CN113361088B - Low-energy-consumption high-speed rail wheel shaft temperature monitoring and diagnosing method based on edge calculation - Google Patents

Low-energy-consumption high-speed rail wheel shaft temperature monitoring and diagnosing method based on edge calculation Download PDF

Info

Publication number
CN113361088B
CN113361088B CN202110605767.7A CN202110605767A CN113361088B CN 113361088 B CN113361088 B CN 113361088B CN 202110605767 A CN202110605767 A CN 202110605767A CN 113361088 B CN113361088 B CN 113361088B
Authority
CN
China
Prior art keywords
cloud computing
temperature data
edge
computing center
shaft temperature
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
Application number
CN202110605767.7A
Other languages
Chinese (zh)
Other versions
CN113361088A (en
Inventor
汪木兰
周雪
刘婷婷
包永强
潘超
贾茜
余卓
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.)
Nanjing Institute of Technology
Original Assignee
Nanjing Institute of Technology
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 Nanjing Institute of Technology filed Critical Nanjing Institute of Technology
Priority to CN202110605767.7A priority Critical patent/CN113361088B/en
Publication of CN113361088A publication Critical patent/CN113361088A/en
Application granted granted Critical
Publication of CN113361088B publication Critical patent/CN113361088B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/08—Probabilistic or stochastic CAD
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/06—Power analysis or power optimisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/10—Noise analysis or noise optimisation

Abstract

The invention provides a low-energy-consumption high-speed rail wheel axle temperature monitoring and diagnosing method based on edge calculation, and provides a low-energy-consumption axle temperature data monitoring and diagnosing scheme utilizing edge calculation, aiming at the condition that the time delay of uploading all real-time axle temperature data of a wheel set to a cloud computing is large and the efficiency of a temperature monitoring system during high-speed rail running is influenced. According to the scheme, the axle temperature monitoring data is divided into two parts, one part of the axle temperature data is sent to the cloud computing center for computing, the other part of the axle temperature data is sent to the edge nodes (a plurality of edge computing devices) for computing, low-delay and low-energy-consumption monitoring and diagnosis are achieved, the real-time performance of wheel set axle temperature data monitoring is improved, and therefore the efficiency of a wheel set axle temperature monitoring system is improved.

