A data collection and processing device and method based on amplitude monitoring

Abstract

The present invention discloses a data collection and processing device and method based on amplitude monitoring. The device includes a vibration sensor, a peak detection circuit, a parallel ADC, and a CPLD. The vibration sensor is used to convert displacement into a voltage signal for output. The peak detection circuit is used to detect an effective peak signal. The parallel ADC is used to sample at the rising edge of the CLK and output data in parallel after a certain delay. The CPLD is used to output a control signal for controlling ADC sampling according to the detected peak signal, perform counting processing on the sampling data output by the parallel ADC, and output the working state information and the counting result in a serial manner. The present invention can discriminate the collected data. At the same time, abnormal data can also be locally stored at the sensor end, which is convenient for on-site fault analysis and maintenance, and can also be used as a reference basis to dynamically correct various monitoring indicators as the equipment ages and wears.

Description

Technical Field

[0001] The present invention belongs to the technical field of multi-industrial field control, and more specifically, relates to a data collection and processing device and method based on amplitude monitoring. Background Art

[0002] In the field of industrial control, it is often necessary to monitor the amplitudes of various devices such as motors and bearings. Usually, sensors are used to detect the vibrations of the devices and output voltage signals. Subsequently, an ADC is used to convert the voltage signals into data, which is then transmitted back to the control host through an industrial bus to achieve the monitoring of the vibration conditions of the devices.

[0003] The above monitoring method can monitor the vibration conditions of the devices in real time, but it still has certain limitations: the result output by the ADC is the value obtained after the amplitude sampled by the sensor at a specific moment is converted. The monitoring system will record the output results in sequence according to the sampling frequency of the ADC. The duration of this working process of the device is generally longer than the sampling period of the ADC, and it is difficult for the device hardware to record all the amplitude data sampled during the working period at a low cost. Under the limitation of cost, the device cannot continuously monitor the amplitude conditions during a working process for a long time and perform data statistics, and the data collected by itself is often insufficient to support actual maintenance and repair work. Summary of the Invention

[0004] Aiming at the defects of the prior art, the purpose of the present invention is to provide a data collection and processing device and method based on vibration monitoring, aiming to achieve long-term data collection of the vibration conditions of the device at a low cost; to solve the problem that the data monitored by the device under the limitation of cost is insufficient to support fault analysis and equipment maintenance.

[0005] The present invention provides a data collection and processing device based on amplitude monitoring, including a vibration sensor, a peak detection circuit, a parallel ADC, and a CPLD; the vibration sensor is used to convert the monitored displacement into a voltage signal for output; the input end of the peak detection circuit is connected to the output end of the vibration sensor, and is used to detect an effective peak signal; the input end of the parallel ADC is connected to the first output end of the peak detection circuit, and the clock signal trigger end of the parallel ADC is connected to the sampling output end of the CPLD. The parallel ADC is used to sample at the rising edge of the CLK and output data in parallel after a certain delay; the first input end of the CPLD is connected to the second output end of the peak detection circuit, and the second input end of the CPLD is connected to the output end of the parallel ADC. The CPLD is used to output a control signal for controlling the ADC sampling according to the detected peak signal, perform counting processing on the sampling data output by the parallel ADC, and output the working state information and the counting result in a serial manner.

[0006] Furthermore, the CPLD includes: a sampling control module, a counting processing module, and a data output module; the sampling control module is used to control the sampling clock of the ADC when receiving the peak signal, and cooperate with the sampling timing to control the ADC output and read the sampling result after a certain time delay; the counting processing module is used to perform counting processing on the ADC result after reading it and generate a counting spectrum and working status information; the data output module is used to transmit the working status information and the counting result to an external control host in a serial manner.

[0007] Furthermore, it further includes: a storage device, connected to the CPLD, used to store configuration parameters, pre-store a standard counting spectrum, and store the abnormal counting spectra generated during the detection process.

[0008] The device provided by the present invention can make statistics on all output values of the vibration sensor within a period of time, can reflect the overall vibration intensity of the device within a period of time, can provide more sufficient and effective data for the monitoring of the device vibration intensity, and at the same time, the present invention provides an effective processing and analysis method, which can distinguish the collected data, and the abnormal data can also be saved locally at the sensor end, facilitating on-site fault analysis and maintenance, and can also be used as a reference basis to dynamically correct various monitoring indicators as the device ages and wears.

