SU1190253A1 - System of automatic defectometry - Google Patents

Abstract

AUTOMATIC DEFECTOMETRIN SYSTEM, containing linear and angular coordinate sensors, measuring unit, the first input of which is connected to the first synchronizer output, the second and third inputs of which are connected to the first and second, automatic system of flaw detection, which input with the second synchronizer output, a control unit whose address bus is connected to the first input of the interface and storage blocks, an output device whose input is connected to the output of the interface block, and the data source of the pack The control unit is connected to the second inputs of the interface and storage units and to the output of the storage unit the third input of which is connected to the output Reading control unit, characterized in that, in order to control the accuracy and speed of processing the signals, the counter, i ,; peca, counting tripper, element .OR, one-shot, first multiplexer, first coordinate memory unit, and a second multiplexer and a second memory block, the output of which is connected to the control unit data bus, are inserted into each measuring channel , the first input is to the output of the second multiplexer, and the second input is to the first input of the first memory block of the coordinates of this channel, to the second inputs of the memory blocks of the remaining measurement channels and to the third input of the memory block, the fourth input of which is connected to the i third output of the interface unit, with the output Record of the control unit W and with the first input Record of the first and second multiplexers, output Confirmation of the capture of the control unit is connected to the control inputs of the first and second multiplexers, through the address counter on It is connected to the first input of the first address (G and the second multiplexer, and through the one-shot to the first input of the OR element and to the second inputs. Record O1 of the first and second multiplexers; OO; the first output of the measuring unit is connected to the first data input of the second multiplexer; the second output is with the second the input of the OR element, the output of which through the counting trigger is connected to the input of the Capture control unit, the output of the first multiplexer is connected to the second input of the first memory block, the output of which is connected to the data bus of the control unit, the second data inputs of the first and second multiplexers and with the input of the read control block, you

Description

sensor strokes linear and angular. The addresses of the control unit are connected. Coordinates are connected by a second input to the second address inputs of the first data of the first multiplexer, the bus and the second multiplexers.

1190253

one, . .-. The invention relates to the field of ultrasonic flaw detection and can be used in conjunction with the primary ultrasound equipment for monitoring weaning products (pipes, rolled products, etc.) in various areas of the machine. Performance control c. the whole is substantially dependent on speed. software input by the microprocessor to the memory through the interface unit of the flaw detection information. This disadvantage is especially pronounced in a multichannel system when entering each echo pulse (meaning connecting the measuring unit to several flaw detectors). At the same time, the availability of information about the whole burst of echo pulses and the joint processing of these signals received from different channels significantly increase the reliability of the control, since they allow to eliminate spurious signals of interference, to classify and determine important defect parameters. The limited speed of program input leads to a decrease in the performance of the control. Suppose - that the known system is to be used to monitor pipes with a speed bridge V 4 m / min, at P 4000 rev / m, the frequency of the probe pulses f 20 kHz and with the number of channels equal to four. The values for V,, n are chosen from the condition of reliable detection of a point defect and the proportions between them should not change. If the K580IK80 microprocessor is used as a calculator, the program implementing the input of the amplitude code of the echo pulse and the angular coordinate in the vector interrupt mode will last 45 ISS. At the same time, the distribution of time according to the functions performed is as follows; RSr response to vector interruption, 5 MKCJ LHlD — formation of the load address by channel number, 13 MKCJ. ) L2) P EMRL - transition to reading, information from the measuring unit, 5 ISS; M + OINAcDdfr - read the code of the amplitude of the echo pulse into the microprocessor, 5 ISS; M + fMOVM, A - transfer of the amplitude code to the memory at the download address, 3.5 μs-, 2INX N - increment of the load address, 2.5 MKCj IN do / r - read the angular coordinates of the microprocessor, 5 μs; , 4 MOVM | A - forwarding the code of the angular coordinate to the download address, 3.5 MKCj L - -51MHN - incrementing the loading address ,. ISS 2.5 and is. 45 MKS. Thus, if you do not reduce the frequency of the probed pulses, then with the number of channels more than one, information loss due to slow data entry will be extremely undesirable. For four, the loss of information will not occur if the frequency is 1/4 45-10 5.5 kHz. Since the proportions between, n, {must remain constant, the number of revolutions will drop to n 1100 r / min, and speed. control kost-do; 1.1 m / min Therefore, in order to increase the reliability of control due to the processing of flaw data, it is necessary to reduce the speed of control by almost 3.7 times.

The purpose of the invention is to increase the reliability of control when using multi-channel systems for the complete collection and processing speed of flaw detection signals by the system.