Description

Low-energy-consumption high-speed rail wheel shaft temperature monitoring and diagnosing method based on edge calculation
Technical Field
The invention belongs to the research fields of edge calculation, axle temperature data monitoring and diagnosis, industrial internet, network resource allocation and the like, and particularly relates to a low-energy-consumption high-speed rail wheel axle temperature monitoring and diagnosis method based on edge calculation.
Background
The rail transit industry has become one of the most important modern urban trips, and a safe, quick, convenient and comfortable trip mode is favored by people. As China enters the high-speed rail era, the speed of the train is greatly improved, and the speed of the high-speed rail reaches 350 km/h. Due to the fact that the running speed is increased, the traction power is increased, impact, power effect and vibration of the high-speed rail and the steel rail are increased, and heating of the train shape-walking portion and the gear box is increased. When the axle and the pinion gear are worn or defective, abnormal heat generation increases. When the temperature exceeds a limit, phenomena such as a hot shaft, a shaft cutting and the like can be caused, and the operation safety of the high-speed rail is seriously damaged. The wheel set shaft temperature data abnormity diagnosis has great significance for timely finding potential safety hazards to guarantee normal operation of high-speed rails. The real-time monitoring data volume of the axle temperature is continuously increased, so that the problems of network transmission congestion, increased calculation complexity and the like are caused, the time delay and energy consumption of the axle temperature data in transmission and calculation are increased, the diagnosis time is increased, and the like, so that the running safety of a high-speed rail faces risks. By utilizing the advantages of edge calculation, the time delay and energy consumption of the axle temperature data diagnosis are reduced, the diagnosis capability of the axle temperature data is improved, and the timeliness of monitoring and diagnosis can be ensured.
Disclosure of Invention
The invention provides a low-energy-consumption high-speed rail wheel axle temperature monitoring and diagnosing method based on edge calculation, aiming at the condition that the time delay of uploading all wheel set real-time axle temperature data to a cloud computing is large and the efficiency of an axle temperature monitoring system during high-speed rail running is influenced in the prior art.
In order to achieve the purpose, the invention adopts the following technical scheme:
a low-energy-consumption high-speed rail wheel-to-shaft temperature monitoring and diagnosing method based on edge calculation is used in a high-speed rail wheel-to-shaft temperature monitoring system for performing shaft temperature diagnosis through data acquisition, and is characterized by comprising the following steps:
s1: establishing a model combining edge computing and cloud computing;
s2: based on the model established in step S1, respectively sending the axle temperature data to the edge nodes (a plurality of edge computing devices) and the cloud computing center for computation, so as to monitor and diagnose the axle temperature data; and the high-speed rail wheel shaft temperature monitoring system completes fault diagnosis on shaft temperature data by combining the edge nodes and the calculation result fed back by the cloud calculation center.
In order to optimize the technical scheme, the specific measures adopted further comprise:
further, step S1 is specifically as follows:
s1.1: when the axle temperature of the high-speed rail wheel pair is monitored, the axle temperature data of the high-speed rail wheel pair is divided into two parts, one part is distributed to a plurality of edge nodes for edge calculation, and the other part is distributed to a cloud calculation center for cloud calculation;
s1.2: respectively calculating energy consumed by edge computing and energy consumed by data transmission to a cloud computing center;
s1.3: and establishing a minimum total energy consumption problem and solving to obtain the optimal shaft temperature data distribution proportion, namely the optimal calculation task distribution proportion according to the energy consumed by the edge calculation and the energy consumed by the data transmission to the cloud calculation center.
Further, in step S1.1, a part of the axle temperature data, that is, the computing tasks, is allocated to n edge nodes that are close to the data source and have weak computing power for computing, and another part of the axle temperature data, that is, the computing tasks, is allocated to a cloud computing center with a number of n +1 that has strong computing power and is far from the data source for computing.
Further, in step S1.2, the energy consumption of the edge node and the cloud computing center is obtained by computing the computing delay of the edge node and the communication delay of the cloud computing center, which is specifically as follows:
setting the calculation time delay t of the shaft temperature data on n edge nodesiComprises the following steps:
Figure BDA0003092006630000021
