CN111830931A - Fault diagnosis method of DCS (distributed control system) - Google Patents

Fault diagnosis method of DCS (distributed control system) Download PDF

Info

Publication number
CN111830931A
CN111830931A CN202010683487.3A CN202010683487A CN111830931A CN 111830931 A CN111830931 A CN 111830931A CN 202010683487 A CN202010683487 A CN 202010683487A CN 111830931 A CN111830931 A CN 111830931A
Authority
CN
China
Prior art keywords
fault
devices
time
state
diagnosis method
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202010683487.3A
Other languages
Chinese (zh)
Other versions
CN111830931B (en
Inventor
孙锴
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Institute of Microelectronics of CAS
Original Assignee
Institute of Microelectronics of CAS
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Institute of Microelectronics of CAS filed Critical Institute of Microelectronics of CAS
Priority to CN202010683487.3A priority Critical patent/CN111830931B/en
Publication of CN111830931A publication Critical patent/CN111830931A/en
Application granted granted Critical
Publication of CN111830931B publication Critical patent/CN111830931B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B23/00—Testing or monitoring of control systems or parts thereof
    • G05B23/02—Electric testing or monitoring
    • G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
    • G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/20—Pc systems
    • G05B2219/24—Pc safety
    • G05B2219/24065—Real time diagnostics

Abstract

The invention provides a fault diagnosis method of a DCS (distributed control system), which comprises the following steps: acquiring an operation and maintenance log of a DCS (distributed control system); establishing a state corpus according to the operation and maintenance log, wherein the state corpus comprises state information of the M devices and an ID (identity) of each device in the M devices; different set values are given to different states in the state corpus; and counting the change condition of the set value of each device in the M devices along with the time. The method comprises the steps of establishing state information containing M devices and ID of each device, giving different set values to different state information, counting the change situation of the set values of the M devices and the n time node devices along with time in a fault matrix and the total fault score of the devices with faults in the fault matrix, and displaying the change of the fault states of the devices along with the time axis through the change of the set values along with the time axis, so that the fault change situation of the devices in the DCS is visualized, and the fault change situation of the devices in the DCS is convenient to diagnose.

