the temperature compensation module of temperature digital signal input microprocessor, in the microprocessor, the output of A/D conversion module and temperature compensation module exports measurement of dip angle value digital signal after nonlinearity compensation module is comprehensive; The A/D conversion that reference power supply module is microprocessor provides Precision reference power supply; Described communication module provides format conversion and interface for measurement of dip angle value digital signal that microprocessor exports exports user to.

As one embodiment of the present of invention, MEMS sensor exports containing the analog voltage signal of obliquity information and the temperature digital signal of temperature information, analog voltage signal is through the A/D conversion module of filtration module input microprocessor, the temperature compensation module of temperature digital signal input microprocessor, in the microprocessor, the output of A/D conversion module and temperature compensation module exports and has high resolving power, high precision dip measured value digital signal after nonlinearity compensation module is comprehensive; Described communication module provides format conversion and interface for measurement of dip angle value digital signal that microprocessor exports exports user to; The A/D conversion that reference power supply module is microprocessor provides Precision reference power supply; Power module of voltage regulation can adapt to the externally fed of wide region, for other modules provide stable supply voltage, and the positive and negative reversal connection of protection externally fed.

Wherein MEMS sensor module can adopt model to be the micro sensing chip of SCA103T.Microprocessor adopts model to be the single-chip microcomputer of ADuC845.This model single-chip microcomputer is integrated with 24 high resolving power sigma-delta type ADC with more corresponding functions in one, and greatly reduce the composition of metering circuit, be conducive to the volume reducing sensor, therefore this type single-chip microcomputer is particularly suitable for precision measurement.

Accurate reference power supply module employing model is high precision, the low noise reference voltage source of ADR4525; Its initial output voltage error is only maximum ± and 0.02%, output voltage noise is 1.25 μ Vp-p only, and temperature coefficient is less than 2ppm/ DEG C.Power module of voltage regulation adopts model to be the low-dropout regulator of LT1962.Output interface and communication module adopt model to be the RS-232 driving chip of MAX232.

In embodiments of the present invention, the temperature compensation step that integrated in microprocessor temperature compensation module adopts is as follows:

(1) temperature of micro compensating module gathers the temperature digital signal that MEMS sensor module exports, and obtains sensor temperature ST through level and smooth, filtering;

(2) MEMS sensor zero-bit V is calculated respectively by the following polynomial equation about sensor temperature ST (1) ~ (4) ₀, sensitivity S _c, quadratic nonlinearity coefficient C ₂, cubic non-linearity coefficient C ₃offset under temperature ST;

V ₀＝p _0,V0+p _1,V0·ST+p _2,V0·ST ²+p _3,V0·ST ³(1)

S _c＝p _0,Sc+p _1,Sc·ST+p _2,Sc·ST ²+p _3,Sc·ST ³(2)

C ₂＝p _0,C2+p _1,C2·ST (3)

C ₃＝p _0,C3+p _1,C3·ST (4)

In formula (1) ~ (4), p _{0, V0}, p _{0, Sc}, p _{0, C2}, p _{0, C3}represent benchmark (refer to 0 DEG C time) zero-bit, sensitivity, quadratic nonlinearity coefficient, cubic non-linearity coefficient respectively; p _{1, V0}, p _{1, Sc}, p _{1, C2}, p _{1, C3}represent the temperature variation coefficient of first order of zero-bit, sensitivity, quadratic nonlinearity coefficient, cubic non-linearity coefficient respectively; p _{2, V0}, p _{2, Sc}represent the temperature variation quadratic coefficients of zero-bit, sensitivity respectively; p _{3, V0}, p _{3, Sc}represent temperature variation three ordered coefficients of zero-bit, sensitivity respectively.

In embodiments of the present invention, the non-linear compensation method step that integrated in microprocessor nonlinearity compensation module adopts is as follows:

(1) microprocessor nonlinearity compensation module gathers the digital voltage signal containing obliquity information that A/D conversion module exports, and this signal obtains the magnitude of voltage SV containing obliquity information after microprocessor digital filtering, smoothing processing;

(2) nonlinearity compensation module obtains the MEMS sensor zero-bit V that temperature compensation module exports ₀, sensitivity S _c, quadratic nonlinearity coefficient C ₂, cubic non-linearity coefficient C ₃, calculate measurement of dip angle value θ by following formula (5) ~ (6) _survey(unit: degree)

$a=\frac{\mathrm{SV}-V_0}{S_c\·g}---\left(5\right)$

Wherein π is circular constant; G is local gravitational acceleration; A is intermediate quantity.

