CN203534650U

CN203534650U - Cloud transmission digital signal processing device with Coriolis mass flow meter

Info

Publication number: CN203534650U
Application number: CN201320656921.4U
Authority: CN
Inventors: 朱邱悦; 高瑞; 赵代岳; 张岩; 朱慧敏; 高璐璐
Original assignee: Shandong University
Current assignee: Shandong University
Priority date: 2013-10-23
Filing date: 2013-10-23
Publication date: 2014-04-09
Anticipated expiration: 2023-10-23

Abstract

The utility model discloses a digital signal processing device for cloud transmission of a Coriolis mass flowmeter, which comprises a Coriolis mass flowmeter. The magnetoelectric sensor transmits the collected signal to the differential amplifier circuit corresponding to the magnetoelectric sensor, and the differential amplifier circuit transmits the processed signal to the DSP through the AD sampling circuit corresponding to the differential amplifier circuit; The module communicates with the DSP. Use the GPS module to collect the location information of the Coriolis mass flowmeter, and use the GPRS remote network communication of the SIM300 module to realize the data transmission of the parameters obtained by the Coriolis mass flowmeter detection from the DSP to the cloud server, and from the cloud server to the mobile network terminal. The digital signal processing device is a high-precision, real-time Coriolis mass flowmeter signal processing device.

Description

A Coriolis mass flowmeter cloud transmission digital signal processing device

技术领域technical field

本实用新型涉及一种科里奥利质量流量计云传输数字信号处理装置。The utility model relates to a digital signal processing device for cloud transmission of a Coriolis mass flowmeter.

背景技术Background technique

由于科氏质量流量计是基于流体振动原理工作，管子振动频率受流体密度等影响，二次仪表测量量为合成波的相位差，且模拟电路对外界噪声比较敏感，因此降低了测量的精确度。为了提高科氏质量流量计的精度和抗干扰能力，国内外研发机构和工程师的普遍做法是将数字信号处理算法应用于科氏质量流量计信号处理过程中。例如，中国专利CN101832803B利用同步调制方法，通过过零比较方法计算得到振动频率，从而计算出相位差。该方法比较简单，但是计算精度不高。北京化工大学采用在DFT基础上引入了线性调频Z变换算法对信号进行频率跟踪，采用滑动Goertzel算法进行信号相位差的测量（林坤，科氏流量计的DSP算法研究及实现，北京化工大学，硕士学位论文，2008）。该方法频率和相位差精度高，但实时性欠佳。合肥工业大学提出采用归一化格型IIR自适应谱线增强器对科氏流量计信号进行增强和频率估计，采用加汉宁窗修正的离散傅里叶变换计算科氏流量计信号的时间差的方法（倪伟，科里奥利质量流量计数字信号处理方法的研究，合肥工业大学，博士学位论文，2004）。该方法实现了频率实时跟踪，但当非整周期采样时相位差计算存在频谱泄露问题，影响测量精度。Because the Coriolis mass flowmeter works based on the principle of fluid vibration, the vibration frequency of the tube is affected by the fluid density, etc., the secondary instrument measurement is the phase difference of the synthetic wave, and the analog circuit is sensitive to external noise, thus reducing the measurement accuracy . In order to improve the accuracy and anti-interference ability of Coriolis mass flowmeters, the common practice of R&D institutions and engineers at home and abroad is to apply digital signal processing algorithms to the signal processing process of Coriolis mass flowmeters. For example, the Chinese patent CN101832803B uses a synchronous modulation method to calculate the vibration frequency through a zero-crossing comparison method, thereby calculating the phase difference. This method is relatively simple, but the calculation accuracy is not high. Beijing University of Chemical Technology adopts the chirp-Z transform algorithm based on DFT to track the frequency of the signal, and uses the sliding Goertzel algorithm to measure the signal phase difference (Lin Kun, DSP algorithm research and implementation of Coriolis flowmeter, Beijing University of Chemical Technology, Master's Thesis, 2008). This method has high accuracy of frequency and phase difference, but poor real-time performance. Hefei University of Technology proposed to use a normalized lattice type IIR adaptive spectral line enhancer to enhance and estimate the frequency of the Coriolis flowmeter signal, and use the discrete Fourier transform with Hanning window correction to calculate the time difference of the Coriolis flowmeter signal Method (Ni Wei, Research on digital signal processing method of Coriolis mass flowmeter, Hefei University of Technology, doctoral dissertation, 2004). This method realizes real-time frequency tracking, but there is a problem of spectrum leakage in phase difference calculation when non-full-period sampling occurs, which affects measurement accuracy.

现有科氏质量流量计信号处理方法存在精度不高，实时性不强，成本较高的问题。因此，实用新型一种高精度、实时性强的低成本数字信号处理装置是当务之急。The existing Coriolis mass flowmeter signal processing method has the problems of low precision, low real-time performance and high cost. Therefore, it is urgent to develop a low-cost digital signal processing device with high precision and strong real-time performance.

实用新型内容Utility model content

为解决现有技术存在的不足，本实用新型公开了一种科里奥利质量流量计云传输数字信号处理装置，特别适用于信号频率变化速度快、相位差不断波动的一种数字信号处理装置。In order to solve the deficiencies in the prior art, the utility model discloses a digital signal processing device for cloud transmission of a Coriolis mass flowmeter, which is especially suitable for a digital signal processing device with fast signal frequency change speed and constant fluctuation of phase difference .

为实现上述目的，本实用新型的具体方案如下：In order to achieve the above object, the concrete scheme of the present utility model is as follows:

一种科里奥利质量流量计云传输数字信号处理装置，包括一个科氏质量流量计，科氏质量流量计自带两个磁电传感器、驱动器和恒流源，两个磁电传感器将采集到的信号传送给与磁电传感器相对应的差分放大电路，差分放大电路将处理后的信号通过与差分放大电路相对应的AD采样电路传送给DSP；A Coriolis mass flowmeter cloud transmission digital signal processing device, including a Coriolis mass flowmeter, the Coriolis mass flowmeter has two magnetoelectric sensors, drivers and constant current sources, and the two magnetoelectric sensors will collect The received signal is transmitted to the differential amplifier circuit corresponding to the magnetoelectric sensor, and the differential amplifier circuit transmits the processed signal to the DSP through the AD sampling circuit corresponding to the differential amplifier circuit;

所述驱动器与DSP通讯连接；The driver is communicated with the DSP;

所述恒流源与PT100相连，恒流源用于给PT100提供电压，PT100测量外界温度，PT100通过与之对应的AD采样电路与DSP相连。恒流源输入电源与dsp的电源模块相连。The constant current source is connected with PT100, and the constant current source is used to provide voltage for PT100, and PT100 measures the external temperature, and PT100 is connected with DSP through corresponding AD sampling circuit. The input power supply of the constant current source is connected with the power supply module of the dsp.