Description

Fault diagnosis method of DCS (distributed control system)
Technical Field
The invention relates to the technical field of industrial control, in particular to a fault diagnosis method of a DCS (distributed control system).
Background
A Distributed Control System (DCS) is a core monitoring System of a modern power enterprise, uploads all production data of a power plant in real time, controls all operation parameters of the System, and plays a key role in the production process of the power plant. According to statistics, among various types of safety accidents, the safety accidents caused by the faults of the DCS system account for about 60% of the whole safety accidents. Therefore, the reliability of the DCS system itself is crucial to the safety production of the power enterprises. At the present stage, the operation state of the DCS system itself is all recorded in the operation and maintenance log of the system. Because the operation and maintenance logs are all recorded by English characters, the formats are various, the devices are numerous, hundreds of thousands of records exist in each hour, and manual processing is difficult to be sufficient.
Disclosure of Invention
The invention provides a fault diagnosis method of a DCS (distributed control system), which is used for diagnosing faults of the DCS by using operation and maintenance logs of the DCS and realizing rapid fault diagnosis and tracing of equipment in the DCS.
The invention provides a fault diagnosis method of a DCS (distributed control system), wherein the DCS comprises a plurality of devices. The fault diagnosis method comprises the following steps: acquiring an operation and maintenance log of a DCS (distributed control system); establishing a state corpus according to the operation and maintenance log, wherein the state corpus comprises state information of the M devices and an ID (identity) of each device in the M devices; different set values are given to different state information in the state corpus; and counting the change condition of the set value of each device in the M devices in the state corpus along with the time.
In the scheme, different set values are given to different state information by establishing the state information containing M devices and the ID of each device, and the change situation of the set value of each device in the M devices along with time is counted, so that the change of the fault state of the devices along with the time axis is displayed through the change of the set value along with the time axis, the fault change situation of the devices in the DCS is visualized, the fault change situation of the devices in the DCS is conveniently diagnosed, and the rapid fault diagnosis and source tracing of the devices in the DCS are realized.
In a specific embodiment, establishing a state corpus according to the operation and maintenance log specifically includes: traversing operation and maintenance logs of the DCS according to rows; the operation and maintenance logs of the DCS are saved into the same set format by using a word segmentation algorithm of natural language processing; and (4) counting keywords representing fault information, and establishing a state corpus of the M devices. The operation and maintenance logs with disordered formats and huge information amount are converted into files with the same set format which are easy to perform statistical analysis by utilizing a natural language processing technology, so that a state corpus is conveniently established by counting keywords representing fault information.
In one embodiment, the same set format is a csv format to facilitate statistical analysis using a word statistics algorithm for natural language processing.
In a specific embodiment, the step of assigning different setting values to different state information in the state corpus is specifically: assigning 0 to the no fault state; assigning a 1 to the first fault condition; assigning 2 to the second fault condition; …, respectively; the nth fault state is given a fault number N. So as to intuitively reveal the fault status category of the device.
In a specific embodiment, the counting the change of the setting value of each of the M devices with time includes: screening M devices from the M devices, wherein each device in the M devices has at least one fault within a set time period; in a set time period, when at least one device of m devices has a fault, picking up a time node, and picking up n time nodes in total; arranging n time nodes according to time sequence to serve as a row of a matrix, and establishing a fault matrix X by taking an ID (identity) with a fault on the corresponding time node in m devices as a column of the matrix:
Figure BDA0002585343720000021
wherein X in the fault matrix XijThe fault number of the ith device at the jth time node is represented, and the state change conditions of the multiple devices are displayed in a matrix mode, so that the visualization is realized.
In a specific embodiment, the counting the change of the setting value of each of the M devices with time further includes: and (4) endowing the fault numbers representing different fault states with different colors, and drawing a fault map. By formulating coloring rules, a system fault color map is constructed, and the fault state and the occurrence time sequence of system equipment are reflected in the form of images, so that the system fault color map is more visualized.
In a specific embodiment, the counting the change of the setting value of each of the M devices with time further includes: counting the ID of equipment with faults of each time node in the n time nodes and the corresponding fault number; and (3) taking the fault occurrence time of each device in the m devices as an axis, and drawing a fault propagation chain of the DCS, so that the fault propagation condition among the devices can be conveniently found, and automatic fault diagnosis and rapid source tracing are realized.
In a specific embodiment, with the failure occurrence time of each of the m devices as an axis, the step of drawing a failure propagation chain of the DCS system is specifically as follows: drawing IDs of m devices; and a one-way arrow is adopted to connect the ID of the equipment with the fault occurring in front and the ID of the equipment with the fault occurring in back in the m equipment between any two adjacent time nodes in the n time nodes, and the ID of the equipment with the fault occurring in front points to the ID of the equipment with the fault occurring in back by the one-way arrow, so that the fault propagation condition among the equipment is more intuitively displayed.
In a specific implementation mode, the fault number sequentially represents the severity of the fault state of the equipment from 1 to N from small to large, so that the fault severity of each equipment is shown from light to heavy through the fault number change condition of each equipment, and the numerical value of the fault number is positively correlated with the fault severity.