The specific implementation step of the nonlinear temperature compensation method of a kind of digital obliquity sensor provided by the invention is as follows:

(1) temperature control horizontal precision turntable is arranged on by without the inclination angle sensor device compensated, obliquity sensor is energized, arrange that inclination angle sensor device exports that described A/D conversion module exports containing the voltage digital signal of obliquity information and described temperature digital signal for computing machine or host computer collection;

(2) within the scope of full temperature to be measured, evenly choose several test temperature points T _i, adjacent temperature spot interval should be not more than 10 DEG C, at each test temperature point T _iunder, temperature control turntable the temperature inside the box should be incubated the sufficiently long time, to make inclination angle sensor device internal and external temperature in a basic balance;

(3) at each test temperature point T _iunder, within the scope of-8 ~ 8 ° of inclination angles, evenly choose several measurement of dip angle points θ _j, adjacent measurement of dip angle point angle intervals should be not more than 1 °;

(4) at each test temperature point T _iunder, after the turntable incubator insulation long enough time, control turntable and rotate to each measurement of dip angle point θ _j; Now gather the described voltage digital signal value V (T containing obliquity information _i, θ _j) and described digital temperature signal value TEMP (T _i);

(5) at each temperature spot T _i, adopt least square fitting to gather digital voltage V (T by following formula (7) _i, θ _j) and inclination angle theta _jcubic polynomial:

V(T _i,θ _j)＝V ₀(T _i)+B ₁(T _i)·θ _j+B ₂(T _i)·θ _j ²+B ₃(T _i)·θ _j ³+ε _V(7)

Each temperature spot T can be obtained _iunder zero measurement value V ₀(T _i) and sensitivity measure value S _c(T _i) (=B ₁(T _i) 180/ π, unit: V/g), and then the acceleration calculation value a (T under each temperature spot, each demarcation angle can be calculated _i, θ _j), computing formula is as follows:

$a(T_i,{\mathrm{\θ}}_j)=\frac{V(T_i,{\mathrm{\θ}}_j)-V_0\left(T_i\right)}{g\·S_c\left(T_i\right)}---\left(8\right)$

ε in formula (7) _vrepresent voltage regression criterion.

(6) the least square method each temperature spot T of matching is as follows adopted _iunder inclination value θ _jwith acceleration calculation value a (T _i, θ _j) three relations:

$\mathrm{\θ}=(a+C_2\left(T_i\right)\·a^2+C_3\left(T_i\right)\·a^3)\frac{180}{\mathrm{\π}}+{\mathrm{\ϵ}}_{\mathrm{\θ}}---\left(9\right)$

Each temperature spot T can be obtained _iunder quadratic nonlinearity co-efficient measurements C ₂(T _i), cubic non-linearity co-efficient measurements C ₃(T _i).ε in formula (9) _θrepresent angle calculation regression criterion.

(7) least square method (10) ~ (13) matching zero-bit V is as follows adopted ₀(T _i), sensitivity S _c(T _i), quadratic nonlinearity coefficient C ₂(T _i), cubic non-linearity coefficient C ₃(T _i) with gather digital temperature TEMP (T _i) polynomial function relation:

$V_0\left(T_i\right)=p_{0,V0}+p_{1,V0}\·\left[\mathrm{TEMP}\left(T_i\right)\right]+p_{2,V0}\·{\left[\mathrm{TEMP}\left(T_i\right)\right]}^2+p_{3,V0}\·{\left[\mathrm{TEMP}\left(T_i\right)\right]}^3+{\mathrm{\ϵ}}_{V_0}---\left(10\right)$

S _c(T _i)＝p _0,Sc+p _1,Sc·[TEMP(T _i)]+p _2,Sc·[TEMP(T _i)] ²+p _3,Sc·[TEMP(T _i)] ³+ε _Sc(11)

$C_2\left(T_i\right)=p_{0,C2}+p_{1,C2}\·\left[\mathrm{TEMP}\left(T_i\right)\right]+{\mathrm{\ϵ}}_{C_2}---\left(12\right)$

$C_3\left(T_i\right)=p_{0,C3}+p_{1,C3}\·\left[\mathrm{TEMP}\left(T_i\right)\right]+{\mathrm{\ϵ}}_{C_3}---\left(13\right)$

In above formula (10) ~ (13) for temperature compensation parameter fitting residual error, wherein and ε _scthe principal element affecting angle error after temperature compensation, affect temperature consistency, ε _scthen affect angle agreement.

