Method and device for measuring rotating speed of steam turbine

Abstract

The invention discloses a method and device for measuring the rotating speed of a steam turbine, and the method comprises the steps: carrying out the blocking and filtering of a rotating speed signal collected in real time, generating a filtering signal, setting a threshold voltage through a hysteresis circuit, and outputting a pulse signal; the pulse signal is subjected to optocoupler isolation and signal wave shaping, and then a square wave signal is output; the frequency value of the rotating speed is calculated based on the square wave signal, the rotating speed value is obtained through further calculation, and finally the obtained rotating speed value is filtered by combining a Hampel filter and a moving average filter. Compared with a traditional rotating speed measurement mode, the frequency measurement precision is guaranteed, and meanwhile the anti-interference capacity of the measurement device is enhanced.

Description

Technical Field

[0001] The invention relates to the field of rotation speed measuring devices and denoising methods, and in particular to a rotation speed measuring device and a measuring method for a steam turbine. Background Art

[0002] A steam turbine is a rotating machine that converts thermal energy into mechanical energy. It is also a commonly used power equipment in the industrial field. The speed is an important parameter that reflects the operating status of the steam turbine. By monitoring the speed signal, the abnormal status of the equipment can be discovered in time, so that corresponding measures can be taken to reduce the occurrence rate of equipment accidents. However, steam turbines often work in harsh environments and complex conditions. The measured speed signal is easily mixed with noise interference. Therefore, in order to ensure the stable and reliable operation of the equipment, it is necessary to ensure the measurement accuracy, and at the same time effectively identify and eliminate abnormal values ​​and wild points in the measurement signal.

[0003] The traditional frequency measurement method and period measurement method will produce a pulse error in the frequency count value. When the frequency of the measured signal changes, the measurement accuracy will increase due to the frequency change. The steam turbine equipment is subject to electromagnetic interference, and the speed measurement conditioning module has some deficiencies in interference suppression adaptability, resulting in the speed signal measured under certain operating conditions being mixed with some wild points, which seriously interferes with the normal operation of the steam turbine. Summary of the invention

[0004] The purpose of the present invention is to solve the defects existing in the prior art and provide a method for measuring the speed of a steam turbine, which can eliminate the error in the number of pulses generated by counting the measured signal, eliminate abnormal values ​​mixed in the speed signal, and ensure the speed detection accuracy and measurement accuracy.

[0005] In order to achieve the above object, the technical solution provided by the present invention is as follows:

[0006] A method for measuring the rotation speed of a steam turbine comprises the following steps:

[0007] Step 1: Collect the speed signal in real time;

[0008] Step 2, the speed signal is subjected to direct current isolation and filtering processing by a filtering circuit to generate a filtered signal;

[0009] Step 3: Based on the filtered signal, a threshold voltage is set through a hysteresis circuit to output a pulse signal;

[0010] Step 4: The pulse signal is isolated by an optical coupler and rectified to output a square wave signal;

[0011] Step 5, calculating the frequency value of the rotation speed based on the square wave signal, and further calculating to obtain the rotation speed value; the frequency value of the rotation speed is calculated using an equal-precision frequency measurement method;

[0012] Step 6: After filtering out abnormal values ​​from the obtained speed value using a Hampel filter, a sliding average filter is used for filtering.

[0013] The present invention adopts an equal-precision frequency measurement method to calculate the frequency value of the speed signal, which can effectively avoid the pulse error caused by counting and ensure the measurement precision and accuracy of the speed; at the same time, the speed value data is filtered by combining Hampel filtering and sliding average filtering to improve the denoising effect.