wherein, ViMiddle V1...VnCalculating the rate of n edge nodes, wherein the unit is bit/s; b is the total calculated amount of the wheel set for monitoring and diagnosing the axle temperature data, and the unit is bit;
the channel for transmitting the shaft temperature data to the cloud computing center is a Rayleigh channel, and the probability density function p (y) of Rayleigh distribution has the expression:
Figure BDA0003092006630000022
wherein y is Rayleigh amplitude and y is more than 0, sigma2Is the noise power;
interruption of a memoryProbability of ProutThe expression is as follows:
Figure BDA0003092006630000023
where Pr () represents a probability function, gamma0Cutting off the channel amplitude corresponding to the channel gain;
the average propagation delay t of the successful transmission of the shaft temperature data to the cloud computing center with the number of n +1n+1Comprises the following steps:
Figure BDA0003092006630000031
wherein alpha isn+1The proportion of the computing tasks allocated to the cloud computing center is calculated; r isn+1The transmission rate of the shaft temperature data transmitted to the cloud computing center with the number of n +1 is bit/s; since the interruption probability is PrtAnd thus the probability of successful transmission is 1-ProutThe average speed of the shaft temperature data transmitted to the cloud computing center through the Rayleigh channel is
Figure BDA0003092006630000032
Unit is bit/s, using Vn+1To represent
Figure BDA0003092006630000033
Establishing a computational energy consumption expression L of n edge nodesiComprises the following steps:
Li＝Pi·αiB·ti，i＝1，2，...，n
wherein, PiCalculating the power to be consumed by unit bit for the ith edge node, wherein the unit is W/bit; alpha is alphaiCalculating a task proportion for the ith edge node; t is tiCalculating time delay of the shaft temperature data on the ith edge node, wherein the unit is s;
establishing a total energy consumption expression L transmitted to the cloud computing center with the number of n +1n+1Comprises the following steps:
Ln+1＝Pn+1·αn+1B·tn+1
wherein, Pn+1Transmitting the power consumed by unit bit to the cloud computing center with the number of n +1 for the data source, wherein the unit is W/bit; alpha is alphan+1The proportion of the computing tasks allocated to the cloud computing center is calculated; t is tn+1The unit is s for the average propagation delay of the successful transmission of the shaft temperature data to the cloud computing center with the number of n + 1.
Further, in step S1.3, the objective function L that minimizes the total energy consumption is finally established as:
Figure BDA0003092006630000034
wherein j is 1, 2., n, n +1, the first n represents an edge node, and the n +1 represents a cloud computing center; pi·αiB·tiRepresents the energy consumption of the ith edge node, i ═ 1, 2. Pn+1·αn+1B·tn+1The energy consumption of the cloud computing center is represented by the number n +1 and the unit is W;
the time delay t will be calculatediAnd propagation delay tn+1The problem of minimizing the total energy consumption is:
Figure BDA0003092006630000041
Figure BDA0003092006630000042
the optimization problem meets the KKT condition, a Lagrange equation is established, the Lagrange multiplier method is used for solving, and the optimal task proportion alpha for minimizing the total energy consumption is obtained through solvingj *Comprises the following steps:
Figure BDA0003092006630000043
optimal task proportion alpha allocated to edge nodesi *Expressed as:
Figure BDA0003092006630000044
wherein alpha isi *Calculating the task proportion of the optimal shaft temperature data distributed to the edge node i;
optimal task proportion alpha distributed to cloud computing centern+1 *Expressed as:
Figure BDA0003092006630000045
wherein alpha isn+1 *The optimal shaft temperature data is distributed to the cloud computing center to calculate the task proportion.
Further, in step S2, the optimal task ratio α obtained by calculation in step S1.3 is obtainedj *And respectively sending the shaft temperature data of the corresponding part to n edge nodes and the cloud computing center with the number of n +1 for diagnosis and computation.
The invention has the beneficial effects that: according to the low-energy-consumption high-speed rail wheel axle temperature monitoring and diagnosing method based on the edge calculation, the low-energy-consumption axle temperature data diagnosing scheme is designed by utilizing the edge calculation, the monitoring and the diagnosis with low time delay and low energy consumption can be realized, the real-time performance of the wheel set axle temperature data monitoring is improved, and the axle temperature data is processed efficiently, so that the efficiency of a wheel set axle temperature monitoring system is improved.
Drawings
FIG. 1 is a schematic diagram of a diagnostic process for a low energy and high rail wheel axle temperature monitoring system.
FIG. 2 is a schematic diagram of comparing energy consumption of the computing task allocation method and the uniform allocation method according to the present invention.
Detailed Description
The present invention will now be described in further detail with reference to the accompanying drawings.