(8) above-mentioned steps (1) ~ (7) provide and to obtain in temperature compensation formula (1) ~ (4) penalty coefficient on the right of equation and (comprise p _{0, V0}, p _{1, V0}, p _{2, V0}, p _{3, V0}, p _{0, Sc}, p _{1, Sc}, p _{2, Sc}, p _{3, Sc}, p _{0, C2}, p _{1, C2}, p _{0, C3}, p _{1, C3}) a kind of effective ways of numerical value;

(9) revise temperature compensation module program in microprocessor, make microprocessor step calculating MEMS sensor zero-bit V as follows ₀, sensitivity S _c, quadratic nonlinearity coefficient C ₂, cubic non-linearity coefficient C ₃temperature compensation value:

(9-1) temperature of micro compensating module gathers the temperature digital signal of MEMS sensor module output and after the disposal of gentle filter, obtains sensor temperature smooth value ST;

(9-2) MEMS sensor zero-bit V is calculated respectively by the following polynomial equation about sensor temperature smooth value ST (14) ~ (17) ₀, sensitivity S _c, quadratic nonlinearity coefficient C ₂, cubic non-linearity coefficient C ₃offset under temperature ST;

V ₀＝p _0,V0+p _1,V0·ST+p _2,V0·ST ²+p _3,V0·ST ³(14)

S _c＝p _0,Sc+p _1,Sc·ST+p _2,Sc·ST ²+p _3,Sc·ST ³(15)

C ₂＝p _0,C2+p _1,C2·ST (16)

C ₃＝p _0,C3+p _1,C3·ST (17)

P in formula (14) ~ (17) _{0, V0}, p _{0, Sc}, p _{0, C2}, p _{0, C3}represent benchmark (refer to 0 DEG C time) zero-bit, sensitivity, quadratic nonlinearity coefficient, cubic non-linearity coefficient respectively; p _{1, V0}, p _{1, Sc}, p _{1, C2}, p _{1, C3}represent the temperature variation coefficient of first order of zero-bit, sensitivity, quadratic nonlinearity coefficient, cubic non-linearity coefficient respectively; p _{2, V0}, p _{2, Sc}represent the temperature variation quadratic coefficients of zero-bit, sensitivity respectively; p _{3, V0}, p _{3, Sc}represent temperature variation three ordered coefficients of zero-bit, sensitivity respectively; These coefficients above-mentioned are obtained by described step (1) ~ (8).

(10) revise nonlinearity compensation module program in microprocessor, make microprocessor step calculating measurement of dip angle value as follows:

(10-1) microprocessor nonlinearity compensation module gathers the digital voltage signal containing obliquity information that A/D conversion module exports, and this signal obtains the magnitude of voltage SV containing obliquity information after microprocessor digital filtering, smoothing processing;

(10-2) nonlinearity compensation module obtains the MEMS sensor zero-bit V that temperature compensation module exports ₀, sensitivity S _c, quadratic nonlinearity coefficient C ₂, cubic non-linearity coefficient C ₃, calculate measurement of dip angle value θ by following formula (18) ~ (19) _survey(unit: degree)

$a=\frac{\mathrm{SV}-V_0}{g\·S_c}---\left(18\right)$

Adopt technical scheme belonging to the present invention, the beneficial effect produced improves full temperature angle measurement accuracy, reduces principle nonlinearity erron.The present invention can realize full temperature angle measurement accuracy and be better than 0.005 °, and nonlinearity erron is less than 0.04%FS.

By the present invention, the described method of formula (14) is adopted to calculate zero-bit V ₀offset can effectively reduce angle error because Zero Drift of Temperature causes than with an order polynomial compensation method.

By the present invention, adopt formula (15) described method meter sensitivity S _coffset can effectively reduce because Sensitivity Temperature floats the angle error caused than using the compensation formula provided in SCA103T chip handbook.

By the present invention, adopt the described non-linear compensation method in formula (16) ~ (19) than directly adopting inverse sine formulae discovery measurement of dip angle value (i.e. θ _survey=(arcsina) * 180/ π) carry out effectively reducing non-linear angle error.