In a specific embodiment, the counting the change of the setting value of each of the M devices with time further includes: summing the rows in the fault matrix X yields:
Figure BDA0002585343720000022
wherein the content of the first and second substances,
Figure BDA0002585343720000023
and the sum of the number of faults of the ith device in a set time period is represented, and the contribution degree of each device to the system fault can be obtained, so that the device needing important attention is determined.
In a specific embodiment, the counting the change of the setting value of each of the M devices with time further includes: taking the number of faults per row in the fault matrix X as an exponential function exIs obtained by using the following formula pairSumming is performed for each row:
Figure BDA0002585343720000024
wherein the content of the first and second substances,
Figure BDA0002585343720000025
the fault severity of the ith equipment in a set time period is shown so as to more truly simulate the damage degree of the fault state of certain equipment in a certain time period to the whole system.
In a specific embodiment, the counting the change of the setting value of each of the M devices with time further includes: summing the columns in the failure matrix X yields:
Figure BDA0002585343720000031
wherein the content of the first and second substances,
Figure BDA0002585343720000032
and the sum of the number of failures of m devices at the jth time node is represented, and the contribution degree of the system failure under each time node can be obtained, so that the time node needing important attention is determined.
In a specific embodiment, the counting the change of the setting value of each of the M devices with time further includes: taking the number of faults in each column in the fault matrix X as an exponential function exThe independent variables of (a) are summed for each column using the following formula:
Figure BDA0002585343720000033
wherein the content of the first and second substances,
Figure BDA0002585343720000034
and the fault severity of the m devices at the jth time node is represented, so that the damage degree of the fault conditions of the m devices at a certain time node to the whole system is simulated more truly.
In a specific embodiment, the counting the change of the setting value of each of the M devices with time further includes: taking all elements in the fault matrix X as an exponential function exIs summed over all elements of the fault matrix X using the following formula:
Figure BDA0002585343720000035
wherein, Fault _ score represents the Fault severity of the m devices in a set time period, so as to more truly simulate the damage degree of the Fault conditions of the m devices in a certain time period to the whole system.
Drawings
Fig. 1 is a flowchart of a fault diagnosis method of a DCS system according to an embodiment of the present invention;
fig. 2 is a schematic diagram for drawing a fault propagation chain according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In order to facilitate understanding of the fault diagnosis method of the DCS system provided by the embodiment of the present invention, an application scenario of the fault diagnosis method of the DCS system provided by the embodiment of the present invention is first described below, where the fault diagnosis method of the DCS system is applied to the field of an industrial control system, where the DCS system includes a plurality of devices, and the fault diagnosis method is used to perform fault diagnosis on the plurality of devices in the DCS system. The fault diagnosis method of the DCS system will be described in detail below with reference to the accompanying drawings.
Referring to fig. 1, a method for diagnosing a fault of a DCS system according to an embodiment of the present invention includes:
s10: acquiring an operation and maintenance log of a DCS (distributed control system);
s20: establishing a state corpus according to the operation and maintenance log, wherein the state corpus comprises state information of the M devices and an ID (identity) of each device in the M devices;
s30: different set values are given to different state information in the state corpus;
s40: and counting the change condition of the set value of each device in the M devices in the state corpus along with the time.
In the scheme, different set values are given to different state information by establishing the state information containing M devices and the ID of each device, and the change situation of the set value of each device in the M devices along with time is counted, so that the change of the fault state of the devices along with the time axis is displayed through the change of the set value along with the time axis, the fault change situation of the devices in the DCS is visualized, the fault change situation of the devices in the DCS is conveniently diagnosed, and the rapid fault diagnosis and source tracing of the devices in the DCS are realized. The above steps will be described in detail with reference to the accompanying drawings.
Firstly, an operation and maintenance log of the DCS is obtained. During specific implementation, the operation and maintenance log can be downloaded and stored in a txt format. It should be understood that the format in which the operation and maintenance log is stored is not limited to txt format.
And secondly, establishing a state corpus according to the operation and maintenance log, wherein the state corpus comprises state information of the M devices and the ID of each device in the M devices. Where the ID of each device serves as a label to distinguish different devices, it may be used to distinguish different devices.
When the state corpus is established according to the operation and maintenance logs, the operation and maintenance logs of the DCS can be traversed according to lines; then, the operation and maintenance logs of the DCS are saved into the same set format by using a word segmentation algorithm processed by natural language; then, the keywords representing the fault information are counted, and a state corpus of the M devices is established. The operation and maintenance logs with disordered formats and huge information amount are converted into files with the same set format which are easy to perform statistical analysis by utilizing a natural language processing technology, so that a state corpus is conveniently established by counting keywords representing fault information. The same set format can be selected as the csv format, so that the statistical analysis can be performed by using a word statistical algorithm of natural language processing. Specifically, when the operation and maintenance log in the txt format is converted into the csv format, the text file in the txt format can be converted into the file in the csv format by using the natural language processing function of python. Specifically, a word segmentation dictionary of English words in the operation and maintenance log can be designed according to the file format characteristics of the operation and maintenance log; traversing all the lines of the txt format operation and maintenance log file, and segmenting words in line units based on the word segmentation dictionary; then, recording the word segmentation result into a csv format file as a later analysis basis; then, sorting the csv format file, and deleting the empty row, the repeated item and the header; and then, counting keywords representing fault information, and establishing a state corpus of equipment faults in the DCS. In addition, in addition to the operation and maintenance log being stored as the file in the csv format, other methods may be adopted. For example, various items of information in the operation and maintenance log of the DCS system may also be saved as a table with a set format, so as to perform statistical analysis by using a word statistical algorithm of natural language processing.