The low-energy-consumption high-speed rail wheel-to-shaft temperature monitoring and diagnosing method based on edge calculation shown in fig. 1 is used in a high-speed rail wheel-to-shaft temperature monitoring system for performing shaft temperature diagnosis through data acquisition, and mainly comprises the following steps:
the method comprises the steps that a part of computing tasks for shaft temperature data diagnosis are distributed to n pieces of edge computing equipment (namely edge nodes) which are close to a data source and have common computing capacity by a wheel pair by utilizing the advantages of an edge computing framework for computing, the other part of computing tasks are handed to a cloud computing center (namely a cloud platform) which has strong computing capacity and is far away from a terminal (data source) for computing, and the computing tasks are distributed to the edge nodes and the cloud computing center according to the optimal proportion calculated by taking the minimum energy consumption as a target in the shaft temperature monitoring process of the high-speed rail wheel, so that diagnosis is completed, and the total energy consumption is minimized.
The following is an explanation of the process and principles of the method of the present invention, taking as an example the axle temperature monitoring data for diagnosing wheel pairs.
1) And establishing a model combining edge computing and cloud computing during wheel set axle temperature monitoring and diagnosis.
Specifically, when the wheel set carries out monitoring diagnosis on the axle temperature data, the total calculated amount for defining the diagnosis of the axle temperature data is B (bit). At the moment, a mode of combining edge computing and a cloud computing center is adopted, wherein the number of edge nodes is 1 to n respectively, and the number of the cloud computing center is n + 1. The ratio of the calculated quantity of n edge nodes and the cloud computing center with the number of n +1 in the shaft temperature monitoring diagnosis to the total calculated quantity of the shaft temperature data in the system monitoring diagnosis is alphajJ is 1,2,., n, n + 1. Wherein the task proportion calculated by distributing to n edge nodes is alphaiRepresents, i 1,2,. n; alpha is used for distributing task proportion calculated by cloud computing centern+1To indicate.
Specifically, the energy consumed by the edge computing and the energy consumed by the cloud computing center are calculated. The computing time delay of the cloud computing center can be ignored under the assumption that the computing capacity of the cloud computing center is infinite. The shaft temperature data is transmitted to the cloud computing center through a wireless channel, and the wireless channel is a rayleigh channel, so that the influence of the wireless channel on the transmission of the shaft temperature data to the cloud computing center with the number n +1 needs to be considered. Because the edge node is close to the data source, the propagation delay can be ignored, but because the edge node has weak calculation capability, the calculation delay needs to be considered.
And obtaining the total energy consumed by the computing task in the processes of propagation and computing by giving the computing time delay of the edge node and the communication time delay of the data source propagated to the cloud computing center.
Setting the calculation time delay t of the shaft temperature data on n edge nodesiComprises the following steps:
Figure BDA0003092006630000051
wherein, ViMiddle V1...VnCalculating the rate of n edge nodes, wherein the unit is bit/s; and B is the total calculated amount of the wheel set in bit when the wheel set carries out axle temperature data monitoring diagnosis.
The channel for transmitting the shaft temperature data to the cloud computing center is a Rayleigh channel, and the probability density function p (gamma) of Rayleigh distribution is expressed as follows:
Figure BDA0003092006630000061
wherein y is a Rayleigh amplitude and y is greater than 0; sigma2Is the noise power.
Interruption probability of ProutThe expression is as follows:
Figure BDA0003092006630000062
where Pr () represents a probability function, gamma0Cutting off the channel amplitude corresponding to the channel gain; sigma2Is the noise power.
The average propagation delay t of the successful transmission of the shaft temperature data to the cloud computing center with the number of n +1n+1Comprises the following steps:
Figure BDA0003092006630000063
wherein alpha isn+1The proportion of the computing tasks allocated to the cloud computing center is calculated; r isn+1The transmission rate of the shaft temperature data transmitted to the cloud computing center with the number of n +1 is bit/s; since the interruption probability is ProutAnd thus the probability of successful transmission is 1-ProutThe average speed of the shaft temperature data transmitted through the Rayleigh channel is
Figure BDA0003092006630000064
The unit is bit/s, and for the sake of simplicity, V is usedn+1To represent
Figure BDA0003092006630000065
Establishing n edge calculation node energy consumption expressions LiComprises the following steps:
Li＝Pi·αiB·ti，i＝1,2，...，n (5)
wherein, PiCalculating the power to be consumed by unit bit for the ith edge node, wherein the unit is W/bit; alpha is alphaiCalculating a task proportion for the ith edge node; t is tiThe unit is s for the calculation time delay of the shaft temperature data on the ith edge node.
Establishing a total energy consumption expression L transmitted to the cloud computing center with the number of n +1n+1Comprises the following steps:
Ln+1＝Pn+1·αn+1B·tn+1 (6)
wherein, Pn+1Transmitting the power consumed by unit bit to the cloud computing center with the number of n +1 for the data source, wherein the unit is W/bit; alpha is alphan+1The proportion of the computing tasks allocated to the cloud computing center is calculated; t is tn+1The unit is s for the average propagation delay of the successful transmission of the shaft temperature data to the cloud computing center with the number of n + 1.
The objective function L that minimizes the total energy consumption is finally established as:
Figure BDA0003092006630000071
wherein j is 1,2,., n, n +1, the first n are edge nodes, and the n +1 is a cloud computing center. Pi·αiB·tiExpressed as the energy consumption of the ith edge node, i ═ 1, 2. Pn+1·αn+1B·tn+1Energy consumption for transmitting the data source to the cloud computing center is represented by the number n +1 and the unit is W/s;
the time delay t will be calculatediAnd propagation delay tn+1The problem of minimizing the total energy consumption is:
Figure BDA0003092006630000072
Figure BDA0003092006630000073
2) the optimization problem meets the KKT condition, a Lagrangian equation is established, and the Lagrangian multiplier method is used for solving, so that:
Figure BDA0003092006630000074
wherein, the above is the established objective function. j is 1,2,., n, n +1, the first n are edge nodes, and the n +1 is a cloud computing center. PiN denotes the power that n edge nodes need to consume to compute a unit bit, Pn+1Propagating the power, ω and β, consumed per bit for a data source to a cloud computing center numbered n +1jIs an auxiliary coefficient in the Lagrange equation.
Solving the problems:
Figure BDA0003092006630000075
therefore, either α j0 or βj＝0。
When beta isjWhen equal to 0, i.e.
Figure BDA0003092006630000081
Solving to obtain:
Figure BDA0003092006630000082
substituting equation (12) into equation (9), i.e.
Figure BDA0003092006630000083
Thereby solving for omega*：
Figure BDA0003092006630000084
By substituting equation (13) into equation (12), the optimum ratio α for the minimum total energy consumption can be solvedj *Comprises the following steps:
Figure BDA0003092006630000085
wherein j is 1,2,., n, n +1, the first n is the number of the edge node, and the n +1 is the number of the cloud computing center.
Optimal task proportion alpha allocated to edge nodesi *Expressed as:
Figure BDA0003092006630000086
wherein, i is 1, 2. Alpha is alphai *The task proportion is calculated for the optimal shaft temperature data distributed to the edge node i.
Optimal task proportion alpha distributed to cloud computing centern+1 *Expressed as:
Figure BDA0003092006630000087
wherein alpha isn+1 *The optimal shaft temperature data is distributed to the cloud computing center to calculate the task proportion.
3) Optimum ratio alpha according to minimum total energy consumptionj *And dividing the wheel set axle temperature monitoring data into two parts, and distributing the two parts to n edge nodes and a cloud computing center to complete corresponding computation, so that the total energy consumption is minimum.
And finally, the high-speed rail wheel-to-shaft temperature monitoring system is combined with the calculation results of the edge calculation part and the cloud calculation part to complete fault diagnosis of the wheel-to-shaft temperature.
Besides monitoring the axle temperature of the wheel pair, the method of the embodiment can also be applied to monitoring the temperature or the axle temperature of other equipment, including monitoring the temperature of a camshaft, monitoring the temperature of a cable and the like.
Fig. 2 is a comparison of energy consumption of the calculation task allocation method and the uniform allocation method provided by the present invention, and it can be seen that the energy consumption varies with the channel amplitude corresponding to the cut-off channel gain. The energy consumption rises with the increase of the channel amplitude corresponding to the cut-off channel gain, but the energy consumption brought by the distribution method provided by the invention is more stable in growth and is obviously lower than that generated by a uniform distribution method.
Compared with the prior art, the method effectively utilizes edge calculation, provides a low-energy-consumption high-speed rail axle temperature monitoring and diagnosing scheme based on the edge calculation, and is suitable for high-timeliness and low-energy-consumption requirements in the monitoring process of axle temperature data of high-speed rail safe operation wheels.
The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above-mentioned embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may be made by those skilled in the art without departing from the principle of the invention.