In the present invention, the nonlinear temperature compensation technique that temperature compensation module and nonlinearity compensation module adopt compensate for that temperature compensation error in existing similar technique is large, the obvious shortcoming of nonlinearity erron; More accurate to nonlinear temperature error compensation.The present invention simultaneously reduces system composition to greatest extent under the prerequisite of assurance function.Obliquity sensor of the present invention has that cost is low, measuring accuracy is high, anti-vibrating and impact, be applicable to the advantage of wide temperature harsh and unforgiving environments, is particularly useful for the measurement of high speed railway track superelevation.

In embodiments of the present invention, temperature compensation module 104-1 principle of work is as follows:

S201 first in the steps below (201-1) ~ (201-7) obtains temperature compensation coefficient and (comprises p _{0, V0}, p _{1, V0}, p _{2, V0}, p _{3, V0}, p _{0, Sc}, p _{1, Sc}, p _{2, Sc}, p _{3, Sc}, p _{0, C2}, p _{1, C2}, p _{0, C3}, p _{1, C3}), and write temperature compensation module in microprocessor.

(201-1) temperature control horizontal precision turntable is arranged on by without the inclination angle sensor device compensated, obliquity sensor is energized, arrange that inclination angle sensor device exports that described A/D conversion module exports containing the voltage digital signal of obliquity information and described temperature digital signal for computing machine or host computer collection;

(201-2) within the scope of full temperature to be measured, evenly choose several test temperature points T _i, adjacent temperature spot interval should be not more than 10 DEG C, at each test temperature point T _iunder, temperature control turntable the temperature inside the box should be incubated the sufficiently long time, to make inclination angle sensor device internal and external temperature in a basic balance;

(201-3) at each test temperature point T _iunder, within the scope of-8 ~ 8 ° of inclination angles, evenly choose several measurement of dip angle points θ _j, adjacent measurement of dip angle point angle intervals should be not more than 1 °;

(201-4) at each test temperature point T _iunder, after the turntable incubator insulation long enough time, control turntable and rotate to each measurement of dip angle point θ _j; Now gather the described voltage digital signal value V (T containing obliquity information _i, θ _j) and described digital temperature signal value TEMP (T _i);

(201-5) at each temperature spot T _i, adopt least square fitting to gather digital voltage V (T by following formula (7) _i, θ _j) and inclination angle theta _jcubic polynomial:

V(T _i,θ _j)＝V ₀(T _i)+B ₁(T _i)·θ _j+B ₂(T _i)·θ _j ²+B ₃(T _i)·θ _j ³+ε _V(20)

Each temperature spot T can be obtained _iunder zero measurement value V ₀(T _i) and sensitivity measure value S _c(T _i) (=B ₁(T _i) 180/ π, unit: V/g), and then the acceleration calculation value a (T under each temperature spot, each demarcation angle can be calculated _i, θ _j), computing formula is as follows:

$a(T_i,{\mathrm{\θ}}_j)=\frac{V(T_i,{\mathrm{\θ}}_j)-V_0\left(T_i\right)}{g\·S_c\left(T_i\right)}---\left(21\right)$

ε in formula (20) _vrepresent voltage regression criterion.According to test of many times result, usually | ε _v| <0.15mV.

(201-6) the least square method each temperature spot T of matching is as follows adopted _iunder inclination value θ _jwith acceleration calculation value a (T _i, θ _j) three relations:

$\mathrm{\&theta;}=(a+C_2\left(T_i\right)\&CenterDot;a^2+C_3\left(T_i\right)\&CenterDot;a^3)\frac{180}{\mathrm{\&pi;}}+{\mathrm{\&epsiv;}}_{\mathrm{\&theta;}}---\left(22\right)$

Each temperature spot T can be obtained _iunder quadratic nonlinearity co-efficient measurements C ₂(T _i), cubic non-linearity co-efficient measurements C ₃(T _i).ε in formula (9) _θrepresent angle calculation regression criterion.According to test of many times result, usually | ε _θ|≤0.0005 °.