Then, different setting values are given to different state information in the state corpus. The fault state information in the csv format operation and maintenance log file can be screened based on the state corpus, and different fault types are marked with different integer values. For example, giving different setting values to different state information in the state corpus can adopt: assigning a fault number of 0 to the fault-free state; assigning a failure number of 1 to the first failure state; assigning a fault number of 2 to the second fault condition; …, respectively; the nth fault state is given a fault number N. So as to intuitively reveal the fault status category of the device. When different set values are marked for different fault types, the fault number can be sequentially represented from 1-N from small to large, the severity of the fault state of the equipment is sequentially represented from light to heavy, the numerical value of the fault number is positively correlated with the severity of the fault, and the fault severity of each equipment is shown through the fault number change condition of each equipment.
Next, the change of the setting value of each of the M devices with time is counted. Specifically, the statistics of the failure matrix may be established, which includes: screening M devices from the M devices, wherein each device in the M devices has at least one fault within a set time period; in a set time period, when at least one device of m devices has a fault, picking up a time node, and picking up n time nodes in total; arranging n time nodes according to time sequence to serve as a row of a matrix, and establishing a fault matrix X by taking an ID (identity) with a fault on the corresponding time node in m devices as a column of the matrix:
Figure BDA0002585343720000041
wherein X in the fault matrix XijThe fault number of the ith equipment in the jth time period is represented, so that the state change conditions of the multiple equipment are displayed in a matrix mode, and the visualization is realized. Where M is less than or equal to M, any M devices that are desired to be focused on can be selected from the M devices. The set time period mentioned above refers to a time period in which at least one device has failed, which may be any time period in which attention is desired.
For example, when the number m of faulty devices is 4, the IDs of 4 faulty devices are ID _1, ID _2, ID _3, and ID _4, respectively. 4 time nodes exist in the 4 devices within a set time period, and are a first time node, a second time node, a third time node and a fourth time node in sequence. At each time node, at least one of the 4 devices fails. For example, at the first time, the ID _2 device and the ID _4 device have faults, wherein the fault number of the fault type of the ID _2 device is 1, and the fault number of the fault type of the ID _4 device is 3; the other two faulty devices are normal and are represented by the fault number 0. At the second time node, the ID _1 device has a failure with a failure type of 2, and the other three devices have no failure. At the third time node, the ID _2 device has a failure of the failure type with a failure number of 2, and the other three devices have no failure. At the fourth time node, the ID _1 device has a failure with a failure type of 1, the ID _3 device has a failure with a failure type of 1, and the other two devices have no failure. The 4 × 4 fault matrix X established is:
Figure BDA0002585343720000051
the state change conditions of the multiple devices are displayed in a matrix mode, and therefore the device is more visual. It should be understood that the above only shows one way of assigning different fault values to different fault types, and other ways may be adopted, for example, a fault number of 0 may be assigned without a fault, and a fault number of 1 may be assigned with a fault, where the established fault matrix X may be:
Figure BDA0002585343720000052
in addition, the counting of the change of the setting value of each of the M devices with time may further include: and (4) endowing different fault numbers with different colors, and drawing a fault map. By formulating coloring rules, a system fault color map is constructed, and the fault state and the occurrence time sequence of system equipment are reflected in the form of images, so that the system fault color map is more visualized.
For example, different colors, 0-green, 1-yellow, 2-red, 3-blue, …, may be identified for different numbers of faults in the following manner. The fault color map established may be:
Figure BDA0002585343720000053
by formulating coloring rules, a system fault color map is constructed, and the fault state and the occurrence time sequence of system equipment are reflected in the form of images, so that the system fault color map is more visualized.
In addition, the counting of the change of the setting value of each of the M devices with time may further include:
counting the ID of equipment with faults of each time node in the n time nodes and the corresponding fault number;
and (3) taking the fault occurrence time of each device in the m devices as an axis, and drawing a fault propagation chain of the DCS, so that the fault propagation condition among the devices can be conveniently found, and automatic fault diagnosis and rapid source tracing are realized.
Specifically, the horizontal axis in the failure matrix X may be reserved as a time axis. The vertical axis in the fault matrix X is compressed into a one-dimensional vector with only a time axis. Specifically, in each column of the fault matrix X, only the device ID with a fault at each time node is referred to the first row of the fault matrix, and the obtained matrix is:
[ID_2+ID_4 ID_1 ID_2 ID1+ID_3]
time axis
Through the mode, the source tracing method can clearly obtain the sequence of different devices along with the fault so as to facilitate the source tracing.
