Identification method and system based on error signals before and after startup and shutdown of equipment

Abstract

The invention discloses an identification method and system based on error signals before and after starting and stopping of equipment, and belongs to the technical field of state monitoring of rotating equipment. The method comprises the following steps: acquiring equipment measurement point data, presetting a start-stop threshold and an alarm threshold, and if the start-stop threshold is not preset, processing the measurement point data to obtain the start-stop threshold; then the start-stop state of the equipment is judged according to the start-stop threshold value; the method comprises the steps of classifying all data, calculating a probability density function and a differential entropy, finally obtaining an index factor of start-stop abnormal data, then identifying the start-stop abnormal data based on a sliding window, marking the start-stop abnormal data, retrieving whether the data exceeding a preset alarm threshold is abnormal data, and if yes, canceling the alarm. Start-stop state identification is carried out based on the equipment vibration condition, then false signal discrimination is further carried out on data before and after start-stop, false alarms caused by working condition fluctuation before and after start-stop of the equipment are reduced, and the alarm accuracy of a monitoring system can be improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of rotating equipment status monitoring, and in particular relates to a method and system for identifying error signals before and after equipment startup and shutdown. Background Art

[0002] The status monitoring of rotating equipment is an important part of equipment operation and maintenance applications. Accurately judging the status of the equipment and uploading the operating status of the equipment to the equipment monitoring system can provide accurate maintenance guidance for on-site maintenance personnel, strengthen the company's management level of equipment, and reduce the losses caused by over-maintenance and under-maintenance of equipment.

[0003] Due to fluctuations in operating conditions before and after the start-up and shutdown of rotating equipment, the vibration intensity of the equipment may fluctuate greatly, which further causes the equipment to issue an alarm. However, the actual analysis of such alarm data is of little significance and will provide wrong guidance to on-site operation and maintenance personnel. In addition, the existing technology only performs simple start-up and shutdown judgments based on the vibration intensity of the equipment or identifies periodic false alarms. The actual start-up and shutdown time of the unit is not periodic. Therefore, a false signal identification method based on the start-up and shutdown status of the equipment is proposed. Summary of the invention

[0004] In view of the deficiencies in the prior art, the purpose of the present invention is to provide a method and system for identifying error signals before and after equipment startup and shutdown, which solves the problems in the prior art.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A method for identifying false signals before and after equipment startup and shutdown, comprising the following steps:

[0007] S1, obtain equipment measurement point information, preset start and stop thresholds, and preset alarm thresholds;

[0008] S2, obtain the historical acceleration and velocity effective values ​​of all vibration sensor measurement points of the equipment;

[0009] S3, determine whether a start / stop threshold is preset, if the start / stop threshold is preset, proceed to S7, if not, proceed to S4;

[0010] S4, there are differences in vibration at different measuring points. The historical data of each measuring point of the equipment are sorted respectively, and the ten points before and after are eliminated. Then the vibration fluctuation rate of the remaining historical data of each measuring point is used as the evaluation index for selecting analysis data;

[0011] S5, selecting the data of the measuring point with the highest vibration fluctuation rate as analysis data to perform start-stop identification;

[0012] S6, classify based on the historical data of the measuring point, use Euclidean distance to calculate the distance sorting to confirm the category, and select the smallest category by voting to obtain the characteristic value as the start-stop threshold;

[0013] S7, using the start / stop threshold to judge the start / stop status of the historical data of each measuring point;

[0014] S8, divide the historical data of each measuring point into three groups: shutdown data, start-stop data, and stable operation data;

[0015] S9, calculate the probability density of the three sets of data using Gaussian kernel;

[0016] S10, calculating differential entropy according to probability density;

[0017] S11, calculating the index factor of the abnormal start and shutdown data according to the differential entropy;

[0018] S12, based on the index factor of the abnormal start and stop data, using a sliding window to identify the abnormal start and stop data;

[0019] S13, marking the abnormal start and shutdown data, and searching whether the data exceeding the preset alarm threshold is abnormal data, and if so, canceling the alarm.

[0020] Furthermore, the Euclidean distance formula:

[0021]

[0022] Among them, x i and x j are two data points respectively, k represents the feature dimension, and n is the total feature dimension.

