CN100385207C - A low-cost intelligent vortex flowmeter signal processing system based on DSP - Google Patents

Abstract

DSP-based low-cost intelligent vortex flowmeter signal processing system, which is characterized by analog signal conditioning module, DSP chip and external expansion RAM, EEPROM, low voltage monitoring circuit, LCD display, keyboard input, analog output, pulse output and Composed of power management modules; the system of the present invention adopts the method of combining periodic map analysis and digital filtering to process the output signal of the vortex flow sensor, and accurately measure the flow of the fluid when there is pipeline vibration and fluid flow field instability on site. The system is small in size, low in cost and complete in function.

Description

A low-cost intelligent vortex flowmeter signal processing system based on DSP

Technical field:

The invention relates to a flow measuring instrument, in particular to a low-cost vortex flowmeter signal processing system with a digital signal processor (DSP) as the core and a combination of periodic graph analysis and digital filtering.

Background technique:

The vortex flowmeter is widely used in the field of fluid flow measurement because it has no moving parts in the measurement part, has small wear and pressure loss, has a large ideal range ratio, and can measure many fluid media. However, the vortex flow sensor is easily affected by fluid pulsation, pipeline or other vibration sources. Traditional signal processing methods, namely amplification, shaping, filtering and counting, cannot effectively suppress noise, so the measurement accuracy is not high; The threshold level will affect the turndown ratio, resulting in the inability to measure small flows; for different ranges and different measured fluids, the traditional signal processing system needs to replace different processing circuits and parameters. For this reason, many researchers at home and abroad have done a lot of research on the signal processing of vortex flowmeters and developed corresponding systems.

Foreign SCHLATTER, Gerald, L. studied the output signal of the vortex street sensor under the condition of strong noise, and found that the noise signal and the vortex street signal have different spectral characteristics, and different templates can be established for the noise and signal, thus proposing the use of frequency domain conversion and According to the cross-correlation power spectrum results, the strong noise in the signal is eliminated according to the established noise template and vortex signal template ("Signal processing method and apparatus for flowmeter", WO 90/04230, 19 April1990). However, the noise situation is complicated, it is not easy to establish a template, and it is difficult to apply it in practice.

Foxboro Corporation of the United States uses two sets of filters to process vortex sensor signals: one set is an instant adaptive band-pass filter, and the other set is a non-instantaneous adjustable band-pass filter ("Adaptive filter with sweep filter analyzer for a vortex flowmeter", US6212975, 10 April 2001). For the vortex flow sensor signal mixed with noise, it is divided into two channels for processing. One way uses an adaptive band-pass filter to track its vortex frequency, filters out noise, and obtains the flow rate; the other uses an adjustable band-pass filter to detect the noise in the signal, observe the fluid flow state, and call the police if abnormal conditions are found. When using an adaptive band-pass filter to process vortex signals, it is necessary to know the specific range of the current vortex frequency before determining the parameters of the filter, and the range of vortex frequency changes is wide. When the flow rate changes greatly, The configuration of the filter parameters takes a long time or is improperly configured, resulting in measurement errors.

Japan's Yokogawa (Yokogawa) company researches the method of spectral analysis first and then bandpass filtering to process the vortex signal, that is, first conduct spectral analysis on the vortex signal to determine the approximate position of the current vortex frequency; then select the corresponding bandpass filter Filtering ("VORTEX FLOWMETER", JP2001153698, 08 June 2001). It is not considered that when the gas flow rate is large, the pipeline vibration will cause strong low-frequency noise interference in the sensor signal, which will make the spectrum analysis result wrong.

Domestic Chongqing University Meng Jianbo et al introduced the use of modern spectral analysis based on the least mean square adaptive algorithm to process flow signals (Acta application in the meter"). The measurement system is composed of vortex generator, hot wire probe, pre-processing circuit and APPLE microcomputer, which verifies the feasibility of spectral analysis method for signal processing of vortex flowmeter. However, the key technical issues such as calculation accuracy, real-time and miniaturization have not been studied, and they have not been applied to actual flowmeters. In addition, our study shows that this method is less capable of suppressing harmonic interference.