所述DSP还与SRAM、EEPROM、ePWM的输出、LCD、键盘及GPS模块相连；Described DSP is also connected with the output of SRAM, EEPROM, ePWM, LCD, keyboard and GPS module;

所述DSP还通过GPRS模块与云服务器相连，云服务器与移动终端相连。移动终端为手机、电脑等。The DSP is also connected to the cloud server through the GPRS module, and the cloud server is connected to the mobile terminal. The mobile terminal is a mobile phone, a computer, and the like.

所述科氏质量流量计为双U型管科氏质量流量计。The Coriolis mass flowmeter is a double U-tube Coriolis mass flowmeter.

所述GPRS模块包括SIM300模块。The GPRS module includes a SIM300 module.

本装置采用DSP作为主控制器，采用反馈型数字驱动模块进行数字驱动，利用GPS模块采集科氏质量流量计位置信息，采用SIM300模块的GPRS远程网络通信，实现科氏质量流量计检测获得的参数从DSP到云服务器，以及从云服务器到移动网络终端的数据传输。通过LCD和键盘实现人机交互功能，ePWM脉冲输出提供4～20mA电流输出。The device adopts DSP as the main controller, adopts the feedback digital drive module for digital drive, uses the GPS module to collect the position information of the Coriolis mass flowmeter, and uses the GPRS remote network communication of the SIM300 module to realize the parameters obtained by the detection of the Coriolis mass flowmeter. Data transmission from DSP to cloud server, and from cloud server to mobile network terminal. Human-computer interaction is realized through LCD and keyboard, and ePWM pulse output provides 4-20mA current output.

GPRS模块能够实现向云服务器中远程传输测量数据，GPS模块能够实现对科氏质量流量计的定位，能够对流量数据，以及管道损坏进行实时监控。The GPRS module can realize the remote transmission of measurement data to the cloud server, the GPS module can realize the positioning of the Coriolis mass flowmeter, and can monitor the flow data and pipeline damage in real time.

云服务器接收到的数据分类存储到数据库中，通过网络开发技术，将其用网页的形式展现出来，能够实现任何网络移动终端对科式质量流量计监测数据的实时读取，数据对比分析，异常状况报警等。The data received by the cloud server is classified and stored in the database, and displayed in the form of a webpage through network development technology, which can realize real-time reading of the monitoring data of the Ko-type mass flowmeter by any network mobile terminal, data comparison and analysis, and abnormal Status alarm, etc.

一种科里奥利质量流量计数字信号处理方法，包括以下步骤：A digital signal processing method for a Coriolis mass flowmeter, comprising the following steps:

步骤一：数字驱动，利用反馈型数字驱动模块使科氏质量流量计起振并维持稳定工作状态；Step 1: Digital drive, use the feedback digital drive module to make the Coriolis mass flowmeter vibrate and maintain a stable working state;

步骤二：科氏质量流量计起振并维持稳定工作状态后信号预处理，采用带通IIR数字滤波器，对AD采样电路采样得到的信号进行数字滤波，保证算法输入数据的精度；Step 2: After the Coriolis mass flowmeter starts to vibrate and maintain a stable working state, the signal is preprocessed, and the band-pass IIR digital filter is used to digitally filter the signal sampled by the AD sampling circuit to ensure the accuracy of the input data of the algorithm;

步骤三：自适应频率跟踪，利用IIR陷波器对两路AD采样得到的带有相位差的传感器振动信号中提取增强信号，再利用牛顿LMS(最小均方误差算法)算法自适应跟踪信号频率；IIR陷波器使陷波频率收敛到流量管振动的基频，让基频周围一个窄频带以外的所有噪声通过，由IIR陷波器参数结合牛顿LMS算法求出基频；Step 3: Adaptive frequency tracking, use the IIR notch filter to extract the enhanced signal from the sensor vibration signal with phase difference obtained by two-way AD sampling, and then use the Newton LMS (minimum mean square error algorithm) algorithm to adaptively track the signal frequency ; The IIR notch filter makes the notch frequency converge to the fundamental frequency of the flow tube vibration, allowing all noise outside a narrow frequency band around the fundamental frequency to pass through, and the fundamental frequency is obtained by combining the parameters of the IIR notch filter with the Newton LMS algorithm;

步骤四：通过离散时间傅里叶变换算法获得两路振动信号相位差；Step 4: Obtain the phase difference of the two vibration signals through the discrete-time Fourier transform algorithm;

步骤五：将相位差进行平滑处理后求得质量流量；Step 5: Obtain the mass flow rate after smoothing the phase difference;

步骤六：温度补偿，检测科氏流量计的敏感管材料的弹性模量温度，根据检测的温度得到补偿系数，计算出补偿后的瞬时流量，从而对温度效应进行数字补偿。Step 6: Temperature compensation, detecting the elastic modulus temperature of the sensitive tube material of the Coriolis flowmeter, obtaining the compensation coefficient according to the detected temperature, and calculating the compensated instantaneous flow rate, thereby digitally compensating for the temperature effect.

所述步骤一中数字驱动的具体过程：在驱动起始阶段，由DSP模块产生初始激振信号激振科氏质量流量计的流量管，当磁电传感器检测幅值达到给定值后，结合牛顿LMS算法估计的频率和DTFT算法估计的相位，合成正弦驱动信号，再利用非线性幅值增益控制方法，得到该时刻驱动信号幅值增益，将合成的正弦信号和非线性幅值增益相乘得到驱动信号，形成反馈回路，使流量管维持在期望幅值附近振动。The specific process of digital driving in the first step: in the initial stage of driving, the initial excitation signal is generated by the DSP module to excite the flow tube of the Coriolis mass flowmeter. When the detection amplitude of the magnetoelectric sensor reaches a given value, combined with The frequency estimated by the Newton LMS algorithm and the phase estimated by the DTFT algorithm are synthesized into a sinusoidal drive signal, and then the nonlinear amplitude gain control method is used to obtain the amplitude gain of the drive signal at this moment, and the synthesized sinusoidal signal is multiplied by the nonlinear amplitude gain The driving signal is obtained to form a feedback loop to keep the flow tube vibrating near the desired amplitude.