Specifically, when a fault propagation chain of the DCS is drawn by taking the fault occurrence time of each device in m devices as an axis, the IDs of the m devices can be drawn firstly; and then, a one-way arrow is adopted to connect the ID of the equipment with the fault occurring in the front and the ID of the equipment with the fault occurring in the rear in the m equipment between any two adjacent time nodes in the n time nodes, and the ID of the equipment with the fault occurring in the front points to the ID of the equipment with the fault occurring in the rear by the one-way arrow, so that the fault propagation condition among the equipment can be more intuitively displayed. For example, referring to fig. 2, a block diagram of different device IDs is first drawn to facilitate the IDs of 4 devices. And the ID _2 equipment and the ID _4 equipment on the first time node have faults, and the faults serve as fault sources. And on the second time node, if the ID _1 device fails, connecting the ID _2 device with the ID _1 device by adopting a one-way arrow, pointing the one-way arrow to the ID _1 device by the ID _2 device, simultaneously connecting the ID _4 device with the ID _1 device by adopting the one-way arrow, and pointing the one-way arrow to the ID _1 device by the ID _4 device. And if the ID _2 device fails at the third time node, connecting the ID _1 device with the ID _2 device by adopting a one-way arrow, wherein the ID _1 device points to the ID _2 device by the one-way arrow. And on the fourth time node, if the ID _1 device and the ID _3 device have faults, the ID _2 device and the ID _3 device are connected by adopting a one-way arrow, the ID _2 device points to the ID _3 device, the ID _2 device and the ID _1 device are connected by adopting the one-way arrow, and the ID _2 device points to the ID _1 device. It should be understood that the above only shows one way to plot a propagation chain, and that other propagation chain diagrams capable of representing fault propagation relationships between devices may be used in addition.
In addition, when the number of faults is from 1 to N, and the severity of the fault state of the equipment is represented from light to heavy in sequence, the counting of the change situation of the set value of each equipment in the M equipment along with time may further include: summing the rows in the fault matrix X yields:
Figure BDA0002585343720000061
wherein the content of the first and second substances,
Figure BDA0002585343720000062
and the sum of the number of faults of the ith device in the set time period can obtain the contribution degree of each device to the system fault, so that the device needing important attention is determined. For example, summing the rows in the fault matrix X described above yields:
sum1＝[3 3 1 3]
the contribution degree of each device to the system fault can be obtained more intuitively, so that the devices needing important attention are determined.
The following method can be adopted for counting the change situation of the set value of each device in the M devices along with the time: taking the number of faults per row in the fault matrix X as an exponential function exThe independent variables of (a) are summed for each row using the following formula:
Figure BDA0002585343720000063
wherein the content of the first and second substances,
Figure BDA0002585343720000064
indicates that the ith device is in the settingSeverity of the fault in the time period. By taking the change of the number of faults of each device under n time nodes as an exponential function exThe independent variable of (2) converts the original fault number which changes linearly into diffused and nonlinear change, thereby more truly simulating the condition that the larger the fault number of the equipment is, the more serious the damage degree of the system is, and more truly simulating the damage degree of the fault state of a certain equipment in a period of time on the whole system.
In addition, when the number of faults is from 1 to N, and the severity of the fault state of the equipment is represented from light to heavy in sequence, the counting of the change situation of the set value of each equipment in the M equipment along with time may further include: summing the columns in the failure matrix X yields:
Figure BDA0002585343720000071
wherein the content of the first and second substances,
Figure BDA0002585343720000072
and displaying the sum of the number of the faults of the m devices at the jth time node, and obtaining the contribution degree of the system fault under each time node, thereby determining the time nodes needing important attention.
In addition, the following method can be adopted for counting the change situation of the set value of each device in the M devices along with the time: taking the number of faults in each column in the fault matrix X as an exponential function exThe independent variables of (a) are summed for each column using the following formula:
Figure BDA0002585343720000073
wherein the content of the first and second substances,
Figure BDA0002585343720000074
indicating the fault severity of the m devices under the jth time node. The failure number of m devices under each time node is used as an independent variable of an exponential function e, and the failure number which originally changes linearly is converted into the diffused and nonlinear change, so that the failure number is further changedThe larger the number of faults of the equipment is, the more serious the damage degree of the system is simulated really, so that the damage degree of the fault conditions of the m equipment at a certain time node to the whole system is simulated more really.
The following method can be adopted for counting the change situation of the set value of each device in the M devices along with the time: taking all elements in the fault matrix X as an exponential function exIs summed over all elements of the fault matrix X using the following formula:
Figure BDA0002585343720000075
wherein the Fault _ score represents a severity of a failure of the m devices within a set period of time. By taking the number of failures of m devices under n time nodes as an exponential function exThe independent variable of (1) converts the original linearly-changed fault number into a diffused and nonlinear change, thereby more truly simulating the condition that the larger the fault number of the equipment is, the more serious the damage degree to the system is, and more truly simulating the damage degree of the fault conditions of m equipment in a certain time period to the whole system.
Different set values are given to different state information in a state corpus by establishing state information containing M devices and ID of each device, and the change situation of the set value of each device in the M devices along with time is counted, so that the change of the fault state of the devices along with the time axis is displayed through the change of the set value along with the time axis, the fault change situation of the devices in the DCS system is visualized, the fault change situation of the devices in the DCS system is conveniently diagnosed, and the rapid fault diagnosis and source tracing of the devices in the DCS system are realized.
The above description is only for the specific embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (14)