[0009] The present invention also provides a data processing method based on amplitude monitoring, including the following steps:

[0010] S1: Predesign a numerical value;

[0011] S2: Clear the counting array I[0:M], and obtain the reference counting array Ir[0:M] corresponding to the mode;

[0012] S3: Set the sampling clock frequency of the output according to the preset parameters, and read the sampling data according to the characteristics and delay of the ADC;

[0013] S4: For an ADC with a precision of N, the sampled data it outputs is an N-bit binary number D[0:N];

[0014] S5: Use an array I[0:M] with a length of M + 1 to count the sampling results. According to each ADC sampling result D[0:N], the value of the corresponding element I[m] in I[0:M] is incremented by 1;

[0015] S6: Continuously repeat step S5 until the number of sampling times reaches the predesigned numerical value. At this time, obtain the final counting result I[0:M], and form a two-dimensional spectrum with the ADC output result as the abscissa and the corresponding result quantity as the ordinate according to the final counting array I[0:M];

[0016] S7: Compare I[0:M] with the preset Ir[0:M]. If there is no obvious difference between the two, proceed to step S8; if there is an obvious difference between the two, proceed to step S9;

[0017] S8: Send a normal working message to the host computer and send the entire array of I[0:M] or its average value to the external control host;

[0018] S9: Generate corresponding exception information according to the difference and send the result data to the external control host.

[0019] Wherein, after step S8 or S9, there is further step S10: Store the abnormal data in the local memory as a basis for on-site fault analysis and maintenance or, considering equipment aging, for correcting the reference array Ir[0:M].

[0020] Furthermore, in step S1, determine the preset value according to the equipment characteristics, actual working conditions, monitoring time, and storage space, and it is advisable that the count of no single channel exceeds 70% of the maximum allowable count per channel.

[0021] Furthermore, perform segmented summation on the two-dimensional spectral shape in step S6 and then make a judgment, and send a fault alarm signal in real time when an abnormal spectral shape is found.

[0022] Furthermore, the principles for judging the spectral shape include:

[0023] (1) When the channel address where the summation result appears is different from the reference spectrum, it indicates that the distribution of the judged count has shifted and the amplitude has deviated from the normal value;

[0024] (2) When the summation result of each segment exceeds a certain value, it indicates that the distribution of the count is too scattered, the amplitude fluctuates greatly, and the equipment is working unstably;

[0025] (3) When there are multiple extreme values in the summation result, it indicates that the count has abnormal aggregation, reflecting abnormal vibration of the equipment.

[0026] Through the above technical solution conceived by the present invention, compared with the prior art, the ADC no longer samples the voltage output by the sensor at a fixed frequency. Instead, the circuit monitors the signal peak and quantifies and records the peak signal. At the same time, the memory no longer stores the numerical results obtained by each ADC. Instead, all possible output results of the ADC are listed in advance, and these results are corresponding to the addresses of the storage space. Each time the ADC outputs a value, the count stored in these addresses will be incremented by one. In this way, when a value appears multiple times, originally multiple addresses were required to store each value, but after using the present invention, only one address is needed to record the number of times it appears, greatly compressing the data volume. Although the time sequence of data generation cannot be recorded, the data volume recorded is greatly increased with little requirement for hardware, providing more powerful data support for maintenance and repair. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 is a schematic block diagram of a data collection and processing device based on vibration monitoring provided by an embodiment of the present invention;

[0028] Figure 2 is a schematic flowchart of the implementation of a data processing method based on amplitude monitoring provided by an embodiment of the present invention;

[0029] Figure 3 is a schematic diagram of counting accumulation into a spectrum. Among them, (a) is a schematic diagram of collecting and quantifying the peak signal of the sensor, and (b) is a schematic diagram of corresponding the quantization result to the storage space and incrementing the count in the corresponding address;

[0030] Figure 4 is a schematic diagram of an abnormal spectrum provided by an embodiment of the present invention. Among them, (a) is the spectrum under normal conditions, (b) is the spectrum with an offset, mostly caused by aging or insufficient lubrication, (c) is the spectrum with non-concentrated counting, generally due to the device being unable to stably maintain a stable power, and (d) is the spectrum with abnormal aggregation, generally due to abnormal resonance occurring during the process of power increase or decrease of the device;

[0031] Figure 5 is a schematic flowchart of the implementation of an abnormal preliminary detection method in a data processing method based on amplitude monitoring provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0032] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0033] The present invention proposes a design scheme for enhancing the sensor data acquisition and processing capabilities by using a programmable device. During a period of time, multiple peak ADC samplings are performed on the vibration signals output by the sensor. At the same time, the collected data is accumulated and counted in channels according to the ADC conversion results, and finally the count statistics of the vibration peaks of the device during this period are formed. The data is stored, processed, and uploaded in the form of a count spectrum, providing sufficient data and a reasonable evaluation method for device status monitoring, and providing sufficient support for fault analysis and equipment maintenance.