Claims (3)

1. A low-energy-consumption high-speed rail wheel-to-shaft temperature monitoring and diagnosing method based on edge calculation is used in a high-speed rail wheel-to-shaft temperature monitoring system for performing shaft temperature diagnosis through data acquisition, and is characterized by comprising the following steps:
s1: establishing a model combining edge computing and cloud computing;
s2: based on the model established in the step S1, respectively sending the axle temperature data to the edge node and the cloud computing center for computation, so as to monitor and diagnose the axle temperature data; the high-speed rail wheel shaft temperature monitoring system completes fault diagnosis on shaft temperature data by combining the edge nodes and the calculation result fed back by the cloud calculation center;
step S1 is specifically as follows:
s1.1: when the axle temperature of the high-speed rail wheel pair is monitored, the axle temperature data of the high-speed rail wheel pair is divided into two parts, one part is distributed to a plurality of edge nodes for edge calculation, and the other part is distributed to a cloud calculation center for cloud calculation;
s1.2: respectively calculating energy consumed by edge computing and energy consumed by data transmission to a cloud computing center;
s1.3: establishing a minimum total energy consumption problem and solving to obtain an optimal shaft temperature data distribution proportion, namely an optimal calculation task distribution proportion according to the energy consumed by edge calculation and the energy consumed by transmitting data to a cloud calculation center;
in the step S1.1, distributing a part of the axle temperature data, namely the computing tasks to n edge nodes which are close to the data source and have weak computing capability for computing, and distributing the other part of the axle temperature data, namely the computing tasks to a cloud computing center which has strong computing capability and is far from the data source and has the number of n +1 for computing;
in step S1.2, based on the computation delay of the edge node and the communication delay of the cloud computing center, the energy consumption of the edge node and the cloud computing center is further obtained, which is specifically as follows:
setting the calculation time delay t of the shaft temperature data on n edge nodesiComprises the following steps:
Figure FDA0003294395870000011
wherein, ViMiddle V1...VnCalculating the rate of n edge nodes, wherein the unit is bit/s; b is the total calculated amount of the wheel set for monitoring and diagnosing the axle temperature data, and the unit is bit;
the channel for transmitting the shaft temperature data to the cloud computing center is a Rayleigh channel, and the probability density function p (gamma) of Rayleigh distribution is expressed as follows:
Figure FDA0003294395870000012
wherein gamma is Rayleigh amplitude and gamma is more than 0, sigma2Is the noise power;
interruption probability of ProutThe expression is as follows:
Figure FDA0003294395870000021
where Pr () represents a probability function, gamma0Cutting off the channel amplitude corresponding to the channel gain;
the average propagation delay t of the successful transmission of the shaft temperature data to the cloud computing center with the number of n +1n+1Comprises the following steps:
Figure FDA0003294395870000022
wherein alpha isn+1The proportion of the computing tasks allocated to the cloud computing center is calculated; r isn+1The transmission rate of the shaft temperature data transmitted to the cloud computing center with the number of n +1 is bit/s; since the interruption probability is ProutAnd thus the probability of successful transmission is 1-ProutThe average speed of the shaft temperature data transmitted to the cloud computing center through the Rayleigh channel is
Figure FDA0003294395870000023
Unit is bit/s, using Vn+1To represent
Figure FDA0003294395870000024
Establishing a computational energy consumption expression L of n edge nodesiComprises the following steps:
Li＝Pi·αiB·ti，i＝1,2,…,n
wherein, PiCalculating the power to be consumed by unit bit for the ith edge node, wherein the unit is W/bit; alpha is alphaiCalculating a task proportion for the ith edge node; t is tiCalculating time delay of the shaft temperature data on the ith edge node, wherein the unit is s;
establishing an energy consumption expression L for transmitting data source to a cloud computing center with the number of n +1n+1Comprises the following steps:
Ln+1＝Pn+1·αn+1B·tn+1
wherein, Pn+1Transmitting the power consumed by unit bit to the cloud computing center with the number of n +1 for the data source, wherein the unit is W/bit; alpha is alphan+1The proportion of the computing tasks allocated to the cloud computing center is calculated; t is tn+1The unit is s for the average propagation delay of the successful transmission of the shaft temperature data to the cloud computing center with the number of n + 1.
2. The low-energy-consumption high-speed rail-wheel axle temperature monitoring and diagnosing method based on the edge calculation as claimed in claim 1, wherein: in step S1.3, the objective function L that minimizes the total energy consumption is finally established as:
Figure FDA0003294395870000031
wherein j is 1, 2., n, n +1, the first n represents an edge node, and the n +1 represents a cloud computing center; pi·αiB·tiRepresents the energy consumption of the ith edge node, i ═ 1, 2. Pn+1·αn+1B·tn+1Energy consumption for data source transmission to the cloud computing center, denoted by the number n +1, sheetThe bit is W;
the time delay t will be calculatediAnd propagation delay tn+1The expression of (1) is substituted, i.e. the problem of minimizing the total energy consumption is established as:
Figure FDA0003294395870000032
Figure FDA0003294395870000033
the optimization problem meets the KKT condition, a Lagrange equation is established, the Lagrange multiplier method is used for solving, and the optimal task proportion alpha for minimizing the total energy consumption is obtained through solvingj *Comprises the following steps:
Figure FDA0003294395870000034
optimal task proportion alpha allocated to edge nodesi *Expressed as:
Figure FDA0003294395870000035
wherein alpha isi *Calculating the task proportion of the optimal shaft temperature data distributed to the edge node i;
optimal task proportion alpha distributed to cloud computing centern+1 *Expressed as:
Figure FDA0003294395870000036
wherein alpha isn+1 *The optimal shaft temperature data is distributed to the cloud computing center to calculate the task proportion.
3. A process as claimed in claim 2The low-energy-consumption high-speed rail wheel shaft temperature monitoring and diagnosing method based on edge calculation is characterized by comprising the following steps of: in step S2, the optimal task ratio α calculated in step S1.3 is obtainedj *And respectively sending the shaft temperature data of the corresponding part to n edge nodes and the cloud computing center with the number of n +1 for diagnosis and computation.
CN202110605767.7A 2021-05-31 2021-05-31 Low-energy-consumption high-speed rail wheel shaft temperature monitoring and diagnosing method based on edge calculation Active CN113361088B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110605767.7A CN113361088B (en) 2021-05-31 2021-05-31 Low-energy-consumption high-speed rail wheel shaft temperature monitoring and diagnosing method based on edge calculation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110605767.7A CN113361088B (en) 2021-05-31 2021-05-31 Low-energy-consumption high-speed rail wheel shaft temperature monitoring and diagnosing method based on edge calculation