1. A fault diagnosis method of a DCS system, wherein the DCS system includes a plurality of devices, the fault diagnosis method comprising:
acquiring an operation and maintenance log of the DCS;
establishing a state corpus according to the operation and maintenance log; the state corpus comprises state information of M devices and an ID (identity) of each device in the M devices;
different set values are given to different state information in the state corpus;
and counting the change condition of the set value of each device in the M devices in the state corpus along with time.
2. The method for fault diagnosis according to claim 1, wherein the establishing the state corpus according to the operation and maintenance log is specifically:
traversing the operation and maintenance logs of the DCS according to rows;
the operation and maintenance logs are saved into the same set format by utilizing a word segmentation algorithm of natural language processing;
and carrying out statistics on keywords representing state information, and establishing a state corpus of the M devices.
3. The fault diagnosis method according to claim 2, characterized in that said same set format is a csv format.
4. The fault diagnosis method according to claim 1, wherein the assigning of different setting values to different state information in the state corpus is specifically:
assigning a fault number of 0 to the fault-free state;
assigning a failure number of 1 to the first failure state;
assigning a fault number of 2 to the second fault condition;
…
the nth fault state is given a fault number N.
5. The fault diagnosis method according to claim 4, wherein said counting the change over time of the set value of each of the M devices comprises:
screening M devices from the M devices, wherein each device in the M devices has at least one fault within a set time period;
in the set time period, picking up a time node when at least one of the m devices fails, and picking up n time nodes in total;
the n time nodes are arranged in time sequence to serve as a row of a matrix, and the ID with the fault on the corresponding time node in the m devices serves as a column of the matrix, so that the following fault matrix X is established:
Figure FDA0002585343710000011
wherein X in the fault matrix XijRepresenting the number of failures of the ith device at the jth time node.
6. The fault diagnosis method according to claim 5, wherein said counting the change over time of the set value of each of the m devices further comprises:
and (4) endowing the fault numbers representing different fault states with different colors, and drawing a fault map.
7. The fault diagnosis method according to claim 5, wherein said counting the change over time of the set value of each of the M devices further comprises:
counting the ID of the equipment with the fault of each time node in the n time nodes and the corresponding fault number;
and drawing a fault propagation chain of the DCS by taking the fault occurrence time of each device in the m devices as an axis.
8. The fault diagnosis method according to claim 7, wherein the step of plotting the fault propagation chain of the DCS system with the time of occurrence of the fault of each of the m devices as an axis is specifically as follows:
drawing the IDs of the m devices;
and connecting the ID of the equipment with the fault occurring in front and the ID of the equipment with the fault occurring in back in the m equipment by adopting a one-way arrow between any two adjacent time nodes in the n time nodes, wherein the ID of the equipment with the fault occurring in front points to the ID of the equipment with the fault occurring in back by the one-way arrow.
9. The fault diagnosis method according to claim 5, characterized in that the number of faults is from 1 to N, from small to large, and the severity of the fault state of the equipment is sequentially represented from light to heavy.
10. The fault diagnosis method according to claim 9, wherein said counting the change over time of the set value of each of the M devices further comprises:
summing the rows in the fault matrix X yields:
Figure FDA0002585343710000021
wherein the content of the first and second substances,
Figure FDA0002585343710000022
and the sum of the number of faults of the ith equipment in the set time period is represented.
11. The fault diagnosis method according to claim 9, wherein said counting the change over time of the set value of each of the M devices further comprises:
taking the number of faults per row in the fault matrix X as an exponential function exThe independent variables of (a) are summed for each row using the following formula:
Figure FDA0002585343710000023
wherein the content of the first and second substances,
Figure FDA0002585343710000024
indicating the severity of the fault of the ith device over the set period of time.
12. The fault diagnosis method according to claim 9, wherein said counting the change over time of the set value of each of the M devices further comprises:
summing the columns in the failure matrix X yields:
Figure FDA0002585343710000025
wherein the content of the first and second substances,
Figure FDA0002585343710000026
and the sum of the failure numbers of the m devices at the j time node is represented.
13. The fault diagnosis method according to claim 9, wherein said counting the change over time of the set value of each of the M devices further comprises:
taking the number of faults in each column in the fault matrix X as an exponential function exThe independent variables of (a) are summed for each column using the following formula:
Figure FDA0002585343710000027
wherein the content of the first and second substances,
Figure FDA0002585343710000028
and indicating the fault severity of the m devices at the j time node.
14. The fault diagnosis method according to claim 9, wherein said counting the change over time of the set value of each of the M devices further comprises:
taking all elements in the fault matrix X as an exponential function exIs summed over all elements of the fault matrix X using the following formula:
Figure FDA0002585343710000029
wherein the Fault _ score represents a severity of the failure of the m devices within the set time period.
CN202010683487.3A 2020-07-15 2020-07-15 Fault diagnosis method of DCS (distributed control system) Active CN111830931B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010683487.3A CN111830931B (en) 2020-07-15 2020-07-15 Fault diagnosis method of DCS (distributed control system)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010683487.3A CN111830931B (en) 2020-07-15 2020-07-15 Fault diagnosis method of DCS (distributed control system)