[0014] Furthermore, the steps of filtering in the above step 6 are as follows:

[0015] 6.1. Express the real-time acquired speed value data as a time series X1 = {x 1 ,x 2 ,x 3 ,...,x t}, x(t) is the speed value data obtained at the tth time, t = 1, 2, ..., N, N is the number of data;

[0016] 6.2. Setting parameters of the Hampel filter, wherein the parameters of the Hampel filter include a window size K and a threshold coefficient μ;

[0017] 6.3. Calculate the median mid in window K i , and calculate the absolute deviation σ between the sample in the window and the median according to the median i ;

[0018] 6.4. Use Hampel filter to filter the data x of sequence X1 i Filtering is performed, and the filtered data is h i , if |x i -mid i |>μ·σ i When i Get the median mid i , that is, x i It is identified as an outlier, otherwise the original data remains unchanged;

[0019] 6.5. x(t+1) is the speed value data obtained at time t+1. x(t+1) is stored in sequence X1, and the earliest cached data x in X1 is discarded. 1 , then the sequence is X2={x 2 ,x 3 ,...,x t ,x t+1}, repeat steps 6.3 and 6.4, so that the speed value data collected at time t+1 is filtered;

[0020] 6.6. When new speed value data is obtained, by repeating steps 6.3 and 6.4, the data H1 processed by the Hampel filter is obtained. 1 ,h 2 ,h 3 ,...,h i}, h i is the data after Hampel filtering at the i-th moment;

[0021] 6.7. Then, the moving average filter is used to filter the data sequence H1={h 1 ,h 2 ,h 3 ,...,h i} Perform filtering again, set the sliding window parameter SW of the sliding average filter, and use the sliding average filter to filter the sequence H1 = {h 1 ,h 2 ,h 3 ,...,h i} data point h i Perform filtering and calculate the arithmetic mean value of the sliding window SW except the maximum and minimum values ​​as the data point h i Filtered data y i ;

[0022] 6.8, h i+1 is the data after Hampel filtering at the i+1th moment, and h i+1 Save to sequence H1, and discard the earliest cached data h 1 , at this time the sequence is H2 = {h 2 ,h 3 ,...,h i ,h i+1}, repeat step 6.7 to complete h i+1 The data is processed by sliding average filtering;

[0023] 6.9. When new data after Hampel filtering is obtained, the data after sliding average filter filtering Y={y 1 ,y 2 ,y 3 ,...,y h}.

[0024] Among them, the calculation formula of the hampel filter is:

[0025]

[0026] The calculation formula of the sliding average filter is:

[0027]

[0028] Where H' 1 For data h i The corresponding sliding average sequence H1' = {h' 1 ,h' 2 ,h' 3 ,...,h' i}; max(H1') and min(H1') are the sequence H' 1 The maximum and minimum values ​​of .

[0029] Furthermore, the medium-precision frequency measurement method in step 5 above uses two counters, one as a counter for the measured signal and the other as a counter for the standard signal. The measured signal is the square wave signal output in step 4, and the standard signal is the clock frequency. The frequency value is calculated using the following formula:

[0030] f x =(N x / N s )·f s

[0031] Where: N x is the frequency count value of the measured signal, N s is the frequency count value of the standard signal, f s is the frequency of the standard signal, f x is the frequency value of the measured signal;

[0032] According to the relationship between the speed and the number of gears, the speed value of the measured signal per minute is calculated. The calculation formula is:

[0033] R=(f x ·Z 2 ·60) / (Z 1 ·Z 3 )

[0034] Where: f x is the frequency value of the measured signal, Z 1 is the number of shaft gears, Z 2 is the measured number of teeth, Z 3 is the number of teeth being measured.

[0035] The present invention also provides a steam turbine speed measuring device using the above-mentioned measuring method, comprising:

[0036] The I / O board transmits the speed signal to the speed function module through the backplane, and leads the output signal of the speed function module to the wiring terminal;

[0037] The speed function module is connected to the I / O board through the backplane, and performs DC isolation, filtering, hysteresis comparison, isolation and signal rectification on the received speed signal in sequence, then outputs a square wave signal and calculates the speed value;

[0038] The backplane, as the middleware between the speed function module and the I / O board, connects the speed function module and the I / O board.

[0039] Furthermore, the I / O board includes a signal input circuit, a buffer signal output circuit, a DO output circuit, and an AO output circuit.

[0040] Furthermore, the speed function module includes a speed measurement channel module, an AO / DO output module, an MCU, a power supply circuit and two 485 circuits.