(201-7) least square method (10) ~ (13) matching zero-bit V is as follows adopted ₀(T _i), sensitivity S _c(T _i), quadratic nonlinearity coefficient C ₂(T _i), cubic non-linearity coefficient C ₃(T _i) with gather digital temperature TEMP (T _i) polynomial function relation:

$V_0\left(T_i\right)=p_{0,V0}+p_{1,V0}\&CenterDot;\left[\mathrm{TEMP}\left(T_i\right)\right]+p_{2,V0}\&CenterDot;{\left[\mathrm{TEMP}\left(T_i\right)\right]}^2+p_{3,V0}\&CenterDot;{\left[\mathrm{TEMP}\left(T_i\right)\right]}^3+{\mathrm{\&epsiv;}}_{V_0}---\left(23\right)$

S _c(T _i)＝p _0,Sc+p _1,Sc·[TEMP(T _i)]+p _2,Sc·[TEMP(T _i)] ²+p _3,Sc·[TEMP(T _i)] ³+ε _Sc(24)

$C_2\left(T_i\right)=p_{0,C2}+p_{1,C2}\&CenterDot;\left[\mathrm{TEMP}\left(T_i\right)\right]+{\mathrm{\&epsiv;}}_{C_2}---\left(25\right)$

$C_3\left(T_i\right)=p_{0,C3}+p_{1,C3}\&CenterDot;\left[\mathrm{TEMP}\left(T_i\right)\right]+{\mathrm{\&epsiv;}}_{C_3}---\left(26\right)$

In above formula (10) ~ (13) for temperature compensation parameter fitting residual error, wherein and ε _scthe principal element affecting angle error after temperature compensation, affect temperature consistency, ε _scthen affect angle agreement.

S202 temperature of micro compensating module gathers the temperature digital signal of MEMS sensor module output and after the disposal of gentle filter, obtains sensor temperature smooth value ST;

S203 calculates MEMS sensor zero-bit V respectively by the following polynomial equation about sensor temperature smooth value ST (27) ~ (30) ₀, sensitivity S _c, quadratic nonlinearity coefficient C ₂, cubic non-linearity coefficient C ₃offset under temperature ST;

V ₀＝p _0,V0+p _1,V0·ST+p _2,V0·ST ²+p _3,V0·ST ³(27)

S _c＝p _0,Sc+p _1,Sc·ST+p _2,Sc·ST ²+p _3,Sc·ST ³(28)

C ₂＝p _0,C2+p _1,C2·ST (29)

C ₃＝p _0,C3+p _1,C3·ST (30)

The temperature compensation that temperature compensation module adopts utilizes three rank polynomial expression (27) ~ (28) to calculate MEMS sensor zero-bit V ₀, sensitivity S _c, closest to MEMS sensor self-characteristic.Utilize accurate temperature control turntable to implement temperature compensation parameter to inclination angle sensor device and demarcate the accuracy that ensure that measurement; Least square fitting technology (formula (20), (22) ~ (26)) calculation compensation parameter is utilized to ensure that unbiasedness and the optimality of parameter estimation.

In embodiments of the present invention, nonlinearity compensation module 104-1 principle of work is as follows:

S301 microprocessor nonlinearity compensation module gathers the digital voltage signal containing obliquity information that A/D conversion module exports, and obtains the magnitude of voltage SV containing obliquity information to this signal after making digital filtering, smoothing processing;

S302 nonlinearity compensation module obtains the MEMS sensor zero-bit V that temperature compensation module exports ₀, sensitivity S _c, quadratic nonlinearity coefficient C ₂, cubic non-linearity coefficient C ₃;

S303 calculates measurement of dip angle value θ by following formula (18) ~ (19) _survey(unit: degree) $a=\frac{\mathrm{SV}-V_0}{g\&CenterDot;S_c}---\left(31\right);$

The three rank polynomial expressions (32) that nonlinearity compensation module 104-3 adopts compensate Nonlinear Error of Transducer, consider the quadratic nonlinearity error of sensor, the accurately calculating of cubic non-linearity error, the method calculating inclination angle than the inverse sine trigonometric function method in existing document, patent can greatly reduce nonlinearity erron.

Those skilled in the art will readily understand; the foregoing is only preferred embodiment of the present invention; not in order to limit the present invention, all any amendments done within the spirit and principles in the present invention, equivalent replacement and improvement etc., all should be included within protection scope of the present invention.