[0023] Furthermore, the start and stop data are data within a time range T before and after the stop data.

[0024] Furthermore, the probability density calculation formula is:

[0025]

[0026] Where f(x) is the probability density, m is the array size, and x i is each data point, h is the bandwidth parameter, and K(μ) is the Gaussian kernel.

[0027] Furthermore, the differential entropy calculation formula is:

[0028]

[0029] Where h(x) is the differential entropy function;

[0030] The calculation formula of the index factor fac of the start-stop abnormal data is:

[0031] fac=(a1×h1+a2×h2+a3×h3) / 3

[0032] Where a1, a2, a3 are coefficients, and h1, h2, h3 are differential entropies.

[0033] Furthermore, the process of identifying the abnormal start-stop data is as follows:

[0034] Set the window length Win, step length Len, index factor fac of abnormal start and shutdown data, calculate the window data mean, window data standard deviation std, and determine whether the current vibration effective value rms satisfies: rms≤mean±std*fac; if satisfied, the data is normal; if not satisfied, it is abnormal data.

[0035] Furthermore, the effective value rms calculation formula is:

[0036] The mean value calculation formula is:

[0037] The standard deviation calculation formula is:

[0038] Where N is the window size.

[0039] A system for identifying false signals before and after equipment startup and shutdown, comprising:

[0040] Preset threshold acquisition module: obtain equipment measurement point information, preset start and stop thresholds, and preset alarm thresholds;

[0041] Measuring point data acquisition module: obtains the historical acceleration and velocity effective values ​​of all vibration sensor measuring points of the equipment;

[0042] Threshold preset judgment module: judge whether the start / stop threshold is preset, if the start / stop threshold is preset, enter the start / stop state judgment module, if the start / stop threshold is not preset, enter the measurement point data processing module;

[0043] Measuring point data processing module: Different measuring points have different vibrations. The historical data of each measuring point of the equipment are sorted respectively, and the ten points before and after are eliminated. The vibration fluctuation rate of the remaining historical data of each measuring point is used as the evaluation index for selecting analysis data;

[0044] Measuring point data screening module: selects the measuring point data with the highest vibration fluctuation rate as analysis data for start-stop identification;

[0045] Start-stop threshold setting module: Classify based on the historical data of the measuring point, use Euclidean distance calculation to sort and confirm the category, and select the smallest category by voting to obtain the characteristic value as the start-stop threshold;

[0046] Start-stop status judgment module: Use the start-stop threshold to judge the start-stop status of each measuring point historical data;

[0047] Measuring point data classification module: divides the historical data of each measuring point into three groups: shutdown data, start-stop data, and stable operation data;

[0048] Probability density function calculation module: calculate the probability density of the three sets of data using Gaussian kernel;

[0049] Differential entropy calculation module: calculates differential entropy according to probability density;

[0050] Index factor calculation module: calculates the index factor of abnormal start and shutdown data based on differential entropy;

[0051] Abnormal data identification module: Based on the indicator factors of abnormal start and stop data, the abnormal start and stop data is identified using a sliding window;

[0052] And, the alarm judgment module: marks the abnormal start and shutdown data, and searches whether the data exceeding the preset alarm threshold is abnormal data. If so, cancel the alarm.

[0053] A computer storage medium stores a readable program, and when the program is running, the above-mentioned method for identifying error signals before and after the equipment is started and stopped can be executed.

[0054] An electronic device, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other through the communication bus;

[0055] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the above-mentioned method for identifying error signals before and after equipment startup and shutdown.

[0056] Beneficial effects of the present invention:

[0057] The present invention can identify the data before and after the equipment is started and shut down based on the historical vibration conditions of the equipment, and identify the false signals in the data, which can reduce the false alarms caused by vibration fluctuations during the start and shutdown of the equipment, improve the alarm accuracy of the monitoring system, and provide accurate guidance and suggestions for on-site operation and maintenance personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0059] Figure 1 It is a flow chart of the signal recognition method of the present invention. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] Example 1

[0062] A method for identifying false signals before and after equipment startup and shutdown, comprising the following steps:

[0063] S1, obtain equipment measurement point information, preset start and stop thresholds, and preset alarm thresholds;

[0064] Among them, the equipment measurement point information refers to obtaining all the measurement point IDs under the equipment, as well as the historical acceleration and speed effective value at each measurement point; the start and stop threshold is the acceleration value, and the start and stop judgment is made using acceleration data; the alarm threshold is the speed value, and the alarm judgment in S13 is made through the speed threshold, and then false alarms are eliminated based on whether it is abnormal data.