[0034] The present invention uses a CPLD to enhance the information acquisition and processing capabilities. During the monitoring period, peak detection is continuously performed on the input vibration analog signal. An ADC is used to sample each detected peak, the sampling results are statistically analyzed, and the number of occurrences of different sampling values is counted. Finally, a sampling count spectrum of the vibration peaks of the device during this period is formed, and an effective fault judgment method is provided, providing more sufficient data support for the monitoring and analysis of device vibration.

[0035] As Figure 1 shown, the data collection and processing device based on vibration monitoring provided by the embodiment of the present invention includes a vibration sensor 1, a peak detection circuit 2, a parallel ADC (analog-to-digital converter) 3, and a CPLD (complex programmable logic device) 4; the vibration sensor 1 is used to convert displacement into a voltage signal for output; the peak detection circuit 2 is used to output a peak signal by the CPLD after detecting a peak. This circuit needs to have a sample and hold function to hold the detected peak signal for a period of time for CPLD to control sampling; the parallel ADC 3 is used to sample at the rising edge of the CLK and output data in parallel after a certain delay; the CPLD 4 is used for sampling control, data processing, uploading, and other tasks.

[0036] Among them, the analog signal output by the vibration sensor 1 is sent to the peak detection circuit 2. After detecting a peak, a peak signal is output to the CPLD. At the same time, the sampling circuit holds the peak signal for a period of time. After receiving the peak signal, the CPLD controls the ADC to perform sampling. The parallel output port D[0:N] of the ADC is connected to the standard IO of the CPLD. Inside the CPLD, a one-dimensional array I[0:M] with a corresponding width is set according to the accuracy of the AD chip to count the conversion results. Each element in the array corresponds one-to-one with each possible output D[0:N] of the ADC. After each AD conversion, the CPLD reads the conversion result D[0:N], and the corresponding I[m] count is incremented by 1.

[0037] In an embodiment of the present invention, the CPLD4 includes a sampling control module 6, a counting and processing module 7, and a data output module 8. The sampling control module 6 is used to control the sampling clock of the ADC when receiving the peak signal, and after a certain time delay in cooperation with the sampling timing, control the ADC to output and read the sampling result. The counting and processing module 7 is used to perform counting processing on the ADC result after reading it, and generate a counting spectrum and working status information. The data output module 8 is used to transmit the working status information and the counting result to the control host in a serial manner.

[0038] In an embodiment of the present invention, it further includes a storage device 5, which is connected to the CPLD4. The storage device 5 is used to store some configuration parameters, and can also pre-store a standard counting spectrum so that the CPLD can compare the real-time data with it, and can also store the abnormal counting spectrum generated during the detection process.

[0039] The device provided by the embodiment of the present invention can make statistics on all output values of the vibration sensor during a relatively long period when the device is working under specific working conditions. This time is jointly determined by the device characteristics and the actual working requirements, and can reflect the overall amplitude situation of the device during this period. It can provide more sufficient and effective data for monitoring the vibration intensity of the device. At the same time, the present invention provides an effective processing and analysis method, which can discriminate the collected data, and the abnormal data can also be saved locally at the sensor end, facilitating on-site fault analysis and maintenance, and can also be used as a reference basis to dynamically correct various monitoring indicators as the device ages and wears.

[0040] The present invention uses the CPLD to realize the acquisition and collation of equipment working condition monitoring data, judges whether the equipment working state is normal by comparing with the standard working condition data, gives an alarm in time for abnormal states, and saves and records the monitoring data.

[0041] As Figure 2 shown, the present invention also provides a data processing method based on vibration monitoring, which specifically includes the following steps:

[0042] Step S1: Predesign a numerical value; specifically, the pre-designed numerical value can be determined according to the device characteristics, actual working conditions, monitoring time, and storage space. Taking the normal situation as an example, it is appropriate that the count of no channel exceeds 70% of the maximum count allowed for each channel.