[0041] Furthermore, the rotation speed measurement channel module includes a signal input detection circuit, a filter circuit, a hysteresis circuit and an isolation and rectification circuit which are connected in sequence.

[0042] Among them, the MCU is based on the CORTEX-M4 processor, the power supply circuit provides power for the speed function module and other devices, and the 2-way 485 circuit provides dual-way redundant serial port communication.

[0043] Compared with the prior art, the present invention has the following advantages:

[0044] 1) The steam turbine speed measuring device of the present invention is based on a modular concept, and the input and output of the signal are separately designed into an I / O module, which reduces the coupling and is conducive to maintenance and prolonging the service life of the product.

[0045] 2) The speed measurement channel module of the device performs DC isolation and filtering on the measured signal, and has a certain anti-interference ability; the MCU uses the equal-precision frequency measurement method to calculate the frequency value of the measured signal, avoiding the pulse error caused by counting, and ensuring the measurement precision and accuracy of the speed.

[0046] 3) The device also has signal buffer output, AO output, and DO output functions. It communicates with the communication controller through the 485 bus and transmits data for real-time monitoring.

[0047] 4) The Hampel filter can filter out outliers in the signal while retaining the characteristics and trends of the original data. The sliding average filter is then used to process the data to ensure the stability of the signal. The speed signal is filtered by combining the Hampel filter and the sliding average filter to improve the denoising effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 The system structure block diagram of the steam turbine speed measuring device of the present invention;

[0049] Figure 2 This is a signal denoising process flow chart based on a steam turbine speed measuring device of the present invention;

[0050] Figure 3 This is an example of a steam turbine original speed signal curve diagram in an embodiment of the present invention;

[0051] Figure 4 The figure is a speed signal curve diagram of a steam turbine after denoising in an embodiment of the present invention. DETAILED DESCRIPTION

[0052] The present invention will be further described below in conjunction with the accompanying drawings.

[0053] like Figure 1 As shown, the present invention provides a steam turbine speed measuring device, which includes three parts: a speed function module, a backplane and an I / O board. The speed function module is connected to the I / O board through the backplane to perform signal acquisition, analog A0 output, signal buffer output, and relay output, and can communicate with the communication controller through the 485 bus and transmit data and configuration information; the speed function module is integrated by a speed measurement channel module, an AO / DO output module, a 2-way 485 circuit, an MCU and a power supply circuit.

[0054] The speed measurement channel module consists of a signal input detection circuit, a filter circuit, a hysteresis circuit, an isolation and rectification circuit. The AO / DO output module includes 4 DO outputs and 1 AO output, and is composed of a channel signal output circuit, an operational amplifier, an isolator, and an optocoupler relay. The MCU is based on a CORTEX-M4 processor, the power supply circuit provides power for the speed function module and other devices, and the 2-channel 485 circuit provides dual redundant serial port communication.

[0055] The I / O board is responsible for transmitting the sensor signal to the speed function module through the backplane, and leading the output signal of the speed function module to the wiring terminal; the I / O board is integrated with the signal input circuit, the buffer signal output circuit, the DO output circuit, and the AO output circuit.

[0056] The backplane serves as middleware between the speed function module and the I / O board, connecting the speed function module and the I / O board.

[0057] The working principle of the speed measurement channel module in the turbine speed measurement device is as follows: the signal input detection circuit detects the sensor signal output by the eddy current sensor or magnetoresistive sensor connected to the I / O board, and then uses the filter circuit to isolate and filter the sensor signal obtained by the signal input detection circuit to generate a filtered signal. The threshold voltage is set based on the filtered signal output by the filter circuit through the hysteresis circuit to output a pulse signal; the pulse signal output by the hysteresis circuit is finally isolated and rectified by the isolation and rectification circuit to output a relatively standard square wave signal.

[0058] The square wave signal output by the speed measurement channel module is input into the MCU for calculating the frequency value of the speed. The MCU uses the equal-precision frequency measurement method to calculate the frequency value of the signal, and then obtains the original speed value based on the relationship between the speed and the number of shaft teeth. The signal denoising method combining the Hampel filter and the sliding average filter is then used to identify and eliminate abnormal values ​​and wild points in the speed signal to obtain a smoother signal.