FIG. 1 is shown. block diagram of the system} in FIG. 2 is a block diagram of a microprocessor unit with auxiliary framing blocks in FIG. 3 is a block diagram of the interface blockade of FIG. 4 is a block diagram of a synchronizer.

The system for automatic defectometry contains a number of identical measuring channels consisting of measuring unit 1 and buffer memory 2 (BZU) of synchronizer 3, the first output of which is intended to connect to the input of the flaw detector, angular sensor 4, sensor 5 linear coordinates, the inputs connected with a scanning device, the storage unit 6, the interface unit 7, the control unit 8, the data bus and addresses of which, together with the Record output, are connected to the memory 6 and the interface 7 blocks, the output last It is connected to the input of the output device 9, and the output of the control unit 8 is fed to the storage unit 6, in addition to the system are connected in series the address counter 10, the input of which is connected to the output of the control unit 8 control, the multiplexer 11 coordinates and the unit 12 memory, the output of which is fed to the data bus of the control unit 8, and, in addition, the counting trigger 13, the output of which is also fed to the input of the capture of the control unit, and the one-shot 14.

Measuring unit 1 consists of series-connected peak detector 15, the first input of which is connected to the flaw detector, and a similarly-digital converter (ADC) 16, the second input of which is connected to the second output of the synchronizer 3, and the third input connected to the flaw detector.

The buffer memory contains serially-connected multiplexer 17 and memory block 18, the output of which is connected to the data bus of control block 8, the first multiplexing inputs

data, addresses and signals Recording at multiplexer 17 are connected respectively to the output of ADC-16, the output of counter 10 addresses and the output of one-vibrator 14, and the second inputs of multiplexing data, addresses and signal Record are fed to the data bus, addresses and output Record of control unit 8, In addition, the output Reading of control block 8 is connected to the BTopijiM input of all memory blocks 12 and 18, and the control inputs of all multiplexers 11 and 17 are connected to the output of the capture of control block 8, element OR 19.

FIG. 1 shows only one measuring channel, the rest consist of analog blocks; FIG. 2

microprocessors are designated 8-1,

clock generator 8-2, start control block 8-3, address bus 8-4 buffer blocks and data bus 8-5, system signal control block 8-6. In addition, the interface unit (Fig. 3) contains a decoder 20, the AND 21 element, the control trigger 22, the character register 23. The synchronizer (Fig. 4) contains a multivibrator 24,

DK-trigger 25, elements And 26 and 27.

Synchronizer 3 is a standard unit of ultrasound equipment and can be made in the form of a two-phase square-wave generator. From the first and second output signalg.alg removed, out of phase from each other. The magnitude of the displacement is determined by the maximum propagation time of ultrasound.

to defect and back. The storage unit 6 and memory blocks 12 and 18 can be implemented on the basis of standard semiconductor memory modules, for example, microcircuits

j K134RU6. A typical functional block diagram of a memory block based on these chips is given in the monograph. The storage unit is used as the RAM of the control unit 8.

Interface block 7 can be

made on the basis of the standard K580IK55 programmable parallel interface configured in the gated or potential information output mode, while setting up the K580IK55 interface and controlling its operation was carried out by the control unit 8 according to its

the program. The control unit 8 acts as a central processor in the system; it must have a developed command system, in particular, the arithmetic and logical operations of the command for exchanging RAM with indirect addressing, I / O commands. A specific feature of the control unit 8 is that it has a direct memory access mode. This mode allows you to stop the microprocessor and disconnect it from the address and data buses for the duration of accessing the external memory of the peripheral device. In connection with this, the microprocessor has a Capture input and an output Capture Confirmation. The output signal of the Capture initiates a microprocessor stop, which, after completing the next machine cycle, stops and sets a signal at the output of the Capture Confirmation. This signal starts the direct memory access mode. The control unit 8 can be implemented on the basis of the standard K580IK80 microprocessor.

The system works as follows.

The inputs of the angular sensor 4 and the linear coordinate sensor 5 receive pulses from the output of a scanning device (not shown), which are transformed last into the current coordinates of the converter when it moves along the monitored product into digital form with the resolution if the coordinate of the angular displacement control of flat products) and with a resolution of 4 along the coordinate of the longitudinal movement.