[0043] Step S2: Clear the counting array I[0:M], and at the same time read the reference counting array Ir[0:M] of the corresponding mode from the external memory for comparative analysis with the final counting result.

[0044] Step S3: The sampling control module controls the ADC to perform sampling, sets the output sampling clock frequency according to the preset parameters, and reads the sampling data according to the characteristics of the ADC with a delay.

[0045] Step S4: For an ADC with a precision of N, the sampled data it outputs is an N-bit binary number D[0:N].

[0046] Step S5: Use an array I[0:M] with a length of M + 1 to count the sampling results. According to each ADC sampling result D[0:N], the value of the corresponding element I[m] in I[0:M] is incremented by 1.

[0047] As Figure 3 shown, in (a), the abscissa is time. Each time the circuit detects a peak signal, the CPLD controls the ADC to sample. The ordinate is the voltage value of the sampled signal, and the voltage sampling result is the binary number D[0:N]. In (b), the abscissa is the channel address, which increases from 0 to M in increments of 1, where M = 2 N+1 - 1, so that each element I[m] in the array I[0:M] can correspond one-to-one with each result of the binary number D[0:N]. The ordinate is the cumulative count. The cumulative count of each channel is I[m]. Each time the ADC generates a sampling result D[0:N], it will map to the corresponding channel address m on the abscissa, and the count of I[m] corresponding to the channel address m is incremented by 1. Eventually, a histogram will be formed on Figure 3 (b). The abscissa of the histogram is the output result of the ADC, which actually represents the amplitude of the device, and the ordinate of the histogram is the number of vibrations with the same amplitude occurring during the monitoring period.

[0048] Step S6: Continuously repeat Step S5 until the number of sampling times reaches the preset value. At this time, the final count result I[0:M] is obtained. This array can form a two-dimensional spectrum with the ADC output result as the abscissa and the corresponding result quantity as the ordinate.

[0049] Step S7: Compare I[0:M] with the preset Ir[0:M].

[0050] Step S8: If there is no obvious difference, send a normal working message to the host computer, and at the same time, the entire array I[0:M] or its average value can be sent to the control host.

[0051] Step S9: If there is an obvious difference, generate corresponding abnormal information according to the difference, such as Figure 4 shown, and send it to the control host along with the result data.

[0052] As Figure 4 shown, under normal circumstances, when the device is working properly, the amplitude is stably maintained within a certain range, and the ADC sampling results are relatively concentrated on the spectrum to form a shape like Figure 4The spikes shown in (a). In addition to directly analyzing the spectral shape, a program can also be designed to roughly judge after segmentally summing the spectral counts. Once an abnormal spectral shape is found, a fault alarm signal can be sent in real time. The specific process is as follows Figure 5 as shown:

[0053] (1) First, find the maximum value among all the summation results. Generally, there is only one maximum value. When there are multiple maximum values, it can be determined that there is an abnormal aggregation in the spectral shape. See Figure 4 (d) for details. It shows that within the monitoring period, the amplitude of the equipment in a certain period is inconsistent with the normal amplitude. Generally, it is due to abnormal resonance when the power of the equipment increases or decreases;

[0054] (2) When the segment where the maximum value is located is inconsistent with the segment position of the preset spectral maximum value, it indicates that the spectral shape has shifted. See Figure 4 (b) for details. The working amplitude of the equipment deviates from the normal value, generally shifting to the right, indicating that the vibration of the equipment is more intense than the normal state during the monitoring period, mostly caused by aging or insufficient lubrication;

[0055] (3) Set a threshold according to the instrument characteristics and actual usage, and record the number of segments where the summation result exceeds this threshold. If the number of segments exceeds the number of segments obtained by performing the same processing on the preset spectrum, it means that the counting is not concentrated. See Figure 4 (c) for details. It shows that there are large fluctuations in the amplitude of the equipment during operation, generally caused by the equipment's inability to stably maintain a stable power.

[0056] Step S10: Store the abnormal data in the local memory as the basis for on-site fault analysis and maintenance, or to correct the reference array Ir[0:M] considering equipment aging.

[0057] In the embodiments of the present invention, in the scenario of working state monitoring, the specific methods for collecting, counting, and forming spectra of the monitoring data are as follows:

[0058] (1) Set a two-dimensional array corresponding to the ADC accuracy. For each output value of the ADC, the value of each element in the two-dimensional array is the count of the corresponding output value of the ADC;

[0059] (2) Every time a peak signal is detected, a sampling is performed, and +1 is added to the corresponding element in the array according to the sampling result;

[0060] (3) Set a monitoring time according to the equipment characteristics, actual working conditions, and storage space. During this period, each element in the array starts from 0 and accumulates according to the sampling result, finally forming a counting spectrum. This counting spectrum is a statistic of the peak signals output by the vibration sensor during this period, reflecting the overall vibration intensity of the equipment during this period.

[0061] In the embodiment of the present invention, the method for directly judging abnormal working conditions at the sensor end is specifically as follows: segmentally sum the obtained counting spectrum, and the general situation of the counting distribution can be obtained, which is convenient for roughly judging the spectrum shape:

[0062] (1) When the channel address where the summation result appears is different from the reference spectrum, it indicates that the distribution of the judged count has shifted and the amplitude has deviated from the normal value;

[0063] (2) When the summation result of each segment exceeds a certain value, it indicates that the distribution of the count is too scattered, the amplitude fluctuates greatly, and the equipment works unstably;

[0064] (3) When there are multiple extreme values in the summation result, it indicates that the count has abnormal aggregation, reflecting abnormal vibration of the equipment.

[0065] In the present invention, the ADC no longer samples the voltage output by the sensor at a fixed frequency, but monitors the signal peak through a circuit and quantifies and records the peak signal. At the same time, the memory no longer stores the numerical result obtained by each ADC. Instead, all possible output results of the ADC are listed in advance, and these results are corresponded to the addresses of the storage space. Each time the ADC outputs a value, the count stored in these addresses will be incremented by one. In this way, when a value appears multiple times, originally multiple addresses were required to store each value, but after using the present invention, only one address is needed to record the number of times it appears, greatly compressing the data volume. Although the time sequence of data generation cannot be recorded, the data volume recorded is greatly increased under the condition of little requirement for hardware, providing more powerful data support for maintenance and repair.

[0066] Those skilled in the art can easily understand that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A data processing method based on amplitude monitoring, characterized in that, it includes the following steps: S1: Predesign a value; S2: Clear the counting array I[0:M], and obtain the reference counting array Ir[0:M] corresponding to the mode; S3: Set the sampling clock frequency of the output according to the preset parameters, and delay the reading of the sampling data according to the characteristics of the ADC; the sampling data output by the ADC is the amplitude when the effective peak signal obtained by vibration monitoring; S4: For an ADC with a precision of N, the sampling data it outputs is an N-bit binary number D[0:N]; S5: Use an array I[0:M] with a length of M + 1 to count the sampling results. According to each ADC sampling result D[0:N], the value of the corresponding element I[m] in I[0:M] is incremented by 1; S6: Continuously repeat step S5 until the number of samplings reaches the predesigned value. At this time, obtain the final counting result I[0:M], and form a two-dimensional spectrum with the ADC output result as the abscissa and the corresponding result quantity as the ordinate according to the final counting array I[0:M]; S7: Compare I[0:M] with the preset Ir[0:M]. If there is no obvious difference between the two, go to step S8; if there is an obvious difference between the two, go to step S9; S8: Send a normal working message to the host computer, and send the entire array of I[0:M] or its average value to the external control host; S9: Generate corresponding abnormal information according to the difference, and send the result data to the external control host; Perform segmented summation on the two-dimensional spectrum in step S6 and then make a judgment, and send a fault alarm signal in real time when an abnormal spectrum is found; The principles for judging the spectrum include: (1) When the channel address where the summation result appears is different from the reference spectrum, it indicates that the distribution of the judged count has shifted, and the amplitude has deviated from the normal value; (2) When the summation result of each segment exceeds a certain value, it indicates that the distribution of the count is too scattered, the amplitude fluctuates greatly, and the device works unstably; (3) When there are multiple extreme values in the summation result, it indicates that the count has an abnormal aggregation, reflecting that the device has abnormal vibration.

2. The data processing method according to claim 1, characterized in that, after step S8 or S9, it further includes step S10: Store the abnormal data in the local memory, as the basis for on-site fault analysis and maintenance or for correcting the reference array Ir[0:M] considering the equipment aging.

3. The data processing method according to claim 1 or 2, characterized in that, in step S1, determine the predesigned value according to the equipment characteristics, actual working conditions, monitoring time and storage space, and it is advisable that the count of no single channel exceeds 70% of the maximum allowable count per channel.

Images