Xu Kejun of Hefei University of Technology, etc. used the periodic map analysis method, the modern spectrum analysis method based on Burg, the adaptive notch filter method and the wavelet transform method to process the signal of the vortex flow sensor (Instrument Technology and Sensors, 1995, ( 5): 22-25 "Software method for signal processing of vortex flowmeter"; Journal of Instrumentation, 2000, 21(2): 222-224 "Adaptive notch method for signal estimation of vortex flowmeter"; Journal of Instrumentation , 2001, 22(3): 255-260 "Signal processing system of vortex flowmeter with spectrum analysis function based on DSP"; Meter signal processing method"), and ADSP2181 was selected as the microprocessor, and the core algorithm was periodic map analysis, and the digital signal processing system of the vortex flowmeter was developed ("Vortex flowmeter digital signal processing system", Chinese invention patent, patent No.: ZL99101209.7, application date: 1999.1.8, authorization announcement date: 2003.4.2). The system has passed the test of Anhui Provincial Flow Meter Calibration Station, and the results prove that the spectral analysis method can improve the ability to suppress noise and expand the range ratio. However, the system has many components, complex control, and large volume, which still has a certain gap with the actual popularization and application.

Zhang Tao and Duan Ruifeng of Tianjin University used the relaxation notch period map method to analyze the signal of vortex flowmeter, and expanded the lower limit of flowmeter range, and with this algorithm as the core, developed a dual-core structure digital signal processing system based on the combination of DSP and MSP430 ( "Low power consumption digital vortex flowmeter", Chinese patent, application number: 200410019008.9, application date: 2004.4.14). The digital signal processing system of the vortex flowmeter with dual-core structure utilizes the control capability of MSP430 single-chip microcomputer to realize data sampling and control of system peripherals; DSP uses its HPI port to receive the data collected by MSP430, perform signal processing, and then send the result back to MSP430 . The system structure is more complicated and the cost is higher.

Yu Haiqing and Li Yalong of Central South University used the adaptive notch filter method to estimate the vortex frequency, and selected TMS320LF2407A as the processor to design the hardware block diagram and software flow chart of the vortex flowmeter digital signal processing system (Computing Technology and Automation, 2001, 20 (3): 159-162 "Design of DSP-based Vortex Flowmeter Signal Processing System"). However, the calculation results of the algorithm, the specific implementation scheme of the system, the experimental test results and the performance indicators of the system are not given in the paper.

Invention content:

The present invention is to avoid the disadvantages of the above-mentioned prior art, and provides a low-cost intelligent vortex flowmeter signal processing system based on DSP with small volume, low cost and complete functions. , Using the combination of periodic map analysis and digital filtering to process the output signal of the vortex sensor and the signal of the temperature and pressure sensor, and accurately measure the flow of the fluid when there is pipeline vibration and fluid flow field instability on site.

The system of the present invention is composed of an analog signal conditioning module, a DSP chip, an externally expanded RAM, an EEPROM, a low voltage monitoring circuit, an LCD display, a keyboard input, an analog output, a pulse output and a power management module;

The system of the present invention is characterized in that the analog signal conditioning module includes a temperature signal conditioning circuit, a pressure signal conditioning circuit and a flow sensor signal conditioning circuit, the temperature signal conditioning circuit is composed of a bridge and a differential amplifier circuit, the The pressure signal conditioning circuit is composed of an I/V conversion circuit, and the flow sensor signal conditioning circuit is composed of a charge amplification circuit, an active band-pass filter circuit with a limiting function and a low-pass filter circuit with a fixed gain; The system accepts the temperature, pressure and flow detection signals output by the sensing unit and sends them to the DSP after being processed by the analog signal conditioning module; the DSP uses its internal integrated A/D conversion module and ring data buffer to pair The three electrical signals are subjected to timing sampling, timing calculation, and calculation while sampling; the DSP performs digital filtering on the flow sensor signal that may be mixed with noise caused by interference, and then uses a periodic map analysis method to filter Perform spectrum analysis and spectrum correction on the final data; average the results of multiple calculations to determine the maximum power spectrum value and obtain its corresponding frequency, which is the frequency of the signal; then according to the instrument coefficient and through temperature and pressure compensation, real-time Calculate the instantaneous flow and cumulative flow; the DSP uses its internal timer comparison output function to output the instantaneous flow data in pulse form from the pulse output unit; the DSP uses its internal integrated SPI module to send the instantaneous flow data to the The above-mentioned analog quantity output unit is output in the form of analog quantity of 4-20mA current; the above-mentioned DSP displays the instantaneous flow, accumulated flow, temperature and pressure on the above-mentioned LCD; Sampling frequency, number of sampling points, average number of times of multiple spectrum analysis, digital filter coefficient; the instrument coefficient input by the keyboard and the accumulated flow calculated are saved by the EEPROM; the low voltage monitoring circuit is responsible for when the system is powered off An interrupt request is raised to the DSP, and the measurement result is saved in the EEPROM before the system is completely powered off.

The system feature of the present invention is also to utilize the periodic interruption of the internal timer of the DSP chip to realize the regular sampling of the temperature, pressure and flow signals, the data obtained by the sampling is put into the ring data buffer, and the sampled data are regularly sampled. The power spectrum analysis is performed on the flow signal, and sampling is performed while calculation is performed, and the flow data of two adjacent analysis processes are partially overlapped and covered.

The system characteristic of the present invention also is that described DSP chip is TMS320LF2407A, and it is internally integrated with 32K FLASH, A/D conversion function module, 2 event managers, watchdog, serial peripheral interface SPI, serial communication interface SCI, CAN function modules.

The feature of the system of the present invention is also that according to the range range of the vortex flowmeter, the sampling frequency and the number of sampling points are set through the keyboard to control the calculation accuracy of the power spectrum analysis; the DSP chip adopts the barycenter spectrum correction method to further Improve the calculation accuracy of the frequency; according to the characteristics of the flow sensor, set the instrument coefficient in sections through the keyboard to improve the measurement accuracy of the flow; through the temperature and pressure compensation table, perform nonlinear correction on the flow calculation results to improve the superheated gas and Measurement accuracy of saturated steam flow.

The feature of the system of the present invention is also that the internal timer of the DSP chip is used to output pulses in a comparison mode, and the output pulse period is dynamically compensated by comparison interruption.

Compared with the prior art, the beneficial effects of the present invention are reflected in:

1. The system of the present invention adopts the signal processing algorithm with periodic map analysis as the core, cooperates with digital filtering, eliminates the influence caused by interference such as strong and low-frequency pipeline vibration on the measurement results, improves the ability of the instrument to suppress noise, and expands the range ratio;

2. The system of the present invention samples at regular intervals, calculates at regular intervals, calculates while sampling, and uses the ring data buffer to update the oldest data with the latest sampling data to realize real-time measurement of the flow rate; according to the working characteristics of instruments with different calibers, the algorithm parameters are adjusted through the panel Optimum configuration, at the same time, spectrum correction, set instrument coefficients, temperature and pressure compensation in sections, improve the accuracy of instrument measurement; according to the characteristics of spectrum analysis algorithm, use limiting amplification instead of program-controlled amplification for vortex street sensor signals to solve vortex street Problems where the signal strength of the sensor varies widely requiring modification of circuit parameters.

3. By selecting the DSP chip TMS320LF2407A with rich on-chip resources, the system of the present invention can reduce the peripheral devices required by the system, reduce the volume of the system, reduce the cost, and realize various functions required by the digital instrument at the same time.

Description of drawings:

Fig. 1 is a hardware block diagram of the system of the present invention.

Fig. 2 is a schematic diagram of the flow sensor signal conditioning circuit in the system of the present invention.

Fig. 3 is a schematic diagram of the electric bridge and amplifying circuit of the temperature sensor signal in the system of the present invention.

Fig. 4 is a schematic diagram of the I/V conversion circuit for processing the output signal of the pressure transmitter in the system of the present invention.

Fig. 5 is a schematic diagram of the pin connection of the DSP chip in the system of the present invention.

Fig. 6 is a schematic diagram of wiring of externally expanded RAM pins of the DSP chip in the system of the present invention.

Fig. 7 is a schematic diagram of EEPROM pin wiring in the system of the present invention.

Fig. 8 is the principle of the low voltage monitoring circuit in the system of the present invention.

Fig. 9 is a schematic diagram of the keyboard circuit in the system of the present invention.

Fig. 10 is a schematic plan view of the mid-segment liquid crystal in the system of the present invention.

Fig. 11 is the connection circuit between the DSP chip and the LCD display circuit in the system of the present invention.

Fig. 12 is a schematic diagram of the pulse output circuit in the system of the present invention.

Fig. 13 is a schematic diagram of the 4-20mA current output circuit in the system of the present invention.

Fig. 14 is a schematic diagram of the power supply circuit of the system of the present invention.

Fig. 15 is a control block diagram of the system of the present invention.

Fig. 16 is a basic flowchart of the signal processing algorithm of the system of the present invention.

Fig. 17 is a keyboard scanning flow chart of the system of the present invention.

Fig. 18 is a flow chart of the main monitoring program of the system of the present invention.

Detailed ways:

Referring to Fig. 1, the present embodiment is made up of analog signal conditioning module, TMS320LF2407A (the DSP chip of TI company) and external expansion RAM, E2PROM, low voltage monitoring circuit, LCD display, keyboard input, analog quantity output circuit, pulse output circuit and power management Module composition.

Vortex flowmeters are made using the Karman vortex principle in fluid dynamics. When the vortex generating body is placed in the fluid pipeline, two regular vortices will be generated on both sides of its downstream, and the frequency of vortex release is linearly related to the fluid flow rate, namely:

Q=k*f

In the formula, Q is the fluid volume flow rate, k is the instrument coefficient, and f is the vortex release frequency, that is, the vortex street frequency.

The sensor (such as: piezoelectric sensor) installed in the vortex generating body picks up the signal representing the frequency of the vortex street and converts it into an electrical signal for output. It is characterized by a small amplitude (millivolt level), an approximate proportional relationship with the square of the flow rate, and a large internal resistance. In an ideal situation, the sensor output signal is a sine wave, but due to the influence of instrument installation conditions, fluid pulsation, pipeline vibration and other factors, the signal is mixed with various noises. In this embodiment, an analog signal conditioning module is provided, as shown in FIG. 1 . Among them is the signal conditioning circuit of the vortex street sensor. The circuit principle is shown in Figure 2. Firstly, the charge amplifier U1A is used to perform impedance transformation on the output signal of the flow sensor and pre-amplify at the same time. The charge amplifier uses a MOS type operational amplifier with high input impedance, C1 and C2 are DC blocking capacitors, C8 is a feedback capacitor, and R2 is a feedback resistor to prevent the feedback capacitor C8 from being saturated. C9 and R3 are balancing capacitors and resistors. Resistor R1 is a matching resistor to prevent sensor output signal saturation. The output signal line of the sensor should be shielded, one end of the shielding layer is connected to the shell of the instrument, and the other end is connected to the ground of the signal processing system. The amplitude of the output signal of the charge amplifier varies greatly. In order to amplify the signal close to the full scale of the A/D converter and make full use of its conversion accuracy, the signal is properly amplified after the charge amplifier. In order to eliminate or reduce the influence of circuit noise, fluid pulsation noise and pipeline vibration on the measurement, it is also necessary to use low-pass and high-pass filter circuits. Therefore, two operational amplifiers U2B and U2C are used to form a two-stage active amplifier filter circuit. The first stage is an active bandpass filter with limiting function. The band-pass filter consists of a high-pass filter and a second-order low-pass filter. The high-pass cut-off frequency is determined by C10, R4 and R5, and the low-pass filter's cut-off frequency is determined by R4, C11, R5, R7, R8 and C12 decided. When the signal amplitude is small, the gain of the entire active filter in the passband is (R7+R8)/(R4+R5): when the input signal amplitude increases and the diode is turned on, the output voltage amplitude is Vd+Vin× (R7)/(R4+R5), where Vd is the forward voltage drop of the diode, and Vin is the output voltage of the charge amplifier. The second stage is a fixed-gain low-pass filter. An active filter circuit with an infinite-gain negative feedback structure is used here, which is mainly composed of an operational amplifier U2C to further reduce the impact of high-frequency noise. After two stages of filtering, the signal is sent to the A/D conversion module in the DSP chip through a buffer circuit composed of U2A.

Considering that temperature and pressure compensation are required when measuring steam, a conditioning circuit for receiving signals from temperature sensors and pressure transmitters is set in the analog signal input and conditioning unit. The temperature sensor adopts Pt100 platinum resistance, the signal conditioning circuit adopts a three-wire bridge structure, and a dedicated instrumentation amplifier U3 is used to amplify the voltage difference between the two bridge arms, as shown in Figure 3. An I/V conversion circuit is used to convert the 4-20mA current signal output by the pressure transmitter into a voltage signal, as shown in Figure 4. These two signals are also sent to the A/D conversion module in the DSP chip.

The DSP chip TMS320LF2407A (hereinafter referred to as 2407A) of TI Company selected by the system of the present invention is the processing core, as shown in FIG. 5 . The DSP chip has strong computing power, the highest computing speed is 40MIPS, and the instruction cycle is 25ns. It has a 16*16-bit hardware multiplier inside, which is enough to realize the FFT algorithm in real time. More importantly, 2407A integrates rich on-chip peripherals, for example, it integrates A/D module inside, which can support up to 16 channels of signal conversion, the fastest conversion cycle is 375ns, and the conversion accuracy is 10 bits. Considering that what the system extracts is the frequency of the vortex signal, the 10-bit A/D can meet the requirements. The chip also integrates 32k FLASH, 2 event managers (including 4 general-purpose timers, supporting functions such as comparison, PWM generation, capture, quadrature encoding, and 2 external interrupts), watchdog, and serial peripherals Interface SPI, serial communication interface SCI, CAN and multi-function multiplexing I/O port, etc., not only easy to realize and expand the function of the instrument, but also can reduce the external components of the system, reduce the design area of the system circuit board, and improve the System reliability. The integrated functional modules of 2407A are arranged as follows: Timer 1 cycle interrupt starts A/D conversion; Timer 3 cycle interrupt is used to regularly scan the keyboard and operate the LCD regularly; use the comparison function of Timer 2 to output pulses, and use its Comparing interrupts dynamically compensates the output pulses to reduce the impact of the working clock on the accuracy of the pulses; the watchdog module monitors the operation of the processing system, and resets the system if the program "runs away"; the SPI module sends analog data to the system The output circuit sends control signals and data, and outputs 4-20mA current signals.

Since 2407A only integrates 544 bytes of dual-port RAM and 2K single-port RAM, which cannot meet the needs of algorithm operations, 64k of RAM is expanded by using 16 external address lines and 16 data lines of 2407A, as shown in Figure 6 shown.

The system of the present invention selects an EEPROM to store measurement results and instrument coefficients, as shown in Figure 7. Use the 2407A multi-function multiplex port to control and read and write data to the EEPROM, where PBO is connected to the chip select CS of the EEPROM, PB1 is connected to CLK, PB2 is connected to the input data line DI, and PB3 is connected to the output data line DO.

The system of the present invention designs a low-voltage detection module, as shown in FIG. 8 . When the module detects that the voltage is lower than the set value, it immediately pulls down the level of the PDPINTA pin of 2407A, and sends an emergency interrupt request to 2407A. Because the interrupt priority is the highest, 2407A responds immediately, and saves the measurement results to EEPEOM before the system is powered off.

The system of the present invention designs a man-machine interface composed of a keyboard and an LCD to realize setting of instrument parameters and displaying measurement results on the spot. There are 4 keys arranged on the keyboard: set key (SET), shift key (SFT), increment key (INC) and confirmation key (ENTR), which are respectively connected with multiplex ports PA4, PA5, PA6 and PA7 of 2407A, through Timer 3 cycle interrupts for regular scanning, the keyboard circuit is shown in Figure 9. The serial interface segment LCD is selected, and the plan view is shown in Figure 10. The LCD has two lines of display: P, T, and Q symbols can be displayed on the left side of the upper line, and 6-digit data can be displayed on the right side; 8-digit data can be displayed on the lower line; in addition, there are 14 prompt barcodes on the right side. 2407A is connected to LCD by multi-function port, PB7 is connected to CS, PB6 is connected to WR, PB5 is connected to DATA, as shown in Figure 11. The LCD has two working states, one is the display state; the other is the setting state. Switch between the two working states by setting the key. In the display state, press the confirmation key to switch between three modes: one is the flow display mode, which is the default mode. The symbol on the left side of the upper line of the LCD is to display "Q", representing the flow rate, and the data on the right side is used to display the instantaneous flow rate. LCD The lower line displays the accumulated flow; the second mode displays pressure, the sign position displays "P", the upper line shows no display on the right, and the lower line displays the measured pressure value; the third mode displays temperature, the sign position displays "T", and the upper line displays "T" on the right side. No display, the lower row displays the measured temperature value. In the setting state, the LCD symbol is not displayed, the parameter number is displayed on the right side of the upper row, and the value of the parameter is displayed on the lower row, and there is an "activation" digit on the lower row, which is flashing. The activation digit can be moved by the shift key, and the increment key can be Change the value of the active digit, and the confirm key will pass the modified value to the parameter. After modifying the parameter, the LCD enters the next instrument parameter modification screen.

The system of the present invention utilizes the comparative output function of the 2407A timer to realize the pulse output, and the pulse output part circuit is shown in FIG. 12 . According to the measured frequency value and the clock frequency used by the system, calculate the preset number of cycles required by the timer, so as to determine the frequency of the output pulse. The set value of the comparison output unit is used to set the duty cycle of the output pulse, usually the duty cycle is set to 1. The final output pulse signal is the processed instantaneous flow signal, which is used for system calibration and actual application of the host computer for measurement and counting. At the same time, it is also set as an integer pulse signal output according to user requirements, that is, one pulse represents a certain volume flow. .

The system of the present invention adopts DAC (digital/analog conversion) and V/I (voltage/current conversion) circuits to realize the current output of 4-20mA, as shown in FIG. 13 . After the 2407A calculates the frequency, according to the upper and lower limit frequency values preset by the user, the frequency is converted into a corresponding digital quantity and sent to the DAC. The DAC converts it into a voltage, and then converts it into a current signal output by V/I. The system of the present invention provides separate power supply for the analog circuit part and the digital circuit part to prevent interference. The analog part requires 5V, and the digital part requires both 5V and 3.3V. Generally, instruments are powered by 24V or 12V. Considering the requirements of low power consumption of the system, the system adopts a high-efficiency DC/DC conversion power supply, which converts the input 12V or 24V voltage into 5V voltage, and then passes through the low dropout voltage regulator ( LDO) completes the 5V/3.3V conversion and provides it to the digital circuit part. The power supply requirements of the analog part of the system are relatively high, so it is directly supplied through the linear voltage regulator circuit, as shown in Figure 14.

The main task of the system of the present invention is to use the selected signal processing algorithm to complete the flow measurement, and it has the functions of pulse output, current output, setting of instrument coefficient and storage of measurement results. Therefore, there are many functions and the software design is complicated. In order to facilitate the design and maintenance of the system, the subroutines that complete specific functions or similar functions are combined into modules, namely function modules, which are then uniformly called by a main monitoring program. These functional modules can also call each other. The software block diagram of the system is shown in Figure 15.

The initialization module mainly includes the initialization of 2407A, the initialization of LCD, and the initialization of program global variables.

The 2407A chip integrates many functional modules, including A/D with sorting function, serial peripheral interface SPI, general I/O port, watchdog, emulation port, event manager, clock phase-locked loop (PLL), etc., each functional module needs to initialize the corresponding register before use to determine its working mode.

Many global variables are set in the system program to reflect the current working status of the system and facilitate the calling of subroutines of each module. Some of the variables need to be initialized to facilitate the next step of the program. These variables are roughly divided into two categories: global variables that need to be saved and global variables that are not saved. The global variables to be saved generally refer to the parameters of the instrument, which can be changed through keyboard settings or automatically saved in EEPROM when the system is powered off. When initializing, take the parameter value from EEPROM and put it in the corresponding global variable; the unsaved global variable does not need to be saved in EEPROM, but some program modules are frequently called, such as the trigonometric function table, if the trigonometric function table is called every time function will reduce the efficiency of the program, so the values to be used can be configured in the initialization module, which can improve the execution speed of the program.

After the system is powered on, the LCD should display such as instantaneous flow, cumulative flow, etc. The specific content is determined according to actual needs. The LCD needs to be initialized before use, and the initialization steps are carried out according to the LCD manual. LCD initialization mainly includes: control pin initialization, write command word, define internal oscillator, start internal oscillator, initialize display content, etc.

Macro definitions are also part of system initialization, implemented by precompiled instructions. There are many constants that need to be used by subroutines in the system, such as the period of timer interrupt, LCD refresh frequency, the measurement range of temperature and pressure, and the corresponding address of the instrument parameters stored in EEPROM. In order to reduce storage space and facilitate system maintenance, these constants can be defined in the macro definition module.

The signal processing module completes tasks such as spectrum analysis and processing of vortex signal, instantaneous flow calculation, temperature and pressure compensation, etc. The basic process of the periodogram analysis algorithm adopted by the system is: perform FFT transformation on the sampled data group, then calculate the power spectrum of each discrete frequency point, and then preliminarily estimate the frequency of the signal according to the principle of maximum energy. Centroid spectrum correction is then required to improve the accuracy of frequency estimation. At the same time, it is also necessary to average the results of multiple calculations to reduce the impact of system random errors. According to the actual situation, parameters such as sampling frequency, number of sampling points and average times are set through the keyboard to control the accuracy of the algorithm. Fig. 16 is a flow chart of the signal processing algorithm of the system of the present invention.

Calculate the instantaneous flow rate based on the obtained signal frequency and the instrument coefficient set by the system. For small flows, due to the limitation of the working principle of the vortex flowmeter or possible noise dominance, the sampled signal does not necessarily represent the flow, so it should be cut off.

The density of the fluid is greatly affected by temperature and pressure, especially when the fluid is a gas. Therefore, temperature and pressure compensation must be performed when measuring the mass flow rate of the fluid. Different fluid flows are affected differently by temperature and pressure. The system is equipped with four different compensation methods according to the fluid to be measured: liquid, general gas, superheated steam and saturated steam. Liquid flow is generally not affected by pressure and temperature factors, so it is not compensated; general gas flow has a certain law with the change of pressure and temperature, and its flow compensation can be carried out through the general gas temperature and pressure compensation formula; the flow of superheated steam and saturated steam Compensation is realized through look-up table interpolation of respective temperature, pressure and density tables.

The 4-20mA current output provided by the present invention is used to reflect the measurement result of the flow rate, so as to meet the needs of the DCS control system on the industrial site. The specific process is: after the system calculates the result, the 10-bit data is serially output to the D/A converter through the serial peripheral interface SPI integrated in the 2407A, converted into a voltage, and then converted by V/I to output a current of 4-20mA . In the software, it is only necessary to write specific data to the send register SPITXBUF in the initialized SPI module at regular intervals.

The pulse output module uses the general timer T4 in the 2407A manager B to provide the pulse output of the system. First initialize manager B. After the system enters the main loop, it continuously writes the data corresponding to the instantaneous flow into the registers T4PR and T4CMPR. The specific formula is:

$\mathrm{<span\; class="notranslate">\; <figure-callout\; id="4"\; label="T"\; filenames="C20051004105800153.png,C20051004105800161.png"\; state="\{\{state\}\}">T</figure-callout></span>}\mathrm{<span\; class="notranslate">\; 4</span>}\mathrm{<span\; class="notranslate">\; PR</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; CLK</span>}}{\mathrm{<span\; class="notranslate">\; flow</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}$

T4CMPR＝0.5*T4PR

Among them, CLK and flow are respectively the counting clock of the timer and the calculated instantaneous flow. The registers T4PR and T4CMPR are 16-bit integers, so the fractional part of the calculation result is ignored, which will inevitably lead to an error in the output frequency. In the software design, the comparison interrupt of the timer can be used to accumulate the fractional part, and dynamically compensate the period of the output pulse to improve the accuracy of the pulse.

The human-machine interface module is composed of a keyboard and an LCD. The LCD is responsible for displaying the measurement results and parameters of the instrument; the keyboard cooperates with the LCD to modify the instrument parameters and switch the working status of the LCD.

The flow chart of keyboard scanning is shown in Figure 17. In the figure, Counter is the counter used for software debounce, First is the variable for distinguishing continuous operation of the keyboard, and T is the time interval for continuous operation of the keyboard.

The 2407A allows processing of 6 core-level maskable interrupts, and uses a centralized peripheral interrupt expansion controller to handle all on-chip peripherals and external pin interrupt priorities and interrupt responses. When the peripheral interrupt is valid, its corresponding interrupt vector is automatically stored in the peripheral interrupt vector register (PIVR). When the CPU responds, it reads the interrupt vector from the PIVR and finds the entry of the interrupt service subroutine in the interrupt vector table. The programming of the interrupt module is mainly the filling of the interrupt vector table and the specific operation of the interrupt service subroutine.

The system uses three core interrupts: INT1, INT2 and INT3.

INT1 contains two peripheral interrupts, which are interrupts from PDPINTA and A/D conversion completion. The PDPINTA pin is used to monitor the system voltage. When the system is powered off, the external monitoring circuit sends an interrupt request to the DSP through PDPINTA, requesting to save the accumulated flow in EEPROM. This interrupt has the highest priority. The second interrupt is used for signal sampling. The system arranges the timer 1 period interrupt of 2407A to automatically trigger and start the A/D module. The A/D module integrates an automatic sorting module, which can sample and convert up to 16 signals at a time. This system needs Take 3 signals: flow, temperature and pressure. After the A/D completes a conversion, it puts the quantized result of the conversion in the result register, and sends an interrupt request to the CPU at the same time. After the interrupt response, the interrupt service subroutine saves the value of the result register in the data buffer, and performs digital filtering of the flow signal at the same time.

INT2 is triggered by the periodic interrupt of Timer3. In the interrupt service subroutine of INT2, there are 4 software counters. The first one is used to process the scanning of the keyboard. When the counter counts to a certain value, the keyboard scanning program is called. If there is a key pressed, the corresponding key operation will be executed. If there is no key, it will exit. The second one is used to refresh the LCD: if the LCD is in the display state, the display value of the LCD will be refreshed regularly. The third is used to complete the blinking action of the selected active digit in the LCD setting state. The last one is used to monitor and restore LCD working status. If the LCD is in the setting state, and there is no key operation for a certain period of time, the system will no longer perform the setting operation by default, and automatically return to the LCD display state and continue to display the measurement results. Because this interrupt is mainly used to operate the keyboard and LCD display, and the time requirement is not very high, so the interrupt priority is low priority.

1NT3 is triggered by the comparison interrupt of timer 4, indicating that the system currently outputs a pulse. Because the pulse frequency required to be output cannot always maintain an integer multiple of the working clock, there is always a certain error in the output pulse. Using this interrupt, the period of the output pulse can be dynamically compensated to improve the accuracy of the output pulse. At the same time, the interrupt is used to accumulate the output pulses, which is convenient for accumulating flow.

In order to facilitate the maintenance of the program, the interrupt periods of all timers are defined in the macro definition module and assigned to the corresponding registers in the initialization module. In addition, fill in the entry of the corresponding interrupt service subroutine in the corresponding position in the interrupt vector table.

The watchdog module integrated in the DSP chip is used to guarantee the recovery of the system failure. The watchdog has a hardware counter, which starts counting continuously after power-on reset, and triggers a system reset when the counter overflows. Therefore, add a watchdog counter reset and clear statement at an appropriate position to prevent the watchdog counter from overflowing. If the system is caught in a software glitch, causing it to freeze, the watchdog counter will overflow and reset the system.

The main monitoring program is the general scheduler of the entire signal processing system, which calls the subroutines in each module to realize all the functions of the instrument. The basic process is as follows: after the system is powered on and reset, it is initialized immediately; after initialization, the periodic map signal is processed according to the sampled vortex signal, and the signal frequency is calculated; then the instantaneous flow is calculated according to the set meter coefficient; according to different Select different temperature and pressure compensation methods for the fluid to be measured, and perform temperature and pressure compensation; after measuring the flow, output the corresponding pulse amount and standard 4-20mA current; after completing the output, if the system needs to modify the instrument parameter signal, modify the instrument parameters, after that, the main monitoring program will return to the signal processing, continue to process the signal, and complete the real-time measurement of the flow of the instrument in a continuous cycle. The process is shown in Figure 18.

Claims (5)

1. low cost intelligent vortex shedding flowmeter signal processing system based on DSP, by analog signal conditioner module, dsp chip and extend out that RAM, EEPROM, low-voltage observation circuit, LCD show, keyboard input, analog quantity output, pulse output and power management module form; It is characterized in that including in the described analog signal conditioner module temperature signal regulation circuit, pressure signal conditioning circuit and flow sensor signal modulate circuit, described temperature signal regulation circuit is made up of electric bridge and differential amplifier circuit, described pressure signal conditioning circuit is made up of the I/V translation circuit, and described flow sensor signal modulate circuit is made up of the active bandwidth-limited circuit of charge amplifying circuit, band amplitude limit function and low-pass filter circuit with fixed gain; Described system accepts temperature, pressure and the flow detecting signal of sensing unit output and deliver to described DSP after described analog signal conditioner resume module; Utilize its inner integrated A/D modular converter and annular data buffer that temperature, pressure and flow three road electric signal are carried out timing sampling and regularly calculating by described DSP, or calculate on the sampling edge limit; Described DSP carries out digital filtering to the flow sensor signal that may be mixed with the noise that is caused by interference that its sampling obtains, and adopts the power spectrum analysis method that filtered data are carried out analysis of spectrum and frequency spectrum correction again; Average according to result of calculation repeatedly, determine peak power spectrum value, obtain its pairing frequency, be the frequency of signal; According to instrument coefficient with by temperature, pressure compensation, calculate instantaneous delivery and integrated flux in real time again; Described DSP utilizes its timer internal to compare output function the instantaneous delivery data is exported by the pulse output unit with impulse form; Described DSP utilizes its inner integrated SPI module that the instantaneous delivery data are sent to described analog quantity output unit, with the analog quantity form output of 4～20mA electric current; Described DSP is presented at instantaneous delivery, integrated flux, temperature and pressure on the described LCD; Include instrument coefficient, systematic sampling frequency, sampling number, repeatedly average time, the digital filter coefficient of analysis of spectrum by what keyboard was provided with; Preserve the instrument coefficient of described keyboard input and the integrated flux that calculates by described EEPROM; Described low-voltage observation circuit is responsible for proposing interrupt request to described DSP when system's power down, and before the complete power down of system measurement result is stored among the described EEPROM.

2. according to the described low cost intelligent vortex shedding flowmeter signal processing system of claim 1 based on DSP, it is characterized in that utilizing the cycle interruption of described dsp chip timer internal, realization is to the timing sampling of described temperature, pressure and flow signal, the data that sampling obtains are put into annular data buffer, and regularly the flow signal that sampling is obtained carries out described analysis of spectrum, sampling edge is calculated on the limit, the covering of overlapping of the data on flows of adjacent twice analyzing and processing.

3. according to the described low cost intelligent vortex shedding flowmeter signal processing system of claim 1 based on DSP, it is characterized in that described dsp chip is TMS320LF2407A, its inside is integrated with FLASH, the A/D conversion functional module of 32K, 2 task managers, house dog, serial peripheral equipment interface SPI, serial communication interface SCI, CAN functional module.

4. according to the described low cost intelligent vortex shedding flowmeter signal processing system of claim 1 based on DSP, it is characterized in that range ability according to described vortex shedding flow meter, by described keyboard sample frequency and sampling number are set, to control the computational accuracy of described power spectrumanalysis; Described dsp chip adopts the computational accuracy of center of gravity spectrum correcting method with further raising frequency; According to described flow sensor characteristic, by described keyboard segment instrument coefficient is set, improve the measuring accuracy of flow; Form by the temperature and pressure compensation carries out gamma correction to flow result of calculation, improves the measuring accuracy of overheated gas and saturated vapour flow.

5. according to the described low cost intelligent vortex shedding flowmeter signal processing system of claim 1 based on DSP, it is characterized in that utilizing described dsp chip timer internal, export pulse with mode relatively, and by interrupting that relatively the recurrence interval of output is carried out dynamic compensation.

Images