The cycle of operation of each measuring channel upon detection of a defect consists of starting the generator of probe pulses of a flaw detector (not shown), receiving a response echo pulse, converting its amplitude into a digital code and writing this code to BZU 2. Consider the execution of this cycle in more detail. Peak 15 detector remembers the maximum amplitude of the echo pulse. The last one is accompanied by a logical signal Defect arriving at A1Sh 16 from the flaw detector. The synchronizer 3 selects a signal with a certain delay in relation to the probe pulse. The start of the conversion, at which the output register of the A / D converter is reset to zero, and if there is a signal, the Defect starts the A / D converter 16. After the conversion of the A / D converter 16 of each measuring unit 1, a pulse is output. the corresponding input OR 19 of the counting trigger 13 and zeroing peak detector 15. The pulse decay at the output of this element OR transfers the counting trigger 13 to the state corresponding to the data input with itelnyh blocks 1 and sensors 4 and 5, the angular and linear coordinates, respectively. In this state, the counting trigger 13 exposes the Capture signal, according to which the control block 8 stops the program and stops, giving a response Accept Capture Confirmation signal.

The signal Confirmation of the capture of the contents of the counter 10 address is increased by one, and the multiplex-p 11 delivers to the input of the memory block 12 the contents of the sensors 5 and 4 linear and angular coordinates together with the address coming from the output of the counter 10 address. Simultaneously, the multiplexers 17 connect the corresponding outputs of the ADC 16 and the address coming from the output of the counter 10 to the address of the single blocks of the single-vibrator 14, receiving a Confirmation signal with a small delay determined by the described switches, and generates a pulse. 18 memory from the corresponding measuring unit 1 amplitude codes of echo pulses. This signal is recorded in memory unit 12 by recording the coordinates (angular and linear) of the input echo pulses. In addition, the Record signal pushes into the input of the counting trigger and returns to its initial state on the falling edge, which leads to the disappearance of the Capture signal.

If the measuring unit 1 did not receive an echo pulse in this cycle, then the ADC 16 of this measuring unit 1 will not turn on and, from its output, zeroes will be written into the corresponding memory block 18.

Thus, the codes of the amplitudes of the echo pulses or zero codes are stored at the same addresses in the memory blocks 18 — received from the corresponding measuring channels in one cycle. The coordinates (angular and linear) related to these echo pulse codes 5 is stored at the same address in memory block 12.

In the absence of a signal, the Capture control unit 8 exchanges information with memory blocks 12 and 18 through the respective multiplexers 11 and 17. The operation of control block 8 consists in sequentially polling the cells of memory blocks 12 and 18 in order to form a multidimensional array of signs according to which defect metrics in relation to the mathematical models stored as programs in the storage unit 6. After processing all the information relating to one product, the recovery unit 8 through the interface unit 7 you Odita control results in .vide technical passport controlled device 9 on the article output.

Let us consider in more detail the process of forming multidimensional arrays of attributes. The coordinates of the echo pulses stored in memory unit 12 refer to the first measuring channel. For the remaining channels, the corresponding coordinates are obtained by correcting the stored coordinates produced by the control unit 8. The latter sticks together echo pulses on each channel that have equal or different coordinates and identical linear coordinates. Thus, a bundle of echo pulse codes is formed, which characterizes one section of defect. Then the microprocessor extracts the value for the mathematical model of defects, the signs of this bundle of echo pulses, for example, the number of pulses in a bundle that exceed some levels of discrimination, the rate of rise and fall amplitudes, the maximum amplitude of the echo pulse, etc.

In addition, the coordinates of the beginning and end of the echo-impulse stack are recorded. Packs in which the angular coordinates intersect, and the linear coordinates differ by one, are referred to by the control unit 8 as a single defect. As a result, for any defect, control unit 8 generates an array of significant features for each channel. If necessary, further information compression is performed taking into account the obtained values for each channel,

In the initial state, all memory blocks are zero, therefore the occurrence of a nonzero value in the cell

12 is considered by the control unit 8 as the arrival of the next portion of the echo pulses. Block 8, the manager remembers the contents of this cell in the storage unit 6 and treats it as a coordinate. Further, this cell of the memory unit 12 is zeroed. Similarly, the non-zero block cell contents

13 for tees is considered a block

8 control as the amplitude code of the echo pulses, which are written during the gluing of the packs, an echo pulse in the storage unit 6. After this rewriting, the cells of the memory block 18 are reset by the microprocessor. In the described manner, control unit 8 distinguishes information recorded in memory blocks 12 and 18.