所述频率估计利用牛顿LMS算法，相位估计利用DTFT算法；The frequency estimation utilizes the Newton LMS algorithm, and the phase estimation utilizes the DTFT algorithm;

所述步骤三中基频的求取过程为：The process of obtaining the fundamental frequency in the step 3 is:

陷波器传递函数如下：The notch filter transfer function is as follows:

$H h (({z z}^{- - 11})) = = \frac{11 + + {wz w}^{- - 11} + + {z z}^{- - 22}}{11 + + {ρwz ρwz}^{- - 11} + + {ρ ρ}^{22} {z z}^{- - 22}} . .$

其中，H(z^-1)为陷波器传递函数，w陷波因子，ρ陷阱带宽，z^-1为延迟因子。本申请中同一个符号表示的意义相同。Among them, H(z ^-1 ) is the transfer function of the notch filter, w notch factor, ρ trap bandwidth, and z ^-1 is the delay factor. In this application, the same symbols have the same meaning.

假设输入信号为随机游动模型时变信号，信号函数表示为

其中，A(n)为信号幅值，ω(n)为信号频率，为信号相位，e(n)为随机噪声信号，n为离散时间点；Assuming that the input signal is a time-varying signal of a random walk model, the signal function is expressed as

Among them, A(n) is the signal amplitude, ω(n) is the signal frequency, is the signal phase, e(n) is the random noise signal, and n is the discrete time point;

当陷波器传递函数中参数w＝-2cosω时，陷波器输出估计为：When the parameter w=-2cosω in the transfer function of the notch filter, the output of the notch filter is estimated to be:

$\overset{^^}{e e} ((n no)) = = \frac{11 + + {wz w}^{- - 11} + + {z z}^{- - 22}}{11 + + {ρwz ρwz}^{- - 11} + + {ρ ρ}^{22} {z z}^{- - 22}} x x ((n no))$

其中，

为e(n)的估计，ω信号角频率；in,

is the estimate of e(n), ω signal angular frequency;

当ρ→1时，

利用牛顿LMS算法对w进行估计；When ρ→1,

Use the Newton LMS algorithm to estimate w;

利用牛顿LMS算法对w进行估计具体过程为：陷波器输出误差为

定义代价函数The specific process of estimating w using the Newton LMS algorithm is as follows: the output error of the notch filter is

Define the cost function

$F f ((w w)) = = \frac{11}{N N} {Σ Σ}_{n no = = 11}^{N N} \frac{11}{22} {ϵ ϵ}^{22} ((n no,, w w))$

其中，N表示采样点个数；ε(n，w)陷波器输出误差，Among them, N represents the number of sampling points; ε(n, w) notch filter output error,

其中，w的估计

可表示为：

由于ρ趋向于1，根据牛顿LMS算法公式，

可由下式递推得到：where the estimate of w

Can be expressed as:

Since ρ tends to 1, according to the Newton LMS algorithm formula,

It can be deduced by the following formula:

$\overset{^^}{w w} ((n no + + 11)) = = \overset{^^}{w w} ((n no)) - - μ μ ((n no)) &dtri; &dtri; ((n no)) \overset{^^}{e e} ((n no))$

其中， $μ (n) = (1 - λ (n)) R^{- 1} (n) = (1 - λ (n)) {(\frac{{&PartialD;}^{2} F (w)}{{&PartialD;}^{2} w^{2}})}^{- 1},$ in, $μ (no) = (1 - λ (no)) R^{- 1} (no) = (1 - λ (no)) {(\frac{{&PartialD;}^{2} f (w)}{{&PartialD;}^{2} w^{2}})}^{- 1},$

λ(n)为遗忘因子，R^-1(n)自相关函数，λ(n)＝λ₀λ(n-1)+(1-λ₀)λ_∞，λ₀，λ_∞分别为遗忘因子初值和终值，μ(n)为自相关因子，▽(n)为离散梯度算子；牛顿LMS就是基于最速下降法，所以梯度算子相当于下降速率。λ(n) is the forgetting factor, R ^-1 (n) autocorrelation function, λ(n)=λ ₀ λ(n-1)+(1-λ ₀ )λ _∞ , λ ₀ , λ _∞ are the forgetting factors Initial value and final value, μ(n) is the autocorrelation factor, ▽(n) is the discrete gradient operator; Newton LMS is based on the steepest descent method, so the gradient operator is equivalent to the rate of descent.

μ(n)可由递推计算得到μ(n) can be obtained by recursive calculation

$μ μ ((n no)) = = \frac{μ μ ((n no - - 11))}{λ λ ((n no)) + + {&dtri; &dtri;}^{22} ((n no)) μ μ ((n no - - 11))},,$

其中，in,

$&dtri; &dtri; ((n no)) = = \frac{&PartialD; &PartialD; \overset{^^}{e e} ((n no))}{&PartialD; &PartialD; w w} = = \frac{y the y ((n no - - 11)) - - ρ ρ \overset{^^}{e e} ((n no - - 11))}{11 + + {ρwz ρwz}^{- - 11} + + {ρ ρ}^{22} {z z}^{- - 22}} . .$

此处，每个陷阱的带宽由ρ的取值确定，在输入信号的先验知识未知的情况下，如果ρ非常趋近于1，即极点靠近零点，将ρ改写为ρ(n)，如下定义：Here, the bandwidth of each trap is determined by the value of ρ. In the case where the prior knowledge of the input signal is unknown, if ρ is very close to 1, that is, the pole is close to zero, rewrite ρ as ρ(n), as follows definition:

ρ(n)＝ρ₀ρ(n-1)+(1-ρ₀)ρ_∞，ρ(n)=ρ ₀ ρ(n-1)+(1-ρ ₀ )ρ _∞ ,

通过仿真选取相应参数ρ₀，ρ_∞的值，这两个为常值，根据信号仿真得到最佳值，初始值和终值。科氏质量流量计信号频率估计

由公式

求得。The values of the corresponding parameters ρ ₀ and ρ _∞ are selected through simulation, these two are constant values, and the optimal value, initial value and final value are obtained according to the signal simulation. Coriolis mass flowmeter signal frequency estimation

by the formula

Get it.

所述步骤四中两路振动信号相位差的获取过程为：The acquisition process of the phase difference of the two vibration signals in the step 4 is:

设观测信号为两路同频率的实正弦信号：Suppose the observed signal is two real sinusoidal signals with the same frequency:

s₁(t)＝A₁cos(2πf₀t+θ₁)，s ₁ (t)=A ₁ cos(2πf ₀ t+θ ₁ ),

s₂(t)＝A₂cos(2πf₀t+θ₂).s ₂ (t)＝A ₂ cos(2πf ₀ t+θ ₂ ).