Claims (7)

1. a digital obliquity sensor, it is characterized in that, comprise MEMS sensor (102), filtration module (103), microprocessor (104), communication module (106), reference power supply module (105) and power module of voltage regulation (101);

Described microprocessor (104) comprises temperature compensation module (104-1), A/D conversion module (104-2) and nonlinearity compensation module (104-3);

The power input of the power input of the power input of described MEMS sensor (102), the power input of described filtration module (103), described microprocessor (104), described reference power supply module (105) and described communication module (106) all connects power module of voltage regulation (101);

The input end of described temperature compensation module (104-1) is connected to the first output terminal of described MEMS sensor (102), the signal input part of described filtration module (103) is connected to the second output terminal of described MEMS sensor (102), the first input end of described A/D conversion module (104-2) is connected to the output terminal of described filtration module (103), second input end of described A/D conversion module (104-2) is connected to the output terminal of described reference power supply module (105), the first input end of described nonlinearity compensation module (104-3) is connected to the output terminal of described temperature compensation module (104-1), second input end of described nonlinearity compensation module (104-3) is connected to the output terminal of described A/D conversion module (104-2), the input end of described communication module (106) is connected to the output terminal of described nonlinearity compensation module (104-3).

2. digital obliquity sensor as claimed in claim 1, it is characterized in that, during work, the temperature digital signal that first output terminal of described MEMS sensor (102) exports carries out temperature compensation calculating to described temperature compensation module (104-1), calculates gained and is obtained by nonlinearity compensation module (104-3); The analog voltage signal containing obliquity information that second output terminal of described MEMS sensor (102) exports is transferred to described A/D conversion module (104-2) after described filtration module (103) filtering, and conversion gained is obtained by nonlinearity compensation module (104-3) containing the voltage digital signal of obliquity information; The temperature compensation parameter that described nonlinearity compensation module (104-3) utilizes described temperature compensation module (104-1) to calculate carries out nonlinear compensation to containing the voltage digital signal of obliquity information, and the measurement of dip angle value digital signal after compensating is exported to communication module (106) exports to user by described communication module (106).

3. digital obliquity sensor as claimed in claim 1, is characterized in that, described microprocessor adopts model to be the single-chip microcomputer of ADuC845.

4., based on a nonlinear temperature compensation method for digital obliquity sensor according to claim 1, it is characterized in that, comprise the steps:

S1: evenly choose several test temperature points T within the scope of full temperature to be measured _i, adjacent temperature spot interval should be not more than 10 DEG C, at each test temperature point T _iunder, temperature control turntable the temperature inside the box should be incubated the sufficiently long time to make inclination angle sensor device internal and external temperature in a basic balance;

S2: at each test temperature point T _iunder, within the scope of-8 ° ~ 8 ° inclination angles, evenly choose several measurement of dip angle points θ _j, adjacent measurement of dip angle point angle intervals should be not more than 1 °;

S3: at each test temperature point T _iunder, after the temperature control turntable case inside holding long enough time, control turntable and rotate to each measurement of dip angle point θ _j, now gather the voltage digital signal V (T containing obliquity information that in inclination angle sensor device, A/D conversion module exports _i, θ _j) and temperature digital signal TEMP (T _i);

S4: at each temperature spot T _i, adopt least square fitting to gather digital voltage V (T _i, θ _j) and inclination angle theta _jcubic polynomial, obtain each temperature spot T _iunder zero measurement value V ₀(T _i) and sensitivity measure value S _c(T _i), and according to described zero measurement value V ₀(T _i) and described sensitivity measure value S _c(T _i) obtain acceleration calculation value under each temperature spot, each demarcation angle $a(T_i,{\mathrm{\&theta;}}_j)=\frac{V(T_i,{\mathrm{\&theta;}}_j)-V_0\left(T_i\right)}{g\&CenterDot;S_c\left(T_i\right)};$

S5: adopt each temperature spot T of least square fitting _iunder inclination value θ _jwith acceleration calculation value a (T _i, θ _j) three relations, obtain each temperature spot T _iunder quadratic nonlinearity co-efficient measurements C ₂(T _i), cubic non-linearity co-efficient measurements C ₃(T _i);

S6: adopt least square fitting zero-bit V ₀(T _i), sensitivity S _c(T _i), quadratic nonlinearity coefficient C ₂(T _i), cubic non-linearity coefficient C ₃(T _i) and described digital temperature signal value TEMP (T _i) polynomial function relation, obtain penalty coefficient p _{0, V0}, p _{1, V0}, p _{2, V0}, p _{3, V0}, p _{0, Sc}, p _{1, Sc}, p _{2, Sc}, p _{3, Sc}, p _{0, C2}, p _{1, C2}, p _{0, C3}, p _{1, C3};