[0059] like Figure 2 As shown, the present invention is based on the signal denoising processing of the steam turbine speed measuring device, comprising the following steps:

[0060] Step 1: The speed signal data obtained by the turbine speed measuring device is expressed as a time series X1={x 1 ,x 2 ,x 3 ,...,x t}, x(t) is the speed signal data collected at the tth time, t = 1, 2, ..., N, N is the number of data;

[0061] Step 2: Use the Hampel filter to perform filtering, and set the window size K and threshold coefficient μ of the Hampel filter;

[0062] Furthermore, in step 2, the window K of the Hampel filter is set to 14, and the threshold coefficient μ is set to 2;

[0063] Step 3: Calculate the median mid in the window i , and calculate the absolute deviation σ between the sample in the window and the median according to the median i ;

[0064] Step 4: Use the Hampel filter to filter the data points x of the sequence X1. i Filtering is performed, and the filtered data is h i , if |x i -mid i |>μ·σ i When i Get the median mid i , that is, x i It is identified as an outlier, otherwise the original data remains unchanged. The calculation formula of the Hampel filter is:

[0065]

[0066] Step 5: When new speed signal data is obtained, the data H1={h 1 ,h 2 ,h 3,...,h i};

[0067] Step 6: For the sequence H1 in step 5 = {h 1 ,h 2 ,h 3 ,...,h i} The data is processed by sliding average filtering, h i is the data after Hampel filtering at the i-th moment;

[0068] Step 7: Set the sliding window SW of the sliding average filter to 8, and use the sliding average filter to filter the sequence H1 = {h 1 ,h 2 ,h 3 ,...,h i} data h i For filtering, the sliding average filter calculation formula is:

[0069]

[0070] Furthermore, in step 7, H' 1 For data h i The corresponding sliding average sequence H1' = {h' 1 ,h' 2 ,h 3 ',...,h' i}; max(H1') and min(H1') are the sequence H' 1 The maximum and minimum values ​​of ;

[0071] Step 8: When the new data after Hampel filtering is obtained, the data after sliding average filtering Y={y 1 ,y 2 ,y 3 ,...,y h}.

[0072] Figure 3 The figure shows the data curve of the steam turbine speed measuring device before filtering. The given value of the steam turbine speed is 2000r / min. Because it is interfered, some abnormal values ​​and wild points appear in the speed curve.

[0073] Figure 4 This is the curve after the speed signal is filtered by combining the Hampel filter and the sliding average filter. As can be seen from the figure, the outliers and wild points of the speed signal are eliminated, while the characteristics of the original data are also maintained, which has a good filtering effect.

Claims

1. A method for measuring the rotation speed of a steam turbine, characterized in that: The following steps are involved: Step 1: Collect the speed signal in real time; Step 2, the speed signal is subjected to direct current isolation and filtering processing by a filtering circuit to generate a filtered signal; Step 3: Based on the filtered signal, a threshold voltage is set through a hysteresis circuit to output a pulse signal; Step 4: The pulse signal is isolated by an optical coupler and rectified to output a square wave signal; Step 5, calculating the frequency value of the rotation speed based on the square wave signal, and further calculating to obtain the rotation speed value; the frequency value of the rotation speed is calculated using an equal-precision frequency measurement method; Step 6: After filtering out abnormal values ​​from the obtained speed value using a Hampel filter, a sliding average filter is used for filtering.

2. The method for measuring the speed of a steam turbine according to claim 1, characterized in that: The steps of filtering in step 6 are as follows: 6.

1. Express the real-time acquired speed value data as a time series X1 = {x1, x2, x3, ..., x t }, x(t) is the speed value data obtained at the tth time, t = 1, 2, ..., N, N is the number of data; 6.

2. Setting parameters of the Hampel filter, wherein the parameters of the Hampel filter include a window size K and a threshold coefficient μ; 6.

3. Calculate the median mid in window K i , and calculate the absolute deviation σ between the sample in the window and the median according to the median i ; 6.