其中，A₁，A₂为不同的信号幅值，f₀为信号频率，θ₁，θ₂为两路信号初始相位，t为采样时间，s₁(t)，s₂(t)为两路连续信号函数；Among them, A ₁ and A ₂ are different signal amplitudes, f ₀ is the signal frequency, θ ₁ and θ ₂ are the initial phases of the two signals, t is the sampling time, s ₁ (t), s ₂ (t) are two Road continuous signal function;

以采样频率f_s(f_s≥2f₀)同时对两路信号采样，获得采样序列：Simultaneously sample the two signals at the sampling frequency f _s (f _s ≥2f ₀ ) to obtain the sampling sequence:

s₁(n)＝A₁cos(ωn+θ₁)，s ₁ (n)=A ₁ cos(ωn+θ ₁ ),

s₂(n)＝A₂cos(ωn+θ₂)，n＝0，1，...，N-1.s ₂ (n)=A ₂ cos(ωn+θ ₂ ), n=0, 1, . . . , N-1.

其中，s₁(n)s₂(n)为采样后离散信号函数；Among them, s ₁ (n) s ₂ (n) is the discrete signal function after sampling;

设ω的估计值为

则s₁(n)在

处离散时间傅里叶变换为：Let the estimated value of ω be

Then s ₁ (n) in

The discrete-time Fourier transform is:

${S S}_{11,, N N} ((\overset{^^}{ω ω})) = = {Σ Σ}_{n no = = 00}^{N N - - 11} {A A}_{11} cos cos ((ωn ωn + + {θ θ}_{11})) . . {e e}^{- - j j \overset{^^}{ω ω} n no} = = {Σ Σ}_{n no = = 00}^{N N - - 11} \frac{{A A}_{11}}{22} [[{e e}^{j j ((ωn ωn + + {θ θ}_{11}))} + + {e e}^{- - j j ((ωn ωn + + {θ θ}_{11}))}]] . . {e e}^{- - j j \overset{^^}{ω ω} n no},,$

${S S}_{11,, N N} ((ω ω)) = = {Σ Σ}_{n no = = 00}^{N N - - 11} {s the s}_{11} ((n no)) * * {e e}^{- - jωn jωn},, {S S}_{11,, N N + + 11} ((ω ω)) = = {Σ Σ}_{n no = = 00}^{N N} {s the s}_{11} ((n no)) * * {e e}^{- - jωn jωn} = = {S S}_{11,, N N} ((ω ω)) + + {s the s}_{11} ((N N)) * * {e e}^{- - jωn jωn} . .$

其中，S_1，N(ω)为一路信号第N采样点经离散时间傅里叶变换后的信号，s₁(n)为一路信号采样后离散信号函数，S_1，N+1(ω)为一路信号第N+1采样点经离散时间傅里叶变换后的信号，为在处离散时间傅里叶变换一路信号函数；Among them, S _{1, N} (ω) is the signal after discrete-time Fourier transform at the Nth sampling point of one signal, s ₁ (n) is the discrete signal function after sampling one signal, S _{1, N+1} (ω) is the signal after the discrete-time Fourier transform of the N+1th sampling point of a signal, for in Discrete-time Fourier transform signal function of one channel;

假设推导即suppose Derivation is

其中，c₁，c₂，c₃，c₄为推导过程中间参数；Among them, c ₁ , c ₂ , c ₃ , and c ₄ are intermediate parameters in the derivation process;

c₁＝sinα₁sinα₂cos(α₁-α₃)+sinα₃sinα₄cos(α₄-α₂)，c ₁ = sinα ₁ sinα ₂ cos(α ₁ -α ₃ )+sinα ₃ sinα ₄ cos(α ₄ -α ₂ ),

c₂＝sinα₁sinα₂sin(α₁-α₃)-sinα₃sinα₄sin(α₄-α₂)，c ₂ = sinα ₁ sinα ₂ sin(α ₁ -α ₃ )-sinα ₃ sinα ₄ sin(α ₄ -α ₂ ),

c₃＝sinα₁sinα₂sin(α₁-α₃)+sinα₃sinα₄sin(α₄-α₂)，c ₃ = sinα ₁ sinα ₂ sin(α ₁ -α ₃ )+sinα ₃ sinα ₄ sin(α ₄ -α ₂ ),

c₄＝sinα₁sinα₂cos(α₁-α₃)-sinα₃sinα₄cos(α₄-α₂)，c ₄ = sinα ₁ sinα ₂ cos(α ₁ -α ₃ )-sinα ₃ sinα ₄ cos(α ₄ -α ₂ ),

并且，α₁，α₂，α₃，α₄为推导过程中间参数；And, α ₁ , α ₂ , α ₃ , α ₄ are intermediate parameters in the derivation process;

${α α}_{11} = = N N ((ω ω - - \overset{^^}{ω ω})) / / 22,, {α α}_{22} = = ((ω ω + + \overset{^^}{ω ω})) / / 22,, {α α}_{33} = = ((ω ω - - \overset{^^}{ω ω})) / / 22,, {α α}_{44} = = N N ((ω ω + + \overset{^^}{ω ω})) / / 22,,$

为S1_，N(ω)的相位，同理，对于s₂(n)，存在着

is the phase of S1 _{, N} (ω), similarly, for s ₂ (n), there exists

$\hat{ω} \approx {ω, \sin α}_{1} / {\sin α}_{3} \approx N, α = N \hat{ω},$ 则相位差Δθ可近似表达为 $\hat{ω} \approx {ω, \sin α}_{1} / {\sin α}_{3} \approx N, α = N \hat{ω},$ Then the phase difference Δθ can be approximately expressed as

其中，m₁，m₂，m₃，m₄为推导过程中间参数，φ₂为S_2，N(ω)的相位，α为推导过程参数。Among them, m ₁ , m ₂ , m ₃ , m ₄ are the intermediate parameters of the derivation process, φ ₂ is the phase of S _{2, N} (ω), and α is the derivation process parameter.