Wherein p _{0, V0}, p _{0, Sc}, p _{0, C2}, p _{0, C3}represent the zero-bit of benchmark, sensitivity, quadratic nonlinearity coefficient, cubic non-linearity coefficient respectively; p _{1, V0}, p _{1, Sc}, p _{1, C2}, p _{1, C3}represent the temperature variation coefficient of first order of zero-bit, sensitivity, quadratic nonlinearity coefficient, cubic non-linearity coefficient respectively; p _{2, V0}, p _{2, Sc}represent the temperature variation quadratic coefficients of zero-bit, sensitivity; p _{3, V0}, p _{3, Sc}represent temperature variation three ordered coefficients of zero-bit, sensitivity;

S7: the zero-bit V obtaining digital obliquity sensor according to above-mentioned penalty coefficient ₀, sensitivity S _c, quadratic nonlinearity coefficient C ₂with cubic non-linearity coefficient C ₃temperature compensation value.

5. nonlinear temperature compensation method as claimed in claim 4, is characterized in that, in step s 4 which according to formula V (T _i, θ _j)=V ₀(T _i)+B ₁(T _i) θ _j+ B ₂(T _i) θ _j ²+ B ₃(T _i) θ _j ³+ ε _vcarry out least square fitting; Wherein, V ₀(T _i) be temperature spot T _iunder zero-bit, B ₁(T _i), B ₂(T _i), B ₃(T _i) be respectively temperature spot T _iunder angle once, secondary and cubic term coefficient, ε _vfor voltage regression criterion.

6. nonlinear temperature compensation method as claimed in claim 4, is characterized in that, press formula in step s 5 $\mathrm{\&theta;}=(a+C_2\left(T_i\right)\&CenterDot;a^2+C_3\left(T_i\right)\&CenterDot;a^3)\frac{180}{\mathrm{\&pi;}}+{\mathrm{\&epsiv;}}_{\mathrm{\&theta;}}$ Carry out least square fitting; Wherein, C ₂(T _i), C ₃(T _i) represent each temperature spot T respectively _iunder quadratic nonlinearity co-efficient measurements and cubic non-linearity co-efficient measurements; ε _θrepresent angle calculation regression criterion.

7. nonlinear temperature compensation method as claimed in claim 4, is characterized in that, in step s 6, carry out least square fitting according to following formula, $V_0\left(T_i\right)=p_{0,V0}+p_{1,V0}\&CenterDot;\left[\mathrm{TEMP}\left(T_i\right)\right]+p_{2,V0}\&CenterDot;{\left[\mathrm{TEMP}\left(T_i\right)\right]}^2+p_{3,V0}\&CenterDot;{\left[\mathrm{TEMP}\left(T_i\right)\right]}^3+{\mathrm{\&epsiv;}}_{V_0};$

S _c(T _i)＝p _0,Sc+p _1,Sc·[TEMP(T _i)]+p _2,Sc·[TEMP(T _i)] ²+p _3,Sc·[TEMP(T _i)] ³+ε _Sc

$C_2\left(T_i\right)=p_{0,C2}+p_{1,C2}\&CenterDot;\left[\mathrm{TEMP}\left(T_i\right)\right]+{\mathrm{\&epsiv;}}_{C_2};$ $C_3\left(T_i\right)=P_{0,C3}+p_{1,C3}\&CenterDot;\left[\mathrm{TEMP}\left(T_i\right)\right]+{\mathrm{\&epsiv;}}_{C_3};$ Wherein p _{0, V0}, p _{0, Sc}, p _{0, C2}, p _{0, C3}represent benchmark (refer to 0 DEG C time) zero-bit, sensitivity, quadratic nonlinearity coefficient, cubic non-linearity coefficient respectively; p _{1, V0}, p _{1, Sc}, p _{1, C2}, p _{1, C3}represent the temperature variation coefficient of first order of zero-bit, sensitivity, quadratic nonlinearity coefficient, cubic non-linearity coefficient respectively; p _{2, V0}, p _{2, Sc}represent the temperature variation quadratic coefficients of zero-bit, sensitivity respectively; p _{3, V0}, p _{3, Sc}represent temperature variation three ordered coefficients of zero-bit, sensitivity respectively; for temperature compensation parameter fitting residual error, wherein and ε _scthe principal element affecting angle error after temperature compensation, affect temperature consistency, ε _scthen affect angle agreement.