4. Use Hampel filter to filter the data x of sequence X1 i Filtering is performed, and the filtered data is h i , if |x i -mid i |>μ·σ i When i Get the median mid i , that is, x i It is identified as an outlier, otherwise the original data remains unchanged; 6.

5. x(t+1) is the speed value data obtained at time t+1. x(t+1) is stored in sequence X1, and the earliest cached data x1 in X1 is discarded. At this time, the sequence is X2 = {x2, x3, ..., x t ,x t+1 }, repeat steps 6.3 and 6.4, so that the speed value data obtained at time t+1 is filtered; 6.

6. When new speed value data is obtained, by repeating steps 6.3 and 6.4, the data H1 = {h1, h2, h3, ..., h i }, h i is the data after Hampel filtering at the i-th moment; 6.

7. Then, the moving average filter is used to filter the Hampel filtered data sequence H1 = {h1, h2, h3, ..., h i } Perform filtering again, set the sliding window parameter SW of the sliding average filter, and use the sliding average filter to filter the sequence H1={h1,h2,h3,...,h i } data point h i Perform filtering and calculate the arithmetic mean value of the sliding window SW except the maximum and minimum values ​​as the data point h i Filtered data y i ; 6.8, h i+1 is the data after Hampel filtering at the i+1th moment, and h i+1 Save it to the sequence H1, and discard the earliest cached data h1. At this time, the sequence is H2 = {h2,h3,...,h i ,h i+1 }, repeat step 6.7 to complete h i+1 The data is processed by sliding average filtering; 6.

9. When the new data after Hampel filtering is obtained, the data after sliding average filter filtering Y={y1,y2,y3,...,y h }.

3. The method for measuring the speed of a steam turbine according to claim 2, characterized in that: The calculation formula of the hampel filter in step 6.4 is: The calculation formula of the sliding average filter in step 6.7 is: Among them, H′1 is the data h i The corresponding sliding average sequence H1' = {h'1,h'2,h'3,...,h' i }; max(H1') and min(H1') are the maximum and minimum values ​​of sequence H'1 respectively.

4. The method for measuring the speed of a steam turbine according to claim 1, characterized in that: The medium-precision frequency measurement method in step 5 uses two counters, one as a counter for the measured signal and the other as a counter for the standard signal. The measured signal is the square wave signal output in step 4, and the standard signal is the clock frequency. The frequency value is calculated using the following formula: f x =(N x / N s )·f s Where: N x is the frequency count value of the measured signal, N s is the frequency count value of the standard signal, f s is the frequency of the standard signal, f x is the frequency value of the measured signal; According to the relationship between the speed and the number of gears, the speed value of the measured signal per minute is calculated. The calculation formula is: R=(f x ·Z2·60) / (Z1·Z3) Where: f x is the measured signal frequency value, Z1 is the number of shaft gears, Z2 is the measured number of teeth, and Z3 is the measured number of teeth.

5. A steam turbine speed measuring device, characterized in that: The steam turbine speed measuring device comprises: The I / O board transmits the speed signal to the speed function module through the backplane, and leads the output signal of the speed function module to the wiring terminal; The speed function module is connected to the I / O board through the backplane, and performs DC isolation, filtering, hysteresis comparison, isolation and signal rectification on the received speed signal in sequence, and then outputs a square wave signal. After calculating the speed value, it performs Hampel filtering and sliding average filtering. The backplane, as the middleware between the speed function module and the I / O board, connects the speed function module and the I / O board.

6. The steam turbine speed measuring device according to claim 5, characterized in that: The I / O board includes a signal input circuit, a buffer signal output circuit, a DO output circuit, and an AO output circuit.

7. The steam turbine speed measuring device according to claim 6, characterized in that: The speed function module includes a speed measurement channel module, an AO / DO output module, an MCU, a power supply circuit and two 485 circuits.

8. The steam turbine speed measuring device according to claim 7, characterized in that: The rotation speed measurement channel module comprises a signal input detection circuit, a filter circuit, a hysteresis circuit and an isolation and rectification circuit which are connected in sequence.

Images