The system for automatic defectometry practically does not impose any restrictions on the speed of input of flaw detection information, since the input time is determined by the speed of the ADC and the memory block. The number of channels does not affect the speed of information input. Microprocessor control unit c. this system stops its operation only for the time of access to the memory when entering a portion of echo pulses. All the rest of the time, it is occupied by the flaw detection data processing described, which is also essential for evaluating system performance. The advantage of the system is its regular structure and ease of data entry into memory blocks 12 and 18, which are controlled by the counting trigger 13 in the system, and the one-shot

14 and counter 10 addresses. Regularity of the structure and ease of organizing data entry leads to high adaptability in the manufacture, configuration and operation of the system. When using a ready-made ADC of the type F7077-1 with the conversion time of 3 MKS. And the installation time of the input voltage A μs, and as a memory of the module 134RU6 with the time of circulation 0.5 μs, we obtain, taking into account the delay in issuing the signal which is for the microprocessor K - 580 IC80 a maximum of 2.5 µs, the period of the following echo pulses is no more than 10 µs. This means that the frequency of the zones of the dying pulses when using this system can be close to 100 KHz. The volume of blocks 1.2 and 18 of memory is determined by the statistical characteristics of the flow of flaw detection signals on all channels and the time of their processing. It is very difficult to obtain an exact value of the volume of the blocks 12 and 18 of the memory, since the flow of the echo pulses and the processing time do not obey the Poisson law. However, it is possible to obtain an upper estimate of the required volume of memory blocks 12 and 1B at a given probability of their overflow, if it is considered that the input stream of echo pulses and their processing time are subject to the Poisson bill. .. We will perform such a calculation for a sufficiently high intensity of the input stream of echo pulses from four channels. L 3960 pulses / s. In this case, the average processing time of a single echo pulse is selected from the ratio of Take TC |) 250 µs. The average occupancy of the microprocessor will be prO, 99, i.e. system on. It is in very heavy mode. The probability of overflow of blocks 12 and 18 of memory to 10 is set. In this case, the value of K is equal to the number of cells in block 12 of memory, which is associated with the probability of overflow by the following ratio H) p. From where we get 915 915 cells. This means that the volume of BZU 2 for each channel can be selected as little as 1 K words, even if the system load is close to the maximum. From the above, it follows that the control speed when using the proposed system is practically independent of the frequency of the probing pulses and the number of channels, the quite acceptable amount of LPD makes it possible to obtain a meager probability of loss of the flaw-detection signal. The speed of control in the proposed system is limited only by the speed of the microprocessor control unit and the complexity of the processing program. The permissible control speed can be found from the relation. . . , where L fff) is the intensity of the total input stream of the ex9 pulse as a function of the speed of control, T {.p is the average processing time of a continuous echo pulse. The use of the proposed system makes it possible to monitor products with greater speed and at the same time with high control accuracy by entering information about the whole burst of impr pulses and joint processing of these signals received from different channels.

Tire

Otter

20

addresses

I

f

Record

Data bus

I

FIG. J

W (/ g

Resolution

nevoffrn

22

Reception of reception from

gcmpoucmSa

23

Claims (1)

SYSTEM FOR AUTOMATIC DEFECTOMETRY containing linear and angular coordinate sensors, a measuring unit, the first input of which is connected to the first output of the synchronizer, the second and third inputs of which are connected to the first and second outputs of the system for automatic flaw detection, the input of which is connected to the second output of the synchronizer , a control unit, the address bus of which is connected to the first input of the interface and storage units, an output device, the input of which is connected to the output of the interface unit, the data bus of the unit is connected to the second inputs of the interface and storage units and to the output of the storage unit _? the third input of which is connected to the Reading output of the control unit, characterized in that, in order to increase the reliability of control and signal processing speed by the system, an address counter, a counting trigger, an .OR element, a one-shot, the first multiplexer, the first coordinate memory block are introduced into it a second multiplexer and a second memory block are introduced into each measuring channel, the output of which is connected to the data bus of the control unit, the first input to the output of the second multiplexer, and the second input to the first input of the first coordinate memory block yes channel, to the second inputs of the memory blocks of the remaining measuring channels and to the third input of the storage unit, the fourth input of which is connected to _the third output of the interface unit, ® with the output Record of the control unit and with the first input Record of the first and second multiplexers, output Capture confirmation the control unit is connected to the control inputs of the first and second multiplexers, through the address counter it is connected to the first input of the address of the first and second multiplexers, and through a single-shot to the first input of the OR element and to second inputs Recording of the first and second multiplexers, the first output of the measuring unit is connected to the first data input of the second multiplexer, the second output is with the second input of the OR element, the output of which is connected through the counting trigger to the input Capture of the control unit, the output of the first multiplexer is connected to the second input of the first memory unit , the output of which is connected to the data bus of the control unit, with the second data inputs of the first and second multiplexers and with the Reading input of the control unit, you SU. „. 1190253 linear and angular sensors the coordinates are connected to the second data input of the first multiplexer, the address bus of the control unit is connected to the second address inputs of the first and second multiplexers.