Publications (2)

Publication Number Publication Date
CN113361088A CN113361088A (en) 2021-09-07
CN113361088B true CN113361088B (en) 2021-12-17

Family

ID=77530746

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110605767.7A Active CN113361088B (en) 2021-05-31 2021-05-31 Low-energy-consumption high-speed rail wheel shaft temperature monitoring and diagnosing method based on edge calculation

Country Status (1)

Country Link
CN (1) CN113361088B (en)

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100908783B1 (en) * 2007-07-25 2009-07-22 한국전력공사 Switching device for transformer having an uninterruptible power function and method of voltage control using the same
CN110554657B (en) * 2019-10-16 2023-03-03 河北工业大学 Health diagnosis system and diagnosis method for operation state of numerical control machine tool
CN111444169A (en) * 2020-03-30 2020-07-24 国网江苏省电力有限公司连云港市赣榆区供电分公司 Transformer substation electrical equipment state monitoring and diagnosis system and method

Also Published As

Publication number Publication date
CN113361088A (en) 2021-09-07

Similar Documents

Publication Publication Date Title
CN113904947B (en) Vehicle-road cooperative distributed edge computing task unloading and resource allocation method and system
CN102612080B (en) Minimum-energy-consumption self-adaption convergent routing method on basis of second-generation wavelet zero tree encoding
CN111114596B (en) Multi-train speed curve collaborative optimization method considering network loss
CN113361113B (en) Energy-consumption-adjustable twin data distribution method for high-speed rail bogie
CN113361088B (en) Low-energy-consumption high-speed rail wheel shaft temperature monitoring and diagnosing method based on edge calculation
US11155419B1 (en) Dynamic reliability evaluation method for coupling faults of middle trough of scraper conveyor
CN107069782B (en) Capacity configuration method applied to rail transit vehicle-mounted hybrid energy storage system
CN105682238A (en) Optimal allocation method of data transmission power in relay system
CN111669768A (en) Wireless energy supply cellular Internet of things resource optimization method based on full-duplex self-return
He et al. Research on multi-objective real-time optimization of automatic train operation (ATO) in urban rail transit
CN107086569B (en) A kind of distribution network electric energy loss computing method based on impedance distribution
CN106780041A (en) Power network project of transmitting and converting electricity investment forecasting method based on particle cluster algorithm
CN112508365B (en) Online coal consumption curve rolling correction method and system
CN103426294B (en) A kind of division methods based on the other multi-layer Fuzzy of emergency traffic flow priority
CN112510704B (en) Online coal consumption curve real-time generation method and system
CN115195697A (en) Power control method, device, equipment and medium applied to hybrid vehicle
CN111756443B (en) Method for pre-estimating key parameters in space chaotic laser communication system
CN112416577A (en) Cooperative intelligent calculation and distribution method suitable for block chain workload certification
CN105828426A (en) Management method and system for managing node energy of high-speed railway infrastructure monitoring sensor network
CN113793022B (en) Power grid investment planning scheme generation method and system based on distributed cloud energy storage
CN111160653A (en) Distributed energy storage system wind power consumption capacity monitoring method based on cloud computing
CN211075912U (en) Safety monitoring system for train running gear
CN116593191A (en) Rail locomotive vehicle performance detection system, detection method, detection equipment and medium
CN107765674A (en) The long-distance monitorng device and method of new-energy automobile
Pan Mechanism of rail potential imbalance in long tunnel based on GA and analysis of improvement and optimization measures

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