$\begin{matrix} {m m}_{11} = = N N {((sin sin \overset{^^}{ω ω}))}^{22} - - {((sin sin α α))}^{22} / / N N \\ {m m}_{22} = = N N {((sin sin \overset{^^}{ω ω}))}^{22} + + {((sin sin α α))}^{22} / / N N - - 22 sin sin \overset{^^}{ω ω} sin sin α α cos cos ((α α - - \overset{^^}{ω ω})) \\ {m m}_{33} = = 22 sin sin \overset{^^}{ω ω} sin sin α α cos cos ((α α - - \overset{^^}{ω ω})) \\ {m m}_{44} = = N N {((sin sin \overset{^^}{ω ω}))}^{22} + + {((sin sin α α))}^{22} / / N N + + 22 sin sin \overset{^^}{ω ω} sin sin α α cos cos ((α α - - \overset{^^}{ω ω})) \end{matrix} . .$

所述步骤五中质量流量计算根据时间差

为采样频率，K为流量计常数，从而算得质量流量。The mass flow calculation in the step five is based on the time difference

Is the sampling frequency, K is the constant of the flowmeter, so as to calculate the mass flow rate.

所述步骤五中相位差平滑处理的方式采用加权平均。The method of phase difference smoothing in the step 5 adopts weighted average.

带反馈控制的数字驱动模块使振动管快速起振并维持稳定振动，通过预处理能有效减少外界干扰对精确度的影响，该数字信号处理算法大大减少软件计算量，牛顿LMS自适应算法能够实时精确跟踪信号频率变化，DTFT算法计算相位差时收敛快，精度高，从而使质量流量的测量精度得以提高，实时性增强。本实用新型阐述的科氏质量流量计数字信号处理算法，采用了带反馈的非线性增益控制算法进行数字驱动，快速稳定，牛顿LMS算法自适应跟踪频率及时准确，精度达到0.01%，带有温度补偿的DTFT算法计算相位差精度达到0.02%工业级别。因此可见，该数字信号处理算法是一种高精度、实时性强的科氏质量流量计信号处理方法。The digital drive module with feedback control enables the vibrating tube to vibrate quickly and maintain stable vibration. The impact of external interference on the accuracy can be effectively reduced through pre-processing. The digital signal processing algorithm greatly reduces the amount of software calculation. The Newton LMS adaptive algorithm can real-time Accurately track signal frequency changes, DTFT algorithm has fast convergence and high precision when calculating phase difference, so that the measurement accuracy of mass flow can be improved and the real-time performance can be enhanced. The Coriolis mass flowmeter digital signal processing algorithm described in the utility model adopts a nonlinear gain control algorithm with feedback for digital driving, which is fast and stable. The adaptive tracking frequency of the Newton LMS algorithm is timely and accurate, and the accuracy reaches 0.01%. The compensated DTFT algorithm calculates the phase difference with an accuracy of 0.02% at the industrial level. Therefore, it can be seen that the digital signal processing algorithm is a high-precision, real-time strong Coriolis mass flowmeter signal processing method.

本实用新型的有益效果：The beneficial effects of the utility model:

利用GPS模块采集科氏质量流量计位置信息，采用SIM300模块的GPRS远程网络通信，实现科氏质量流量计检测获得的参数从DSP到云服务器，以及从云服务器到移动网络终端的数据传输。通过LCD和键盘实现人机交互功能，ePWM脉冲输出提供4～20mA电流输出。Use the GPS module to collect the location information of the Coriolis mass flowmeter, and use the GPRS remote network communication of the SIM300 module to realize the data transmission of the parameters obtained by the detection of the Coriolis mass flowmeter from the DSP to the cloud server, and from the cloud server to the mobile network terminal. Human-computer interaction is realized through LCD and keyboard, and ePWM pulse output provides 4-20mA current output.

附图说明Description of drawings

图1科氏质量流量计结构图；Figure 1 Coriolis mass flowmeter structure diagram;

图2一次仪表与二次仪表连接图；Figure 2 The connection diagram of the primary instrument and the secondary instrument;

图3算法实施方案硬件结构图；Fig. 3 algorithm implementation scheme hardware structural diagram;

图中，1平行的U形测量管，2磁电传感器B，3磁电传感器A，4驱动器，5变送器，6，10芯输出电缆，7插头及连接电缆，8传感器。In the figure, 1 parallel U-shaped measuring tube, 2 magnetoelectric sensor B, 3 magnetoelectric sensor A, 4 driver, 5 transmitter, 6, 10-core output cable, 7 plug and connecting cable, 8 sensor.

具体实施方式：Detailed ways:

下面结合附图对本实用新型进行详细说明：Below in conjunction with accompanying drawing, the utility model is described in detail:

如图1所示，科里奥利质量流量计（下简称为科氏质量流量计），能够直接测量质量流量，测量精度高，应用前景广阔。科氏流量计按其结构分为直管型和弯管型。该实用新型专利将以双U型管科氏质量流量计为例进行设计，As shown in Figure 1, a Coriolis mass flowmeter (hereinafter referred to as a Coriolis mass flowmeter) can directly measure mass flow with high measurement accuracy and broad application prospects. Coriolis flowmeters are divided into straight tube type and curved tube type according to their structure. This utility model patent will be designed with a double U-tube Coriolis mass flowmeter as an example.

科氏流量计工作原理描述如下：当有流体流经流量计测量管时，在测量管振动频率一定的情况下，流入和流出测量管的两路正弦波信号会存在相位差，而且该相位差正比于流过测量管的流体质量流量。因此，科氏质量流量计关键在于两路传感器信号的振动频率和相位差的获取。The working principle of the Coriolis flowmeter is described as follows: When a fluid flows through the measuring tube of the flowmeter, when the vibration frequency of the measuring tube is constant, there will be a phase difference between the two sine wave signals flowing into and out of the measuring tube, and the phase difference Proportional to the mass flow rate of fluid flowing through the measuring tube. Therefore, the key to the Coriolis mass flowmeter lies in the acquisition of the vibration frequency and phase difference of the two sensor signals.

如图2所示，科氏质量流量计包括一次仪表和二次仪表，其中一次仪表包括平行的U形测量管1、传感器8、驱动器、温度传感器和连接电缆插头，传感器包括磁电传感器B2和磁电传感器A3，其中温度传感器位于U型管与连接法兰交叉处。一次仪表与二次仪表之间由插头及连接电缆7连接。二次仪表主要由系统反馈数字驱动模块、信号采集模块、信号处理模块和SIM300模块构成，即DSP变送器。二次仪表主要由系统反馈数字驱动模块、信号采集模块、信号处理模块构成，即变送器，作用是为驱动器提供驱动信号，测量传感器信号的频率以及相位差。传统处理方法是基于模拟电路的信号处理方式，对传感器输出信号进行放大、滤波、整形、鉴相和计数，测量相位差大小。As shown in Figure 2, the Coriolis mass flowmeter includes a primary meter and a secondary meter, wherein the primary meter includes a parallel U-shaped measuring tube 1, a sensor 8, a driver, a temperature sensor and a connecting cable plug, and the sensor includes a magnetic sensor B2 and Magnetic sensor A3, wherein the temperature sensor is located at the intersection of the U-shaped pipe and the connecting flange. The primary meter and the secondary meter are connected by a plug and a connecting cable 7 . The secondary instrument is mainly composed of a system feedback digital drive module, a signal acquisition module, a signal processing module and a SIM300 module, that is, a DSP transmitter. The secondary instrument is mainly composed of a system feedback digital drive module, a signal acquisition module, and a signal processing module, that is, a transmitter, which is used to provide a drive signal for the driver and measure the frequency and phase difference of the sensor signal. The traditional processing method is based on the signal processing method of the analog circuit, which amplifies, filters, shapes, phases and counts the output signal of the sensor, and measures the phase difference.