[0065] S2, obtain the historical acceleration and velocity effective values ​​of all vibration sensor measurement points of the equipment;

[0066] The device is usually equipped with multiple vibration sensor measurement points to read historical data from the server's database; all vibration sensors under the device collect acceleration and velocity data that are processed by the vibration collection unit module and uploaded to the platform.

[0067] S3, determine whether a start / stop threshold is preset, if the start / stop threshold is preset, proceed to S7, if not, proceed to S4;

[0068] S4, there are differences in vibration at different measuring points. The historical data of each measuring point of the equipment are sorted respectively, and the ten points before and after are eliminated. Furthermore, the vibration fluctuation rate of the remaining historical data of each measuring point is used as an evaluation index for selecting analysis data;

[0069] S5, selecting the data of the measuring point with the highest vibration fluctuation rate as analysis data to perform start-stop identification;

[0070] S6, classify based on the historical data of the measuring point, select the appropriate K value, use Euclidean distance to calculate the distance sorting to confirm the category, and select the smallest category by voting to obtain the characteristic value S as the start-stop threshold;

[0071] Among them, the Euclidean distance formula is:

[0072]

[0073] Among them, x i and x j are two data points respectively, k represents the feature dimension, and n is the total feature dimension.

[0074] S7, using the start / stop threshold to judge the start / stop status of the historical data of each measuring point;

[0075] S8, divide the historical data of each measuring point into three groups: shutdown data data1, start-stop data data2, and stable operation data data3, where the start-stop data is the data within the time range T before and after the shutdown data;

[0076] S9, using Gaussian kernel to calculate probability density functions f1, f2, and f3 for the three sets of data respectively;

[0077] The probability density function is:

[0078]

[0079] Where f(x) is the probability density, m is the array size, and x i is each data point, h is the bandwidth parameter, and K(μ) is the Gaussian kernel.

[0080] S10, calculating differential entropies h1, h2, h3 according to the probability density function;

[0081] The calculation formula of differential entropy is:

[0082]

[0083] Where h(x) is the differential entropy function.

[0084] S11, calculating the index factor of the abnormal data of start and stop according to the differential entropy, which is used for subsequent judgment of abnormal data;

[0085] The calculation formula of the index factor fac of abnormal start and shutdown data is:

[0086] fac=(a1×h1+a2×h2+a3×h3) / 3

[0087] Where a1, a2, a3 are coefficients, and h1, h2, h3 are differential entropies.

[0088] S12, based on the index factor of the abnormal start and stop data, using a sliding window to identify the abnormal start and stop data;

[0089] Set the window length Win, step length Len, index factor fac of abnormal start and shutdown data, calculate the window data value mean, window data standard deviation std, and determine whether the current vibration effective value rms satisfies: rms≤mean±std*fac; if satisfied, the data is normal; if not satisfied, it is abnormal data.

[0090] Among them, the effective value rms formula is:

[0091] Mean formula:

[0092] Standard deviation formula:

[0093] In the formula, x i is each data point in the window, and N is the window size.

[0094] S13, based on the above judgment, the abnormal start and shutdown data are marked, and the data exceeding the preset alarm threshold is searched to see whether it is abnormal data. If so, the alarm is cancelled.

[0095] Based on similar inventive concepts, an embodiment of the present invention also provides a computer storage medium storing a readable program, which can execute the above-mentioned method for identifying error signals before and after equipment startup and shutdown when the program is running.

[0096] Based on similar inventive concepts, an embodiment of the present invention provides an electronic device, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other through the communication bus;

[0097] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the above-mentioned method for identifying error signals before and after equipment startup and shutdown.

[0098] Based on similar inventive concepts, an embodiment of the present invention further provides a computer program product, including computer instructions, which instruct a computing device to execute operations corresponding to the above-mentioned method for identifying error signals before and after equipment startup and shutdown.