Publications (2)

Publication Number Publication Date
CN111830931A true CN111830931A (en) 2020-10-27
CN111830931B CN111830931B (en) 2021-08-20

Family

ID=72924093

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010683487.3A Active CN111830931B (en) 2020-07-15 2020-07-15 Fault diagnosis method of DCS (distributed control system)

Country Status (1)

Country Link
CN (1) CN111830931B (en)

Citations (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1433535A (en) * 2000-01-29 2003-07-30 Abb研究有限公司 Method for automatic fault tree synthesis
JP2007100976A (en) * 2005-09-30 2007-04-19 Mitsubishi Heavy Ind Ltd Device and method for diagnosing failure of gas engine driven-air conditioner
US7403850B1 (en) * 2005-09-29 2008-07-22 Dynalco Controls Corporation Automated fault diagnosis method and system for engine-compressor sets
CN103761173A (en) * 2013-12-28 2014-04-30 华中科技大学 Log based computer system fault diagnosis method and device
CN106406229A (en) * 2016-12-20 2017-02-15 吉林大学 Numerical control machine tool fault diagnosis method
CN107246959A (en) * 2017-07-12 2017-10-13 西安因联信息科技有限公司 The diagnostic system and method for a kind of equipment fault based on wireless senser
CN109343395A (en) * 2018-10-17 2019-02-15 深圳中广核工程设计有限公司 A kind of abnormality detection system and method for nuclear power plant DCS operation log
CN109491365A (en) * 2018-11-23 2019-03-19 陕西千山航空电子有限责任公司 A kind of fault diagnosis report rapid generation
CN109684447A (en) * 2018-12-13 2019-04-26 贵州电网有限责任公司 A kind of dispatching of power netwoks running log fault information analysis method based on text mining
JP2019116377A (en) * 2017-12-27 2019-07-18 株式会社日立ビルシステム Elevator failure diagnosis system
CN110288004A (en) * 2019-05-30 2019-09-27 武汉大学 A kind of diagnosis method for system fault and device excavated based on log semanteme
CN110855502A (en) * 2019-11-22 2020-02-28 叶晓斌 Fault cause determination method and system based on time-space analysis log
CN111061235A (en) * 2019-12-20 2020-04-24 中核控制系统工程有限公司 DCS equipment diagnosis method with fault early warning function
CN111061584A (en) * 2019-11-21 2020-04-24 浪潮电子信息产业股份有限公司 Fault diagnosis method, device, equipment and readable storage medium
CN111090973A (en) * 2019-11-26 2020-05-01 北京明略软件系统有限公司 Report generation method and device and electronic equipment
CN111190415A (en) * 2020-01-21 2020-05-22 北京市劳动保护科学研究所 Industrial control system availability testing method and system
CN111337792A (en) * 2020-04-14 2020-06-26 上海海事大学 Power system fault diagnosis method based on improved Petri network