如图3所示，一种科里奥利质量流量计数字信号处理装置，包括一个科氏质量流量计，科氏质量流量计自带两个磁电传感器、驱动器和恒流源，两个磁电传感器将采集到的信号传送给与磁电传感器相对应的差分放大电路，差分放大电路将处理后的信号通过与差分放大电路相对应的AD采样电路传送给DSP；As shown in Figure 3, a Coriolis mass flowmeter digital signal processing device includes a Coriolis mass flowmeter. The Coriolis mass flowmeter has two magnetoelectric sensors, drivers and constant current sources. The electrical sensor transmits the collected signal to the differential amplifier circuit corresponding to the magnetoelectric sensor, and the differential amplifier circuit transmits the processed signal to the DSP through the AD sampling circuit corresponding to the differential amplifier circuit;

驱动器通过反馈型数字驱动模块与DSP通讯连接；DSP还与SRAM、EEPROM、ePWM的输出、LCD、键盘及GPS模块相连；DSP还通过GPRS模块与云服务器相连，云服务器与移动终端相连。科氏质量流量计为双U型管科氏质量流量计。GPRS模块包括SIM300模块。The driver communicates with the DSP through the feedback digital drive module; the DSP is also connected with the output of SRAM, EEPROM, ePWM, LCD, keyboard and GPS module; the DSP is also connected with the cloud server through the GPRS module, and the cloud server is connected with the mobile terminal. The Coriolis mass flowmeter is a double U-tube Coriolis mass flowmeter. GPRS module includes SIM300 module.

步骤三：自适应频率跟踪，利用IIR陷波器对两路AD采样得到的带有相位差的传感器振动信号中提取增强信号，再利用牛顿LMS算法自适应跟踪信号频率；IIR陷波器使陷波频率收敛到流量管振动的基频，让基频周围一个窄频带以外的所有噪声通过，由IIR陷波器参数结合牛顿LMS算法求出基频；Step 3: Adaptive frequency tracking, use the IIR notch filter to extract the enhanced signal from the sensor vibration signal with phase difference obtained by two-way AD sampling, and then use the Newton LMS algorithm to adaptively track the signal frequency; the IIR notch filter makes the trap The wave frequency converges to the fundamental frequency of the flow tube vibration, allowing all noise outside a narrow frequency band around the fundamental frequency to pass through, and the fundamental frequency is obtained by combining the IIR notch filter parameters with the Newton LMS algorithm;

所述步骤一中数字驱动的具体过程：在驱动起始阶段，由DSP模块产生初始激振信号激振流量管，当磁电传感器检测幅值达到给定值后，结合频率估计和相位估计，合成正弦驱动信号，再利用非线性幅值增益控制方法，得到该时刻驱动信号幅值增益，将合成的正弦信号和非线性幅值增益相乘得到驱动信号，形成反馈回路，使流量管维持在期望幅值附近振动。利用反馈型数字驱动模块测量得到相位，幅值信号经过非线性控制算法反馈控制驱动信号。The specific process of digital driving in the first step: in the initial stage of driving, the initial excitation signal is generated by the DSP module to excite the flow tube, and when the detection amplitude of the magnetoelectric sensor reaches a given value, combined with frequency estimation and phase estimation, Synthesize the sinusoidal driving signal, and then use the nonlinear amplitude gain control method to obtain the amplitude gain of the driving signal at this moment, multiply the synthesized sinusoidal signal and the nonlinear amplitude gain to obtain the driving signal, and form a feedback loop to maintain the flow tube at Vibrates around the desired amplitude. The phase is measured by the feedback digital drive module, and the amplitude signal is fed back to control the drive signal through a nonlinear control algorithm.

本装置采用的DSP是TI公司的TMS320F28335DSP芯片，由于GPRS模块的供电电压为3.4～4.5（典型值为4.2），采用5V供电时，需要进行5V到4.2V的转换，本装置使用MIC29300为SIM300提供电压，其输出电流达到3A，可满足SIM300的要求。The DSP used in this device is the TMS320F28335DSP chip of TI Company. Since the power supply voltage of the GPRS module is 3.4~4.5 (typical value is 4.2), when using 5V power supply, it needs to convert from 5V to 4.2V. This device uses MIC29300 to provide SIM300 Voltage, its output current reaches 3A, which can meet the requirements of SIM300.

GPS模块采用GS-15C GSP接收机采集位置信息，精度达到5～10米。GPS模块集成度高，通过串口与DSP通信。The GPS module adopts GS-15C GSP receiver to collect location information, and the accuracy reaches 5-10 meters. The GPS module is highly integrated and communicates with the DSP through the serial port.

SIM300模块内部集成了GSM控制器，两个串口，一个SIM卡接口，两个模拟音频接口等。所有这些硬件接口除天线接口外，都是通过60针的板对板连接器进行连接，对传输中用不到的接口进行引脚悬空即可。The SIM300 module integrates a GSM controller, two serial ports, one SIM card interface, two analog audio interfaces, etc. All these hardware interfaces except the antenna interface are connected through 60-pin board-to-board connectors, and the pins of the interfaces that are not used in transmission can be left floating.

GPRS模块工作过程如下：SIM300模块上电后，观察NetworkLED引脚上的网络指示灯，等到网络指示灯的闪烁频率变为64ms ON/3000ms OFF，此时表示模块已经连接到GPRS网络上，通过DSPF28335引脚对PWKEY引脚输出一个大于1500ms的低脉冲来开启SIM300模块。SIM300内部集成了TCP/IP协议，DSPF28335通过串口向SIM300发送AT指令，就可控制SIM300实现数据传输功能。The working process of the GPRS module is as follows: After the SIM300 module is powered on, observe the network indicator on the NetworkLED pin, and wait until the flashing frequency of the network indicator becomes 64ms ON/3000ms OFF, which means that the module has been connected to the GPRS network. The pin outputs a low pulse greater than 1500ms to the PWKEY pin to turn on the SIM300 module. SIM300 integrates TCP/IP protocol inside, and DSPF28335 sends AT commands to SIM300 through the serial port to control SIM300 to realize data transmission function.