[0099] Example 2

[0100] In this embodiment, a system for identifying error signals before and after equipment startup and shutdown is proposed, which specifically includes:

[0101] Preset threshold acquisition module: obtain equipment measurement point information, preset start and stop thresholds, and preset alarm thresholds;

[0102] Measuring point data acquisition module: obtains the historical acceleration and velocity effective values ​​of all vibration sensor measuring points of the equipment;

[0103] Threshold preset judgment module: judge whether the start / stop threshold is preset, if the start / stop threshold is preset, enter the start / stop state judgment module, if the start / stop threshold is not preset, enter the measurement point data processing module;

[0104] Measuring point data processing module: Different measuring points have different vibrations. The historical data of each measuring point of the equipment are sorted respectively, and the ten points before and after are eliminated. Furthermore, the vibration fluctuation rate of the remaining historical data of each measuring point is used as an evaluation index for selecting analysis data;

[0105] Measuring point data screening module: selects the measuring point data with the highest vibration fluctuation rate as analysis data for start-stop identification;

[0106] Start-stop threshold setting module: classify based on the historical data of the measuring point, select the appropriate K value, use Euclidean distance calculation to sort and confirm the category, and select the smallest category by voting to obtain the characteristic value S as the start-stop threshold;

[0107] Start-stop status judgment module: Use the start-stop threshold to judge the start-stop status of each measuring point historical data

[0108] Measuring point data classification module: divides the historical data of each measuring point into three groups: shutdown data data1, start-stop data data2, and stable operation data data3, where the start-stop data is the data within the time range T before and after the shutdown data;

[0109] Probability density function calculation module: use Gaussian kernel to calculate probability density f1, f2, and f3 for three sets of data respectively;

[0110] Differential entropy calculation module: calculate differential entropy h1, h2, h3 according to probability density;

[0111] Index factor calculation module: calculates the index factor of abnormal start and shutdown data based on differential entropy, which is used for subsequent judgment of abnormal data;

[0112] Abnormal data identification module: Based on the indicator factors of abnormal start and stop data, the abnormal start and stop data is identified using a sliding window;

[0113] And, the alarm judgment module: based on the above judgment, the abnormal start and shutdown data is marked, and the data exceeding the preset alarm threshold is searched to see whether it is abnormal data. If so, the alarm is cancelled.

[0114] The method of the present invention may be implemented in hardware, firmware, or as software or computer code that may be stored in a recording medium (such as a CDROM, RAM, floppy disk, hard disk, or magneto-optical disk), or as computer code that is originally stored in a remote recording medium or a non-temporary machine-readable medium downloaded over a network and will be stored in a local recording medium, so that the method described herein may be stored in such software processing on a recording medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware (such as an ASIC or FPGA). It is understood that a computer, processor, microprocessor controller, or programmable hardware includes a storage component (e.g., RAM, ROM, flash memory, etc.) that can store or receive software or computer code, and when the software or computer code is accessed and executed by a computer, processor, or hardware, the method described herein is implemented. In addition, when a general-purpose computer accesses the code for implementing the method shown herein, the execution of the code converts the general-purpose computer into a dedicated computer for executing the method shown herein.

[0115] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. A method for identifying error signals before and after equipment startup and shutdown, characterized in that: The following steps are involved: S1, obtain equipment measurement point information, preset start and stop thresholds, and preset alarm thresholds; S2, obtain the historical acceleration and velocity effective values ​​of all vibration sensor measurement points of the equipment; S3, determine whether a start / stop threshold is preset, if the start / stop threshold is preset, proceed to S7, if not, proceed to S4; S4, there are differences in vibration at different measuring points. The historical data of each measuring point of the equipment are sorted respectively, and the ten points before and after are eliminated. Then the vibration fluctuation rate of the remaining historical data of each measuring point is used as the evaluation index for selecting analysis data; S5, selecting the data of the measuring point with the highest vibration fluctuation rate as analysis data to perform start-stop identification; S6, classify based on the historical data of the measuring point, use Euclidean distance to calculate the distance sorting to confirm the category, and select the smallest category by voting to obtain the characteristic value as the start-stop threshold; S7, using the start / stop threshold to judge the start / stop status of the historical data of each measuring point; S8, divide the historical data of each measuring point into three groups: shutdown data, start-stop data, and stable operation data; S9, calculate the probability density of the three sets of data using Gaussian kernel; S10, calculating differential entropy according to probability density; S11, calculating the index factor of the abnormal start and shutdown data according to the differential entropy; S12, based on the index factor of the abnormal start and stop data, using a sliding window to identify the abnormal start and stop data; S13, marking the abnormal start and shutdown data, and searching whether the data exceeding the preset alarm threshold is abnormal data, and if so, canceling the alarm.