Patent Citations (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1433535A (en) * 2000-01-29 2003-07-30 Abb研究有限公司 Method for automatic fault tree synthesis
US7403850B1 (en) * 2005-09-29 2008-07-22 Dynalco Controls Corporation Automated fault diagnosis method and system for engine-compressor sets
JP2007100976A (en) * 2005-09-30 2007-04-19 Mitsubishi Heavy Ind Ltd Device and method for diagnosing failure of gas engine driven-air conditioner
CN103761173A (en) * 2013-12-28 2014-04-30 华中科技大学 Log based computer system fault diagnosis method and device
CN106406229A (en) * 2016-12-20 2017-02-15 吉林大学 Numerical control machine tool fault diagnosis method
CN107246959A (en) * 2017-07-12 2017-10-13 西安因联信息科技有限公司 The diagnostic system and method for a kind of equipment fault based on wireless senser
JP2019116377A (en) * 2017-12-27 2019-07-18 株式会社日立ビルシステム Elevator failure diagnosis system
CN109343395A (en) * 2018-10-17 2019-02-15 深圳中广核工程设计有限公司 A kind of abnormality detection system and method for nuclear power plant DCS operation log
CN109491365A (en) * 2018-11-23 2019-03-19 陕西千山航空电子有限责任公司 A kind of fault diagnosis report rapid generation
CN109684447A (en) * 2018-12-13 2019-04-26 贵州电网有限责任公司 A kind of dispatching of power netwoks running log fault information analysis method based on text mining
CN110288004A (en) * 2019-05-30 2019-09-27 武汉大学 A kind of diagnosis method for system fault and device excavated based on log semanteme
CN111061584A (en) * 2019-11-21 2020-04-24 浪潮电子信息产业股份有限公司 Fault diagnosis method, device, equipment and readable storage medium
CN110855502A (en) * 2019-11-22 2020-02-28 叶晓斌 Fault cause determination method and system based on time-space analysis log
CN111090973A (en) * 2019-11-26 2020-05-01 北京明略软件系统有限公司 Report generation method and device and electronic equipment
CN111061235A (en) * 2019-12-20 2020-04-24 中核控制系统工程有限公司 DCS equipment diagnosis method with fault early warning function
CN111190415A (en) * 2020-01-21 2020-05-22 北京市劳动保护科学研究所 Industrial control system availability testing method and system
CN111337792A (en) * 2020-04-14 2020-06-26 上海海事大学 Power system fault diagnosis method based on improved Petri network

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
孙锴: "流程工业系统故障预警及故障诊断模型", 《数字技术与应用》 *
魏晨辉: "基于高端容错计算机的故障日志分析系统", 《清华大学学报（自然科学版）》 *

Also Published As

Publication number Publication date
CN111830931B (en) 2021-08-20

Similar Documents

Publication Publication Date Title
US5847972A (en) Method and apparatus for graphically analzying a log-file
KR20210019564A (en) Operation maintenance system and method
US7702078B2 (en) Method, system and computer program product for facilitating the analysis of automatic line insulation testing data
CN111309505B (en) Page exception handling method and device and electronic equipment
CN109885478A (en) A kind of localization method and system of error code
CN107193730A (en) A kind of interface test method of automation
CN108108445A (en) A kind of data intelligence processing method and system
CN110766100B (en) Bearing fault diagnosis model construction method, bearing fault diagnosis method and electronic equipment
CN111367782B (en) Regression testing data automatic generation method and device
CN111830931B (en) Fault diagnosis method of DCS (distributed control system)
CN114647558A (en) Method and device for detecting log abnormity
CN112256576B (en) Man-machine dialogue corpus testing method, device, equipment and storage medium
CN116955071A (en) Fault classification method, device, equipment and storage medium
CN110609761B (en) Method and device for determining fault source, storage medium and electronic equipment
CN108335236A (en) A kind of source of houses leakage broker's detection method and device
CN110554956B (en) BDMS automatic test method
CN108279013A (en) The inspection error correction method and device and navigation system of electronic map incremental data
CN115660251A (en) Enterprise health degree evaluation system based on AI big data
CN116414587A (en) Fault data acquisition method, fault processing method, electronic device and storage medium
CN113986900A (en) Data quality problem grading processing method, storage medium and system
CN112906891A (en) Expert system knowledge base construction method and device based on machine learning
CN115599621A (en) Micro-service abnormity diagnosis method, device, equipment and storage medium
CN117439899B (en) Communication machine room inspection method and system based on big data
CN116126738B (en) Interface abnormality identification method and device and electronic equipment
CN117234762A (en) Exception handling method and system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant