CN114962239B - Equipment fault detection method based on intelligent Internet of things - Google Patents
Equipment fault detection method based on intelligent Internet of things Download PDFInfo
- Publication number
- CN114962239B CN114962239B CN202210627061.5A CN202210627061A CN114962239B CN 114962239 B CN114962239 B CN 114962239B CN 202210627061 A CN202210627061 A CN 202210627061A CN 114962239 B CN114962239 B CN 114962239B
- Authority
- CN
- China
- Prior art keywords
- pump
- abnormal
- key
- vibration
- flow
- 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.)
- Active
Links
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
- F04B—POSITIVE-DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS
- F04B49/00—Control, e.g. of pump delivery, or pump pressure of, or safety measures for, machines, pumps, or pumping installations, not otherwise provided for, or of interest apart from, groups F04B1/00 - F04B47/00
- F04B49/10—Other safety measures
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
- F04B—POSITIVE-DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS
- F04B2201/00—Pump parameters
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Control Of Positive-Displacement Pumps (AREA)
Abstract
The invention relates to the field of artificial intelligence, and provides an equipment fault detection method based on an intelligent Internet of things, which comprises the following steps: acquiring a temperature curve and a vibration curve of a key pump; acquiring an abnormal section of a vibration curve in an abnormal time period of a key pump; determining the number of abnormal vibration cycles and the vibration amplitude of the abnormal cycles; determining a temperature amplitude corresponding to the abnormal vibration period; calculating an abnormal coefficient; determining the distribution flow weight value carried by the common pump corresponding to the key pump; obtaining the flow rate to be distributed by the key pump; determining a flow distribution scheme corresponding to the key pump; calculating the abnormal degree of the distribution scheme; and selecting the minimum value of the abnormal degree, and processing the key pump according to the minimum value of the abnormal degree. The invention can master the state of the equipment in real time, effectively reduce the times of unplanned shutdown and ensure the production benefit.
Description
Technical Field
The invention relates to the field of artificial intelligence, in particular to an equipment fault detection method based on an intelligent Internet of things.
Background
The iron and steel coking enterprise belongs to an equipment asset intensive enterprise, field equipment comprises equipment such as a tower tank, a pipeline, a reactor, a pump group, a compressor, a turbine and the like, the automation level is high, the production continuity is high, and the iron and steel coking enterprise has the characteristics of high temperature, high pressure, flammability, explosiveness, easy corrosion and easy poisoning.
The maintenance of the existing coke-oven plant pump is mainly in a mode of combining preventive maintenance with manual point inspection. Enterprises can regularly overhaul certain devices and equipment, but lack real-time effective data, so that the good overhaul effect is difficult to achieve, and a large amount of maintenance work and cost increase are brought. The on-site security system lacks a diagnosis and analysis function or is difficult to effectively utilize the diagnosis function, and only can find late-stage faults. Meanwhile, maintenance decisions are influenced due to lack of equipment operation data.
Aiming at the situation, the invention provides the equipment fault detection method based on the intelligent Internet of things by installing the detection heads on the vibration obvious parts of a plurality of key pumps of the coking device to obtain the vibration parameters and the temperature parameters and analyzing the obtained parameter information.
Disclosure of Invention
The invention provides an equipment fault detection method based on an intelligent Internet of things, which aims to solve the problem that the pump state cannot be monitored in real time in the prior art.
The invention discloses an equipment fault detection method based on an intelligent Internet of things, which adopts the following technical scheme:
acquiring a temperature curve and a vibration curve of each key pump in a pump group;
acquiring a vibration curve abnormal section in an abnormal time period when the key pump is abnormal by using a vibration curve of the key pump;
determining the number of abnormal vibration periods and the vibration amplitude of each abnormal period in the abnormal section of the vibration curve by using the abnormal section of the vibration curve;
extracting a temperature amplitude corresponding to each abnormal vibration period in a corresponding time period in the temperature curve according to the abnormal section of the vibration curve;
calculating the abnormal coefficient of the abnormal section of the vibration curve according to the difference value of the vibration amplitude of each abnormal period in the abnormal section of the vibration curve and the vibration amplitude of the standard period, the average value of the vibration amplitudes of all the vibration periods in the abnormal section of the vibration curve and the vibration amplitude difference value of the standard period and the temperature amplitude corresponding to each abnormal vibration period;
acquiring common pumps corresponding to abnormal key pump shunts, and determining a distribution flow weight value borne by each common pump corresponding to the key pump according to the flow interaction quantity and the distance between the key pump and the common pump corresponding to the key pump;
obtaining the flow rate to be distributed by the key pump through the abnormal coefficient of the key pump and the flow rate carried by the key pump;
determining a distribution scheme for distributing the flow to each corresponding common pump by the key pump according to the distribution flow to be distributed by the key pump;
calculating the abnormal degree of the distribution scheme according to the flow distributed by each common pump in each distribution scheme and the distribution flow weight carried by the common pump;
and selecting the minimum value of the abnormal degree of all the obtained distribution schemes, wherein the distribution scheme corresponding to the minimum value of the abnormal degree is an optimal scheme, carrying out manual overhaul on the key pump when the minimum value of the abnormal degree does not belong to a set threshold range, and distributing the flow to be distributed by the key pump according to the optimal distribution scheme when the minimum value of the abnormal degree is in the set threshold range.
Further, the method for detecting the equipment fault based on the intelligent internet of things further comprises the following steps after the distribution is completed:
and detecting a temperature curve and a vibration curve of the key machine pump after distribution is finished, and when an abnormal section of the vibration curve occurs, calculating an abnormal coefficient of the abnormal section of the vibration curve, and further distributing the flow of the key machine pump.
Further, in the method for detecting the equipment fault based on the intelligent internet of things, the method for acquiring the abnormal section of the vibration curve in the abnormal time period when the key pump is abnormal is as follows:
if the key pump has continuous abnormal periods, counting an abnormal curve in a set time period from the first abnormal period, and taking the abnormal curve as an abnormal section of a vibration curve in the abnormal time period.
Further, in the method for detecting the equipment fault based on the intelligent internet of things, the expression of the abnormal coefficient of the abnormal section of the vibration curve is as follows:
in the formula: gamma represents the abnormal coefficient of the key pump in the abnormal section of the vibration curve, alpha represents the alpha-th vibration period in the abnormal stage, n represents the vibration period number in the abnormal stage, b α The difference value between the vibration amplitude value of the key pump in the alpha-th vibration period and a standard value is shown,
the average value of the vibration amplitude difference values of all vibration periods in the abnormal section of the vibration curve and the vibration amplitude difference values of the standard period is represented, m (n) represents the permutation entropy of the vibration amplitude difference values, and H (n) represents the permutation entropy of the temperature amplitude difference values.
Further, in the method for detecting the equipment fault based on the intelligent internet of things, the method for determining the distribution flow weight value carried by each common pump corresponding to the key pump is that:
and obtaining a flow distance ratio through the flow interaction amount and the distance between the key pump and the corresponding common pump, and obtaining a distribution flow weight value carried by each common pump corresponding to the key pump and distributed to the key pump through the flow distance ratio corresponding to the key pump and the total flow interaction amount of the common pump.
Further, in the method for detecting the equipment fault based on the intelligent internet of things, the expression of the flow to be distributed by the key pump is as follows:
C＝c*γ
in the formula: c represents the flow rate to be dispensed by the critical mechanical pump, and C represents the flow rate carried by the critical mechanical pump.
Further, in the method for detecting the equipment fault based on the intelligent internet of things, the expression of the abnormal degree is as follows:
in the formula: q represents the abnormal degree of the key pump and the common pump with flow interaction with the key pump, O represents the number of the common pumps with flow interaction with the key pump, O represents the O-th common pump with flow interaction with the key pump, rho represents a hyper-parameter, i represents the i-th common pump, j represents the j-th common pump, y represents the abnormal degree of the key pump and the common pump with flow interaction with the key pump, O represents the number of the common pumps with flow interaction with the key pump, p represents a hyper-parameter, i represents the i-th common pump, y represents the j-th common pump, and i shows the flow rate after the i-th ordinary pump is dispensed, g (y) i ) Indicating the temperature change, w, caused by an increased flow of the ordinary pump j Represents the assistance capability of the jth common pump and the key pump, deltax i ＝y i -x i Indicating the flow rate, x, dispensed by the ith conventional pump i Indicates the current flow rate, w, of the ith conventional pump i Represents the assistance capability of the ith ordinary pump and the key pump, deltax j ＝y j -x j Indicating the flow rate, y, dispensed by the jth conventional pump j Denotes the flow rate, x, after the j-th ordinary pump is dispensed j Indicating the current flow rate of the jth conventional pump.
Further, in the method for detecting the equipment fault based on the intelligent internet of things, in the flow distribution scheme, the flow distributed to each common pump by the key pump does not exceed the distribution flow weight value carried by the common pump and the distribution flow value obtained by the maximum flow carried by the common pump.
The beneficial effects of the invention are: the method acquires the vibration and temperature data of the real-time operation of the key pump of the equipment by using the sensor, acquires the abnormal condition of the abnormal key pump, performs self-adaptive adjustment and alarm control according to the correlation degree between the key pump and the common pump, can grasp the equipment state in real time, realizes predictive maintenance and repair, reduces over-repair or under-repair to the maximum extent, and can effectively reduce the times of unplanned shutdown and ensure the production benefit; compared with the prior art, the safety accident risk can be reduced to the maximum extent by judging and predicting the equipment state.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a schematic flow chart of an embodiment of an equipment fault detection method based on an intelligent internet of things.
Detailed Description
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.
Example 1
The embodiment of the invention relates to an equipment fault detection method based on an intelligent internet of things, which is shown in fig. 1 and comprises the following steps:
101. acquiring a temperature curve and a vibration curve of each key pump in a pump group;
because coking device has a plurality of machine pumps, the effect diverse that every machine pump played, each machine pump is coordinated with each other and is worked and accomplish the coking task, wherein the position and the control ability that partial pump is located are strong, are in coking device's node position, vibration is comparatively obvious when this kind of machine pump work or take place unusually, the data of gathering through this kind of machine pump more possess the representativeness, when coking device operation is unusual, the feedback that data that this kind of machine pump gathered can be better is unusual. The machine pump is marked as a key machine pump, the vibration obvious parts of a plurality of key machine pumps of the coking device are respectively provided with a detection head, the detection heads are used for detecting the running state parameters of the vibration of the key machine pumps, the detection heads comprise a vibration sensor, a temperature sensor and other required sensors selected according to the types of the machine pumps, for example, a sensor used for detecting the wind speed or the wind volume can be added to the machine pumps such as fans, and the detection heads are only arranged at the vibration obvious parts of the key machine pumps, so that real and effective equipment data such as the vibration condition or the heating condition can be obtained.
And respectively acquiring the temperature data and the vibration data of each key pump node through the detection head of each key pump node to obtain the temperature curve and the vibration curve of each key pump node.
Therefore, the real-time vibration and temperature data of each key pump node are obtained by arranging the detection head at the position where the vibration of the key pump node is obvious.
102. Acquiring a vibration curve abnormal section in an abnormal time period when the key pump is abnormal by using a vibration curve of the key pump;
103. determining the number of abnormal vibration cycles and the vibration amplitude of each abnormal cycle in the abnormal section of the vibration curve by using the abnormal section of the vibration curve;
104. extracting a temperature amplitude corresponding to each abnormal vibration period in a corresponding time period in the temperature curve according to the abnormal section of the vibration curve;
when the key pump works normally, the vibration of the pump usually changes periodically or approximately periodically, when there are some external influences, the vibration frequency of the pump fluctuates in a small range, and the original cycle is disturbed, but the change is the change in a small range, and the vibration cycle is not affected basically, when the key pump works abnormally, for example, when the flow (air flow or liquid flow) rises suddenly or the key pump works for a long time, the load capacity of the pump increases suddenly, or when an abnormality (part failure) occurs in the pump, the vibration cycle is far different from the vibration cycle when the pump works normally, and when the pump works abnormally, the abnormal working time of the pump is obtained, along with the change of the temperature of the pump, the more the abnormal risk of the pump is increased, the more the possibility of causing afterward breakdown is increased, so that the running condition of the pump at the current time is judged according to the vibration curve and the temperature change curve of the pump, the abnormal working time of the pump is obtained, the abnormal severity degree of the abnormal condition is predicted according to carry out load sharing, and the load sharing is obtained by calculating the load sharing method with the minimum influence. The common mechanical pump has low degree of mission criticality, so that the coking process is less influenced when some common mechanical pumps have problems, and the operation state of the key mechanical pump has larger influence on the coking process because the mission criticality carried by the key mechanical pump is critical. Therefore, the parameters of the key pump need to be monitored in real time, and the parameters of the common pump can be acquired according to historical data.
In this embodiment, the key pump with abnormal operation needs to be subjected to abnormality judgment and accurate control, and a relatively accurate abnormal operation time and an accurate abnormal level need to be obtained through calculation, and the specific process is as follows:
the minimum vibration period of the key pump is obtained according to the vibration curve, when the key pump works normally, the power and the running state of the key pump are normal, so that the pump vibrates periodically, and at the moment, the minimum vibration period template is obtained through manual marking. However, due to the influence of environmental factors and some external factors (such as ground vibration), the amplitude of the vibration curve of the pump has slight difference, so that the vibration periodicity is not accurate through the calculation of the original vibration data, and at the moment, in order to eliminate the interference of the external factors on the minimum period acquisition, the original vibration data needs to be modified, and the fault-tolerant range is increased. And (3) adding a fault tolerance range to the data at each moment in the period by taking the standard minimum vibration period template as a reference, namely adding a certain value in the template as a, and adding a value after the fault tolerance range as a +/-beta, wherein beta is the fault tolerance range, namely modifying the real-time data into a when the value of the corresponding moment in the corresponding period in the implementation data is within the range of a +/-beta, and if the value is not within the range, retaining the original data to obtain a modified vibration curve of the corresponding moment.
In the process, the vibration of the key pump is considered to be periodic when the key pump works normally, a template with the minimum vibration period is obtained, the vibration data of the key pump is monitored in real time, the real-time data is compared with the template data, and whether the key pump works abnormally or not is monitored; meanwhile, in order to eliminate the influence of external factors on the monitoring result, a fault tolerance range is added to the data, for example, if the vibration degree of the real-time monitored vibration data is a ', the vibration degree value corresponding to the vibration period of the template is a, the fault tolerance range is considered, if a-beta is not less than a' and not more than a + beta, the vibration degree at the moment is modified into a, and if a 'is not between [ a-beta, a + beta ], the vibration degree at the moment is kept a'.
When the real-time data has a vibration value exceeding the fault-tolerant range, marking the current moment (vibration period), recording the vibration value of the next vibration period starting from the vibration period, and if the vibration value of the next vibration period is also beyond the fault-tolerant range, monitoring the working state of the key machine pump mainly at the moment.
105. Calculating the abnormal coefficient of the abnormal section of the vibration curve according to the difference value of the vibration amplitude of each abnormal period in the abnormal section of the vibration curve and the vibration amplitude of the standard period, the average value of the vibration amplitudes of all the vibration periods in the abnormal section of the vibration curve and the vibration amplitude difference value of the standard period and the temperature amplitude corresponding to each abnormal vibration period;
the expression of the abnormal coefficient of the key pump in the abnormal stage is as follows:
in the formula: gamma is the abnormal coefficient of the key pump in the abnormal stage, alpha is the alpha-th vibration period of the abnormal stage, n is the vibration period number of the abnormal stage, b α The difference value between the vibration amplitude value of the key pump in the alpha-th vibration period and a standard value is shown,
the average value of the vibration amplitude difference values of all vibration periods in the abnormal section of the vibration curve and the vibration amplitude difference values of the standard period is represented, m (n) represents the permutation entropy of the vibration amplitude difference values, and H (n) represents the permutation entropy of the temperature amplitude difference values.
The arrangement entropy m (n) of the vibration amplitude difference value is a difference value sequence (b) 1 ，b 2 ，b 3 …b n ) Degree of disorder, difference b in the difference sequence n When the linear growth is performed, the disorder degree of the sequence is low, namely the abnormal degree of the corresponding abnormal stage is serious, namely the degree of deviation of the operation condition of the next moment from the normal operation is larger than that of the previous moment; the principle of the temperature amplitude difference value permutation entropy H (n) is consistent with that of the vibration amplitude difference value permutation entropy, and the temperature amplitude difference value sequence is (t) 1 ，t 2 ，t 3 …t n ) The permutation entropy, which is the prior art, the calculation method is not outlined here.
The larger the value of the abnormal coefficient gamma obtained by calculation is, the closer the value is to 1, which shows that the state trend of the machine pump is worse and worse when the key machine pump continues to operate according to the operation state, and the larger the value of the abnormal coefficient gamma is, the higher the probability of emergency repair caused by serious failure of the machine pump is.
The process shows that only when the vibration value exceeds the fault-tolerant range in the continuous vibration period, the part of the vibration period is taken as an abnormal stage, the abnormal coefficient of the key pump in the abnormal stage is calculated through the vibration amplitude difference, the vibration amplitude difference arrangement entropy and the temperature amplitude difference arrangement entropy of the key pump in the abnormal stage, and the abnormal degree of the key pump is determined according to the abnormal coefficient.
106. Acquiring common pumps corresponding to abnormal key pump shunts, and determining a distribution flow weight value borne by each common pump corresponding to the key pump according to the flow interaction quantity and the distance between the key pump and the common pump corresponding to the key pump;
acquiring the control capability of each pump: when the pump works, different power of the pump is different, the controllable flow is different (the flow is intake air, liquid flow and the like, and is called flow in general), the control function of the key pump is stronger than that of the common pump, therefore, the control capability of each pump is graded into 10 grades, the higher the grade is, the stronger the control effect is, the more the flow is, and the larger the key degree is.
Analyzing the relation between the common pump and the key pump, and obtaining the assistance capability between the common pump and the key pump (namely the allocation flow weight value carried by each common pump corresponding to the key pump allocated by the key pump): flow interaction exists between the common pump and the key pump, the larger the flow interaction amount is, the fastest the adjustment effect on the associated pump can be achieved by changing the flow control parameter, and the larger the association degree between the common pump and the key pump is, the ratio of the flow interaction amount to the distance between a certain common pump and the key pump and the ratio of the flow interaction amount between the common pump and all pumps are obtained (the larger the flow interaction between the common pump and the key pump is and the closer the flow interaction is, the faster the flow pressure of the key pump can be shared), the ratio is used as the assistance capability of the common pump and the key pump, the size of the capability of the common pump assisting the key pump is represented, the larger the value is, the stronger the assistance capability is represented, and the pressure of the key pump can be shared more, so that the normal operation can be recovered as soon as possible.
According to the distribution condition of the pumps, the pumps are taken as nodes, the nodes are divided into key nodes and common nodes (the key pumps are the key nodes, and the common pumps are the common nodes), and the edge weight between the key nodes and the common nodes is the assistance capability between the key pumps and the common pumps, so that graph structure data G1 is constructed. The value of the pump node is a vector formed by the pump control capacity and the abnormal coefficient.
107. Obtaining the flow rate to be distributed by the key pump through the abnormal coefficient of the key pump and the flow rate carried by the key pump;
when the key node is abnormal, tasks of the key node are expected to be distributed to related common nodes, namely the control capability of the key pump node is expected to be reduced, and the control quantity is reduced. Each control level corresponds to a control quantity, and the flow rate required to be distributed when the key pump works abnormally is as follows:
C＝c*γ
in the formula: c represents the flow rate which needs to be distributed by the key machine pump, C represents the flow rate (namely the flow rate of the load) controlled by the key machine pump, gamma represents an abnormal coefficient, and the larger the abnormal degree is, the larger the flow rate which needs to be distributed is.
When the key machine pump is abnormal, the flow needing to be distributed is determined through the formula, and after part of flow is distributed, the flow controlled by the key machine pump is reduced, so that the pressure of the key machine pump is relieved, and the influence of equipment shutdown on production benefits can be avoided.
108. Determining a distribution scheme for distributing the flow to each corresponding common pump by the key pump according to the distribution flow to be distributed by the key pump;
109. calculating the abnormal degree of the distribution scheme according to the flow distributed by each common pump in each distribution scheme and the distribution flow weight carried by the common pump;
110. and selecting the minimum value of the abnormal degree of all the obtained distribution schemes, carrying out manual maintenance on the key pump when the minimum value of the abnormal degree does not belong to the range of the set threshold, and distributing the flow to be distributed by the key pump according to the optimal distribution scheme when the minimum value of the abnormal degree is in the range of the set threshold.
Under the normal condition, the stronger the assisting capability of the common node and the key node is, the more the task amount is allocated, but the load amount of the common pump increases due to the allocation according to the method, so that the common pump is abnormal, at the moment, the tasks of the key node need to be allocated to the common node to ensure that the key node returns to normal, and the abnormal degree of all nodes is ensured to be minimum, so that an abnormal degree model of the key pump and the common pump with flow interaction with the key pump is established, and the model is as follows:
in the formula: q represents the abnormal degree of the key pump and the common pump with flow interaction with the key pump, O represents the number of the common pumps with flow interaction with the key pump, O represents the O-th common pump with flow interaction with the key pump, rho represents a super parameter, the empirical value is 1,i to represent the i-th common pump, j represents the j-th common pump, y represents the flow interaction with the key pump, and i denotes the flow rate after the i-th ordinary pump is dispensed, g (y) i ) The temperature change caused by the increase of the flow rate of the common pump can be obtained through historical data (namely a temperature curve table corresponding to different flow rates of the common pump), and w j Represents the side weight value, deltax, of the jth common pump and the key pump i ＝y i -c i Indicating the flow rate, x, dispensed by the ith conventional pump i Indicates the current flow rate, w, of the ith conventional pump i Represents the side weight value, deltax, of the ith normal pump and the key pump j ＝y j -x j Indicating the flow rate, y, dispensed by the jth conventional pump j Denotes the flow rate, x, after the j-th ordinary pump is dispensed j Indicating the current flow rate of the jth conventional pump.
In the above formula,. DELTA.x 1 +...Δx i ...+...Δx j +...Δx O C, Q represents the abnormal degree of the key pump and the common pump with flow interaction with the key pump, the value of the abnormal degree Q is expected to be minimum by the system, and the flow distributed to the common pump does not exceed the distributed flow value obtained by the distributed flow weight value carried by the common pump and the maximum flow carried by the common pump, at the moment, the distributed flow can reduce the burden of the key node, and can be recovered to be normal as soon as possible, and the abnormal degree of the common pump can also be minimum. And solving by adopting a gradient descent method to obtain the optimal distribution amount and the minimum abnormal degree.
The optimal distribution amount is solved through the model, so that the abnormal degree Q obtained after the flow is distributed by each common pump is as small as possible, namely, the partial flow of the key pump is distributed, and meanwhile, the common pumps can normally work in respective load ranges.
The method comprises the steps of carrying out grade division according to the abnormal degree Q values of all nodes related to key nodes obtained by solving, namely when the Q value is within a certain interval range, the abnormity of the key pump can be adjusted according to a related common pump, when the Q value exceeds a threshold value range, the risk caused by adjusting the flow of the related common pump is large, namely the risk of the whole system is large due to self-adaptive adjustment of the system, alarming is carried out when the self-adaptive adjustment of the system is not applicable, and the abnormal key pump is manually adjusted and overhauled through alarming prompt to return to a normal operation state.
And performing manual spot inspection on the abnormal key pump according to the abnormal coefficient of the pump in the abnormal stage, detecting the reason of the abnormal operation of the key pump, providing a corresponding adjusting method, verifying the manual adjusting method and the system self-adaptive adjusting method, and adjusting the hyperparameter rho if a plurality of spot inspection results have larger difference with the system self-adaptive adjusting result, so that the difference between the self-adaptive adjusting result and the manual spot inspection adjusting method is as small as possible.
The beneficial effects of the invention are: the method acquires the vibration and temperature data of the real-time operation of the key pump of the equipment by using the sensor, acquires the abnormal condition of the abnormal key pump, performs self-adaptive adjustment and alarm control according to the correlation degree between the key pump and the common pump, can grasp the equipment state in real time, realizes predictive maintenance and repair, reduces over-repair or under-repair to the maximum extent, and can effectively reduce the times of unplanned shutdown and ensure the production benefit; compared with the prior art, the safety accident risk can be reduced to the maximum extent by judging and predicting the equipment state.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and should not be taken as limiting the scope of the present invention, which is intended to cover any modifications, equivalents, improvements, etc. within the spirit and scope of the present invention.
Claims (7)
1. An equipment fault detection method based on an intelligent Internet of things is characterized by comprising the following steps:
acquiring a temperature curve and a vibration curve of each key pump in a pump group;
acquiring a vibration curve abnormal section in an abnormal time period when the key pump is abnormal by using a vibration curve of the key pump;
determining the number of abnormal vibration periods and the vibration amplitude of each abnormal period in the abnormal section of the vibration curve by using the abnormal section of the vibration curve;
extracting a temperature amplitude corresponding to each abnormal vibration period in a corresponding time period in the temperature curve according to the abnormal section of the vibration curve;
calculating the abnormal coefficient of the abnormal section of the vibration curve according to the difference value of the vibration amplitude of each abnormal period in the abnormal section of the vibration curve and the vibration amplitude of the standard period, the average value of the vibration amplitudes of all the vibration periods in the abnormal section of the vibration curve and the vibration amplitude difference value of the standard period and the temperature amplitude corresponding to each abnormal vibration period; the expression of the abnormal coefficient of the abnormal section of the vibration curve is as follows:
in the formula: gamma is the abnormal coefficient of the key pump in the abnormal section of the vibration curve, alpha is the alpha-th vibration period of the abnormal stage, n is the vibration period number of the abnormal stage, b α The difference value between the vibration amplitude value of the key pump in the alpha-th vibration period and a standard value is shown,
representing the average value of the vibration amplitude difference values of all vibration periods in the abnormal section of the vibration curve and the vibration amplitude difference values of the standard periods, wherein m (n) represents the permutation entropy of the vibration amplitude difference values, and H (n) represents the permutation entropy of the temperature amplitude difference values; acquiring common pumps corresponding to abnormal key pump shunts, and determining a distribution flow weight value borne by each common pump corresponding to the key pump according to the flow interaction quantity and the distance between the key pump and the common pump corresponding to the key pump;
obtaining the flow rate to be distributed by the key pump through the abnormal coefficient of the key pump and the flow rate carried by the key pump;
determining a distribution scheme for distributing the flow to each corresponding common pump by the key pump according to the distribution flow to be distributed by the key pump;
calculating the abnormal degree of the distribution scheme according to the flow distributed by each common pump in each distribution scheme and the distribution flow weight carried by the common pump;
and selecting the minimum value of the abnormal degree of all the obtained distribution schemes, wherein the distribution scheme corresponding to the minimum value of the abnormal degree is the optimal scheme, when the minimum value of the abnormal degree does not belong to the range of the set threshold value, manually overhauling the key pump, and when the minimum value of the abnormal degree is in the range of the set threshold value, distributing the flow to be distributed by the key pump according to the optimal distribution scheme.
2. The method for detecting the equipment fault based on the intelligent Internet of things according to claim 1, wherein after the allocation is completed, the method further comprises the following steps:
and detecting a temperature curve and a vibration curve of the key machine pump after distribution is finished, and when an abnormal section of the vibration curve occurs, calculating an abnormal coefficient of the abnormal section of the vibration curve, and further distributing the flow of the key machine pump.
3. The equipment fault detection method based on the intelligent Internet of things according to claim 1, wherein the method for acquiring the abnormal section of the vibration curve in the abnormal time period when the key pump is abnormal is as follows:
if the key pump has continuous abnormal periods, counting an abnormal curve in a set time period from the first abnormal period, and taking the abnormal curve as an abnormal section of a vibration curve in the abnormal time period.
4. The method for detecting the equipment fault based on the intelligent Internet of things according to claim 1, wherein the method for determining the distribution flow weight value carried by each common pump corresponding to the key pump is as follows:
and obtaining a flow distance ratio through the flow interaction quantity and the distance between the key pump and the corresponding common pump, and obtaining a distribution flow weight value carried by each corresponding common pump distributed to the key pump by the key pump through the flow distance ratio corresponding to the key pump and the total flow interaction quantity of the common pumps.
5. The intelligent Internet of things-based equipment fault detection method according to claim 1, wherein the expression of the flow to be distributed by the key pump is as follows:
C＝c*γ
in the formula: c represents the flow rate to be dispensed by the critical mechanical pump, and C represents the flow rate carried by the critical mechanical pump.
6. The intelligent Internet of things-based equipment fault detection method according to claim 1, wherein the expression of the abnormal degree is as follows:
in the formula: q represents the abnormal degree of the key pump and the common pump with flow interaction with the key pump, O represents the number of the common pumps with flow interaction with the key pump, O represents the O-th common pump with flow interaction with the key pump, rho represents a hyper-parameter, i represents the i-th common pump, j represents the j-th common pump, y represents the abnormal degree of the key pump and the common pump with flow interaction with the key pump, O represents the number of the common pumps with flow interaction with the key pump, p represents a hyper-parameter, i represents the i-th common pump, y represents the j-th common pump, and i shows the flow rate after the i-th ordinary pump is dispensed, g (y) i ) Indicating the temperature change, w, caused by an increased flow of the ordinary pump j Represents the assistance capability of the jth common pump and the key pump, deltax i ＝y i -x i Indicating the flow rate, x, dispensed by the ith conventional pump i Indicates the current flow rate, w, of the ith conventional pump i Represents the assistance capability of the ith ordinary pump and the key pump, deltax j ＝y j -x j Indicating the flow rate, y, assigned by the jth conventional pump j Denotes the flow rate, x, after the j-th ordinary pump is dispensed j Indicating the current flow rate of the jth conventional pump.
7. The method for detecting the equipment fault based on the intelligent internet of things as claimed in claim 1, wherein in the flow distribution scheme, the flow distributed to each ordinary pump by the key pump in the key pump does not exceed a distribution flow value obtained by a distribution flow weight carried by the ordinary pump and a maximum flow carried by the ordinary pump.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210627061.5A CN114962239B (en) | 2022-06-01 | 2022-06-01 | Equipment fault detection method based on intelligent Internet of things |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210627061.5A CN114962239B (en) | 2022-06-01 | 2022-06-01 | Equipment fault detection method based on intelligent Internet of things |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114962239A CN114962239A (en) | 2022-08-30 |
| CN114962239B true CN114962239B (en) | 2023-03-24 |
Family
ID=82958875
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210627061.5A Active CN114962239B (en) | 2022-06-01 | 2022-06-01 | Equipment fault detection method based on intelligent Internet of things |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114962239B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117516670B (en) * | 2024-01-04 | 2024-03-29 | 山东晨晖电子科技有限公司 | Intelligent data monitoring method for ultrasonic water meter of Internet of things |
| CN117846941A (en) * | 2024-02-22 | 2024-04-09 | 江苏冠德昇数字科技有限公司 | Remote intelligent control system of air compressor |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2013140135A (en) * | 2011-12-09 | 2013-07-18 | Tokyo Electron Ltd | Abnormality detection apparatus for periodic driving system, processing apparatus including periodic driving system, abnormality detection method for periodic driving system, and computer program |
| US10662959B2 (en) * | 2017-03-30 | 2020-05-26 | General Electric Company | Systems and methods for compressor anomaly prediction |
| CN110008350A (en) * | 2019-03-06 | 2019-07-12 | 杭州哲达科技股份有限公司 | A kind of pump Ankang knowledge base lookup method based on Bayesian inference |
| CN114135477B (en) * | 2021-10-11 | 2024-04-02 | 昆明嘉和科技股份有限公司 | Dynamic threshold early warning method for monitoring state of machine pump equipment |
-
2022
- 2022-06-01 CN CN202210627061.5A patent/CN114962239B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN114962239A (en) | 2022-08-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114962239B (en) | Equipment fault detection method based on intelligent Internet of things | |
| CN110298455B (en) | Mechanical equipment fault intelligent early warning method based on multivariate estimation prediction | |
| CN110320892B (en) | Sewage treatment equipment fault diagnosis system and method based on L asso regression | |
| CN112784373B (en) | Fault early warning method for wind turbine generator gearbox | |
| CN101799320B (en) | Fault prediction method and device thereof for rotation equipment | |
| CN111679952B (en) | Alarm threshold generation method and device | |
| US20040049715A1 (en) | Computer networked intelligent condition-based engine/equipment management system | |
| CN110378042B (en) | Wind turbine generator gearbox oil temperature anomaly detection method and system based on SCADA data | |
| CN112132394B (en) | Power plant circulating water pump predictive state evaluation method and system | |
| CN112836941B (en) | Online health condition assessment method for high-pressure system of steam turbine of coal motor unit | |
| CN112861350B (en) | Temperature overheating defect early warning method for stator winding of water-cooled steam turbine generator | |
| CN111103137A (en) | Wind turbine gearbox fault diagnosis method based on deep neural network | |
| CN116976857B (en) | Predictive maintenance method and system for railway power equipment based on deep learning | |
| Abbasi et al. | Development of predictive maintenance interface using multiple linear regression | |
| CN114925718A (en) | Transformer core grounding current online prediction method and device, storage medium and fault diagnosis method | |
| CN115131946A (en) | Adaptive equipment alarm method and system considering working condition fluctuation | |
| CN113780375A (en) | Virtual-real interaction wind power plant wind power monitoring system based on digital twins and application | |
| CN113539382B (en) | Early warning positioning method and system for key technological parameters of dimethyl phosphite | |
| CN117523793A (en) | Power plant equipment fault early warning method and computer equipment | |
| CN116050613B (en) | Method for predicting residual service life of top cover drainage pump of hydroelectric generating set | |
| CN112668794A (en) | Maintenance optimization method for aircraft engine under uncertain life prediction information | |
| CN201637483U (en) | Rotary equipment failure prediction module | |
| CN113632025A (en) | Methods, systems, and computer program products for assessing energy consumption in an industrial environment | |
| CN113955149B (en) | Health diagnosis method and device for motor system | |
| CN114254834A (en) | Wind power plant wind speed prediction method based on machine learning |
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 |