Application of ARM's equal-precision frequency measurement in unit speed measurement and control

There are two traditional measurement methods. One is the frequency measurement method (M method), which counts the pulses of the measured signal within the gate time (T-Nfo, the time of N reference signal pulses) (the count value is M). The frequency of the measured signal is , and the error is

The other is the cycle measurement method (T method), which is to count the reference pulses within one cycle of the measured signal (the count value is N). The frequency of the measured signal is and the error is .

Among them, is the accuracy of the reference signal frequency, which can usually reach ; for the frequency measurement method, within the same gate time, for any f, it cannot be guaranteed that there are exactly M T within the T time, so a maximum quantization error of ±1 T will be generated, and as the measured frequency f decreases, M decreases, and the error increases. Therefore, the frequency measurement method only has good measurement accuracy for high-frequency signals; for the period measurement method, as the measured frequency f increases, N becomes smaller, and the error increases. Therefore, the period measurement method only has good measurement accuracy for low-frequency signals. When the measurement range is relatively wide, the combination of the above two methods is undoubtedly effective in improving the measurement accuracy, but there are the following problems: first, the measurement accuracy of the entire frequency band is inconsistent; second, the measurement method is frequently switched near the mid-frequency, the error is large, and the real-time performance is poor.

2 Principle and error analysis of equal precision measurement method

The equal-precision frequency measurement method is a frequency measurement method developed on the basis of the traditional frequency measurement method, and has been increasingly used in frequency measurement in various fields.

The principle of the equal-precision frequency measurement method is shown in Figure 1.

Set up two counters, counter 1 counts the measured signal, and counter 2 counts the reference signal. A gate time T is set in advance. After the measurement starts, when the next leading edge of the measured signal arrives, counter 1 and counter 2 are opened synchronously to start counting. After the gate time arrives, counter 1 and counter 2 do not stop counting until the leading edge of the measured signal arrives, and then counter 1 and counter 2 are closed synchronously. The frequency of the measured signal can be expressed as: The error is: , where M is the count value of counter 1, N is the count value of counter 2, and f is the reference frequency. It can be seen that it is the same as the expression of the traditional frequency measurement method. The difference is that the work of counter 1 is synchronously opened and closed by the measured pulse, so there is no counting error, that is, 99. It can be seen that the measurement accuracy of this method does not change with the frequency of the measured signal. The effective number of digits displayed by the measured value is the same within the full range, that is, equal precision measurement. In general, 1010, so the measurement error of this method is mainly caused by the ±1 error in the counting of the reference signal. Therefore, it can be seen that the higher the frequency of the reference signal, the higher the measurement accuracy under the same gate time. In addition, the longer the gate time T is, the more counts N are, and the higher the measurement accuracy is. However, T and N are restricted by many factors and cannot be increased arbitrarily. First of all, it is the requirements of the project. To reflect and understand the degree of change in speed, a shorter time must be used. The speed of the turbine unit changes from 0 to 50 Hz during the start-up and shutdown process, and does not exceed 100 Hz. In practical applications, the gate time and reference signal frequency can be appropriately selected to enable the measurement to achieve high-precision and fast measurement in the full frequency band. [page]

3 Implementation in ARM Measurement System

Using the equal precision measurement method, the unit speed measurement and control system uses ARM7's LPC2214 as the CPU. LPC2214 has two 32-bit timers/counters, each of which has the following characteristics:

1) 32-bit timer/counter with programmable 32-bit prescaler.

2) Each timer has four 32-bit capture channels that can capture the instantaneous value of the timer when the input signal jumps. The capture event can optionally generate an interrupt.

3) Four 32-bit match registers: ① Continuous operation, with the option to generate an interrupt on match; ② Stop timer on match, with the option to generate an interrupt; ③ Reset timer on match, with the option to generate an interrupt.

4) Each timer has 4 external outputs corresponding to the match registers, with the following characteristics: ① Set to low level when matching; ② Set to high level when matching; ③ Flip when matching; ④ Remain unchanged when matching.

In this system, the matching function of timer/counter 0 is used to control the gate time, and the capture function of timer/counter 1 is used to monitor the measured signal. The measured signal is transformed into a square wave signal with the same frequency through the hardware shaping circuit, connected to the capture pin of timer/counter 1, and its capture interrupt function is opened. The software responds to the interrupt and performs corresponding processing. The counting frequency is 2.211 840 MHz, the gate time is 0.5 s, and the error is 1111, of which 1212, so it can be concluded that, in theory, the measurement accuracy of equal-precision frequency measurement in this system is better than that of the other, which can fully meet the needs of the project.

In the turbine speed measurement and control system, in order to improve the reliability of measurement, a measurement method is adopted in which two types of signals, electrical (machine end residual pressure) signal and gear disk mechanical pulse signal, are input simultaneously. The calculated frequency value is used for display, output control, etc. The circuit structure block diagram of the measurement system is shown in Figure 2.

The system program mainly includes subroutines such as initializing timer/counter, capturing interrupt processing, frequency calculation, display, and output control. The system flow is shown in Figure 3.

Use the RIGOL function/arbitrary waveform generator to generate a frequency-varying sine waveform: ① The frequency of the simulated waveform decreases from 100 Hz to 0.2 Hz within 500 s, with a decreasing gradient of 0.1996 Hz/s; ② The frequency of the simulated waveform decreases from 100 Hz to 1 Hz and then increases from 1 Hz to 100 Hz, with each 10 Hz lasting for 18 s; ③ The measured data is recorded every 0.5 s, and waveforms 1 and 2 are obtained as shown in Figure 4.

Experiments have shown that using the equal-precision frequency measurement method can achieve relatively high measurement accuracy in the entire measurement range. The maximum relative error of this test system is less than 10~. The gate time is 0.5 s, which can quickly respond to changes in the unit speed.

4 Conclusion

Compared with the traditional frequency measurement method and period measurement method, the equal-precision measurement method can achieve equal precision in the entire frequency band, greatly improving the measurement accuracy. The test results show that the maximum relative error of the measurement is better than that of the conventional frequency measurement method and the period measurement method in the turbine speed measurement and control system, and the gate time is variable, which can quickly respond to the change of the unit speed. The developed SJ-22D microcomputer speed measurement and control device has been put into use in many hydropower stations (plants), and its operation is accurate and reliable.