云服务器设置为固定IP的数据库系统，SIM300通过GPRS访问云服务器的某一设定端口（如80端口），从而进行GPRS模块与云服务器之间的通信。云服务器将收到的监测数据按数据类别存储到数据库中，数据类别主要有质量流量、温度、GPS定位位置以及采集时间等。利用网络开发技术，将数据库内容以网页的形式展现，实现实时定位，实时检测，实时查询。这样，任何一个移动网络终端都可以通过网址进行访问云空间，在线实时观测科氏流量计工作状态。The cloud server is set as a database system with a fixed IP, and SIM300 accesses a certain port (such as port 80) of the cloud server through GPRS to communicate between the GPRS module and the cloud server. The cloud server stores the received monitoring data in the database according to the data category. The data categories mainly include mass flow rate, temperature, GPS location and collection time. Using network development technology, the database content is displayed in the form of a web page to realize real-time positioning, real-time detection, and real-time query. In this way, any mobile network terminal can access the cloud space through the website, and observe the working status of the Coriolis flowmeter online in real time.

利用CPUTIMER0定时器中断，得到1s累积流量，保存在外扩的EEPROM中。Utilize CPUTIMER0 timer to interrupt, get 1s accumulative flow, save in the EEPROM that expands outside.

PWM输出功能由DSP的ePWM中一路PWM比较功能，获得带有流量信息的脉冲信号。The PWM output function obtains the pulse signal with the flow information by one PWM comparison function in the ePWM of the DSP.

人机接口由LCD和键盘组成，利用DSP多功能复用GPIO口，来实现特定功能。LCD显示测量结果有瞬时流量、累积流量、温度等。键盘主要用来设定仪表系数。The man-machine interface is composed of LCD and keyboard, and uses DSP multi-function multiplexing GPIO port to realize specific functions. LCD display measurement results include instantaneous flow, cumulative flow, temperature, etc. The keyboard is mainly used to set meter coefficients.

由DSP提供初始正弦激振信号，信号幅度由小变大，当振动幅度达到给定值时，利用牛顿LMS陷波算法以及DTFT算法估计信号频率和相位，从而合成正弦驱动信号。然后，根据振动幅度的变化，通过与给定幅度的差值确定幅值增益。最后，将合成正弦波与增益的乘积作为反馈驱动信号，保持流量管振动幅度维持在稳定工作状态。The initial sinusoidal excitation signal is provided by DSP, and the signal amplitude changes from small to large. When the vibration amplitude reaches a given value, the Newton LMS notch algorithm and DTFT algorithm are used to estimate the signal frequency and phase, thereby synthesizing the sinusoidal drive signal. Then, according to the variation of the vibration amplitude, the amplitude gain is determined by the difference with the given amplitude. Finally, the product of the synthesized sine wave and the gain is used as the feedback driving signal to keep the vibration amplitude of the flow tube in a stable working state.

流量管稳定振动工作后，利用两路AD采集两路磁电式传感器信号，经过多通道缓冲串口Mcbsp，利用DMA传输到内部存储器临时数组，当临时数组放满后，产生DMA接收中断，将两个临时数组数据转移到外扩的SRAM缓冲数组中。After the flow tube stabilizes and vibrates, use two ADs to collect two channels of magnetoelectric sensor signals, pass through the multi-channel buffer serial port Mcbsp, and use DMA to transfer to the temporary array of the internal memory. When the temporary array is full, a DMA receiving interrupt is generated, and the two The temporary array data is transferred to the externally expanded SRAM buffer array.

根据DSP运算速度，每次取500点数据，当信号幅值大于设定值，开始调用算法模块。对数据进行预处理，根据流量管参数设计切比雪夫带通滤波器，滤波后数据也存放在外扩SRAM数组中。According to the DSP operation speed, 500 points of data are taken each time, and when the signal amplitude is greater than the set value, the algorithm module is called. The data is preprocessed, and the Chebyshev bandpass filter is designed according to the parameters of the flow tube, and the filtered data is also stored in the externally expanded SRAM array.

临时数组中滤波后的数据进入牛顿LMS陷波算法模块，一方面，由牛顿LMS算法自适应估计两路信号基频。为了保证频率精度，对基频进行平均处理，并设置波动范围，当波动幅度大于设定值时，则不更新频率，反之，则更新频率，将频率值保存在外扩数组中。另一方面，两路信号经过陷波器后为滤除噪声的增强信号，保存此增强信号，为相位差计算提供精确输入数据。The filtered data in the temporary array enters the Newton LMS notch algorithm module. On the one hand, the Newton LMS algorithm adaptively estimates the fundamental frequency of the two signals. In order to ensure the frequency accuracy, the fundamental frequency is averaged and the fluctuation range is set. When the fluctuation range is greater than the set value, the frequency will not be updated. Otherwise, the frequency will be updated and the frequency value will be stored in the extended array. On the other hand, after the two signals pass through the notch filter, they become the enhanced signal for filtering noise, and the enhanced signal is saved to provide accurate input data for phase difference calculation.

调用DTFT算法对增强信号求得相位差时，将增强信号进行DTFT后，针对算法计算特点，分别将信号实部和虚部保存在两个外扩数组中，从而得到两路信号相位差，然后对相位差进行平滑处理，结合频率值得到时间差，继而得到瞬时流量。仪表系数保存在外扩的EEPROM。When calling the DTFT algorithm to obtain the phase difference of the enhanced signal, after performing DTFT on the enhanced signal, according to the calculation characteristics of the algorithm, the real part and the imaginary part of the signal are stored in two external expansion arrays, so as to obtain the phase difference of the two signals, and then The phase difference is smoothed, and the time difference is obtained by combining the frequency value, and then the instantaneous flow rate is obtained. Meter coefficients are stored in the externally expanded EEPROM.

采集温度传感器信号，通过串行外设接口SPI，进入DSP转换为温度值，根据流量计材质获得相应温度补偿系数，对瞬时流量进行温度补偿。Collect the temperature sensor signal, enter the DSP to convert it into a temperature value through the serial peripheral interface SPI, and obtain the corresponding temperature compensation coefficient according to the material of the flowmeter, and perform temperature compensation on the instantaneous flow rate.