2. According to claim 1, a method for identifying error signals before and after equipment startup and shutdown is characterized in that: The Euclidean distance formula: Among them, x i and x j are two data points respectively, k represents the feature dimension, and n is the total feature dimension.

3. The method for identifying error signals before and after equipment startup and shutdown according to claim 1 is characterized in that: The start and stop data refers to data within a time range T before and after the stop data.

4. The method for identifying error signals before and after equipment startup and shutdown according to claim 1 is characterized in that: The probability density calculation formula is: Where f(x) is the probability density, m is the array size, and x i is each data point, h is the bandwidth parameter, and K(μ) is the Gaussian kernel.

5. The method for identifying error signals before and after equipment startup and shutdown according to claim 4 is characterized in that: The differential entropy calculation formula is: Where h(x) is the differential entropy function; The calculation formula of the index factor fac of the start-stop abnormal data is: fac=(a1×h1+a2×h2+a3×h3) / 3 Where a1, a2, a3 are coefficients, and h1, h2, h3 are differential entropies.

6. The method for identifying error signals before and after equipment startup and shutdown according to claim 5 is characterized in that: The process of identifying the abnormal start and stop data is as follows: Set the window length Win, step length Len, index factor fac of abnormal start and shutdown data, calculate the window data mean, window data standard deviation std, and determine whether the current vibration effective value rms satisfies: rms≤mean±std*fac; if satisfied, the data is normal; if not satisfied, it is abnormal data.

7. The method for identifying error signals before and after equipment startup and shutdown according to claim 6 is characterized in that: The effective value rms calculation formula: The mean value calculation formula is: The standard deviation calculation formula is: Where N is the window size.

8. A system for identifying false signals before and after equipment startup and shutdown, characterized in that: include: Preset threshold acquisition module: obtain equipment measurement point information, preset start and stop thresholds, and preset alarm thresholds; Measuring point data acquisition module: obtains the historical acceleration and velocity effective values ​​of all vibration sensor measuring points of the equipment; Threshold preset judgment module: judge whether the start / stop threshold is preset, if the start / stop threshold is preset, enter the start / stop state judgment module, if the start / stop threshold is not preset, enter the measurement point data processing module; Measuring point data processing module: Different measuring points have different vibrations. The historical data of each measuring point of the equipment are sorted respectively, and the ten points before and after are eliminated. The vibration fluctuation rate of the remaining historical data of each measuring point is used as the evaluation index for selecting analysis data; Measuring point data screening module: selects the measuring point data with the highest vibration fluctuation rate as analysis data for start-stop identification; Start-stop threshold setting module: Classify based on the historical data of the measuring point, use Euclidean distance calculation to sort and confirm the category, and select the smallest category by voting to obtain the characteristic value as the start-stop threshold; Start-stop status judgment module: Use the start-stop threshold to judge the start-stop status of each measuring point historical data; Measuring point data classification module: divides the historical data of each measuring point into three groups: shutdown data, start-stop data, and stable operation data; Probability density function calculation module: calculate the probability density of the three sets of data using Gaussian kernel; Differential entropy calculation module: calculates differential entropy according to probability density; Index factor calculation module: calculates the index factor of abnormal start and shutdown data based on differential entropy; Abnormal data identification module: Based on the indicator factors of abnormal start and stop data, the abnormal start and stop data is identified using a sliding window; And, the alarm judgment module: marks the abnormal start and shutdown data, and searches whether the data exceeding the preset alarm threshold is abnormal data. If so, cancel the alarm.

9. A computer storage medium storing a readable program, characterized in that: When the program is running, it can execute the method for identifying error signals before and after equipment startup and shutdown as described in any one of claims 1 to 7.

10. An electronic device, characterized in that: include: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to a method for identifying error signals before and after equipment startup and shutdown as described in any one of claims 1-7.

Images