温度补偿：科氏流量计的敏感管材料的弹性模量是随温度变化而变化的，可以适时检测温度，根据温度得到补偿系数，计算出补偿后的瞬时流量，从而对温度效应进行数字补偿。Temperature compensation: The elastic modulus of the sensitive tube material of the Coriolis flowmeter changes with the change of temperature. The temperature can be detected in time, and the compensation coefficient can be obtained according to the temperature, and the instantaneous flow rate after compensation can be calculated, so as to digitally compensate the temperature effect.

对于单相流信号，提高计算精度、扩展量程下限是科氏质量流量计数字信号处理算法的目标。尤其是对于小流量信号，信号较弱、信噪比低，DTFT计算得到的相位差波动比较大，所以需要采用加权平均对相位差进行平滑处理。但流量突变会产生测量误差，为此设置相位差限定值，如果连续10个相位差计算结果都超过限定值，则把这10个相位差值之和的平均值作为当前相位差，加快流量变化反应速度。For single-phase flow signals, improving the calculation accuracy and extending the lower limit of the range is the goal of the digital signal processing algorithm of the Coriolis mass flowmeter. Especially for small flow signals, the signal is weak and the signal-to-noise ratio is low, and the phase difference calculated by DTFT fluctuates greatly, so it is necessary to use weighted average to smooth the phase difference. However, sudden changes in the flow rate will cause measurement errors. For this purpose, set the phase difference limit value. If the calculated results of 10 consecutive phase differences exceed the limit value, the average value of the sum of the 10 phase difference values will be used as the current phase difference to speed up the flow change. reaction speed.

通过DTFT递推算法分别计算出两路信号在

处的DTFT，其相位相减之后即可得到两路信号的相位差Δθ，此即为DTFT递推算法测量相位差的基本原理。Through the DTFT recursive algorithm, the two signals in the

The phase difference Δθ of the two signals can be obtained after the phase subtraction of the DTFT at , which is the basic principle of the DTFT recursive algorithm to measure the phase difference.

牛顿LMS自适应算法在收敛后根据信号特征的变化持续调整陷波器参数，跟踪振动频率的变化。采用零极点约束的IIR陷波器，将零点固定在单位圆上，并且位于陷波频率处，极点设在单位圆内，与零点同一角度。After the Newton LMS adaptive algorithm converges, it continuously adjusts the parameters of the notch filter according to the change of the signal characteristics, and tracks the change of the vibration frequency. The IIR notch filter with pole-zero constraints fixes the zero point on the unit circle and is located at the notch frequency, and the pole is set within the unit circle at the same angle as the zero point.

由于工业现场噪声多在5KHz以上，所以在高采样频率下，采用简单的带通滤波器即可达到滤波效果。Since the industrial field noise is mostly above 5KHz, the filtering effect can be achieved by using a simple band-pass filter at a high sampling frequency.

一般情形下，当信噪比不是特别低的时候，通过自适应格型陷波器收敛后求得的信号频率值与真实值很接近，即可以认为

，In general, when the signal-to-noise ratio is not particularly low, the signal frequency value obtained after the convergence of the adaptive lattice notch filter is very close to the real value, that is, it can be considered

,

对本专利进行了仿真实验，仿真中，采样参数：信号频率f=200hz，采样频率fs=2000。在频率计算不到100点时刻，牛顿LMS算法达到收敛，并且频率跟踪快速准确，精度为0.01%级别。A simulation experiment was carried out on this patent. In the simulation, sampling parameters: signal frequency f=200hz, sampling frequency fs=2000. When the frequency calculation is less than 100 points, the Newton LMS algorithm reaches convergence, and the frequency tracking is fast and accurate, with an accuracy of 0.01% level.

DTFT算法能够快速并精确计算两路传感器信号相位差。仿真中，相位差参数为：phasediff=0.01，。在相位差计算300点左右时刻，相位差已经达到收敛，精度在0.02%左右。The DTFT algorithm can quickly and accurately calculate the phase difference of two sensor signals. In the simulation, the phase difference parameter is: phasediff=0.01,. When the phase difference calculation is about 300 points, the phase difference has reached convergence, and the accuracy is about 0.02%.

实验还仿真了从0.01度到0.4度区间内5个相位，每个相位差取5次测量结果的平均值，其计算精度如表1。The experiment also simulates 5 phases in the interval from 0.01 degree to 0.4 degree, and the average value of 5 measurement results is taken for each phase difference, and its calculation accuracy is shown in Table 1.

表1DTFT算法相位差计算仿真数据Table 1 DTFT algorithm phase difference calculation simulation data

给定相位差given phase difference 相位差计算平均值Calculate the average value of the phase difference 相对误差Relative error 0.01度0.01 degree 0.01000256度0.01000256 degrees 0.025600%0.025600% 0.05度0.05 degrees 0.05000116度0.05000116 degrees 0.023000%0.023000% 0.1度0.1 degrees 0.10001652度0.10001652 degrees 0.016520%0.016520% 0.2度0.2 degrees 0.20002175度0.20002175 degrees 0.010875%0.010875% 0.4度0.4 degrees 0.40005483度0.40005483 degrees 0.013707%0.013707%

Claims

1. A Coriolis mass flowmeter cloud transmission digital signal processing device is characterized by comprising a Coriolis mass flowmeter, wherein the Coriolis mass flowmeter is provided with two magnetoelectric sensors, a driver and a constant current source, the two magnetoelectric sensors transmit acquired signals to a differential amplification circuit corresponding to the magnetoelectric sensors, and the differential amplification circuit transmits the processed signals to a DSP through an AD sampling circuit corresponding to the differential amplification circuit;

the driver is in communication connection with the DSP;

the constant current source is connected with the PT100, the constant current source is used for providing voltage for the PT100, the PT100 measures the external temperature, and the PT100 is connected with the DSP through the AD sampling circuit corresponding to the PT 100;

the DSP is also connected with a cloud server through a GPRS module, and the cloud server is connected with the mobile terminal.

2. The coriolis mass flowmeter cloud transfer digital signal processing device of claim 1, wherein the DSP is further connected to the output of the SRAM, the EEPROM, the ePWM, the LCD, the keyboard, and the GPS module.

3. The coriolis mass flowmeter cloud transfer digital signal processing device of claim 1, wherein the coriolis mass flowmeter is a dual U-tube coriolis mass flowmeter.

4. The coriolis mass flowmeter cloud transfer digital signal processing device of claim 1, wherein the GPRS module comprises a SIM300 module.