CN107220712B - Converter valve state monitoring and maintenance decision method and system based on Markov process - Google Patents
Converter valve state monitoring and maintenance decision method and system based on Markov process Download PDFInfo
- Publication number
- CN107220712B CN107220712B CN201710338753.7A CN201710338753A CN107220712B CN 107220712 B CN107220712 B CN 107220712B CN 201710338753 A CN201710338753 A CN 201710338753A CN 107220712 B CN107220712 B CN 107220712B
- Authority
- CN
- China
- Prior art keywords
- state
- converter valve
- states
- degradation
- determining
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 132
- 238000000034 method Methods 0.000 title claims abstract description 106
- 238000012423 maintenance Methods 0.000 title claims abstract description 92
- 238000006731 degradation reaction Methods 0.000 claims abstract description 187
- 230000015556 catabolic process Effects 0.000 claims abstract description 173
- 238000012546 transfer Methods 0.000 claims abstract description 13
- 230000007704 transition Effects 0.000 claims description 37
- 230000006866 deterioration Effects 0.000 claims description 18
- 238000004364 calculation method Methods 0.000 claims description 9
- 230000001419 dependent effect Effects 0.000 claims description 6
- 238000001514 detection method Methods 0.000 claims description 6
- 238000005457 optimization Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000002250 progressing effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000007562 laser obscuration time method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000013024 troubleshooting Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
Abstract
The invention discloses a converter valve state monitoring and state maintenance decision method based on a Markov process, which comprises the following steps: determining different degradation states describing the converter valve, and determining state monitoring corresponding to the degradation states and state maintenance decisions corresponding to the degradation states according to the different degradation states; according to the historical data, obtaining the transfer rate of different degradation states of the converter valve, and determining the different degradation states, state monitoring corresponding to the degradation states and expected benefits of state maintenance decisions corresponding to the degradation states; determining a state frequency and an average duration of time that the converter valve based on the Markov process stays in different degradation states; determining the available time of the converter valve according to the average duration, and determining the optimal state monitoring frequency; and determining different degradation states of the converter valve according to the average duration, state monitoring and the expected profit sum of the state maintenance decision, and determining the optimal state maintenance decision.
Description
Technical Field
The invention relates to the field of converter valve maintenance, in particular to a converter valve state monitoring and maintenance decision method and system based on a Markov process.
Background
The converter valve often faces the problems of state monitoring, troubleshooting and the like in the using process. Compared with the overhauling modes such as overhauling after an accident, regular overhauling and the like, the overhauling efficiency is low, and the labor cost input is high. Therefore, the condition maintenance of the equipment can be implemented to improve the maintenance efficiency. At present, under the premise of an intelligent power grid concept, an intelligent substation technology is implemented, the implementation conditions of state maintenance of substation equipment are increasingly mature, and the process of comprehensively implementing the state maintenance of the equipment by a power grid is also promoted. For a converter valve, on the basis of state maintenance, on the basis of a monitoring technology, no technology for developing a state maintenance strategy in a research direction from a qualitative state to a quantitative state exists.
Therefore, a technique is needed to solve the problem of optimization of converter valve condition monitoring and maintenance decisions.
Disclosure of Invention
The invention provides a converter valve state monitoring and maintenance decision method and system based on a Markov process, and aims to solve the problem of how to optimize the converter valve state monitoring and maintenance decision.
In order to solve the above problems, the present invention provides a method for converter valve status monitoring and status maintenance decision based on a markov process, the method comprising:
determining different degradation states describing the converter valve, and determining state monitoring corresponding to the degradation states and state maintenance decisions corresponding to the degradation states according to the different degradation states;
according to historical data, obtaining the transfer rate of different degradation states of the converter valve, and determining the expected benefits of the different degradation states, the state monitoring corresponding to the degradation states and the state maintenance decision corresponding to the degradation states;
determining a state frequency at which the converter valve based on the Markov process stays in the different degradation state and an average duration in the different degradation state;
determining the available time of the converter valve according to the average duration time of the different degradation states, establishing a state monitoring frequency model by taking the longest available time of the converter valve as a target, and determining the optimal state monitoring frequency;
and determining the different degradation states of the converter valve, the state monitoring corresponding to the degradation states and the expected income sum of the state maintenance decision corresponding to the degradation states according to the average duration of the different degradation states, establishing a state maintenance decision model by taking the expected income sum of the converter valve as the highest target, and determining the optimal state maintenance decision.
Preferably, the expected revenue of the different degradation states of the converter valve is an operational revenue gained by maintaining the degradation states;
the expected benefit of the condition monitoring corresponding to the degraded condition is the fund for implementing the condition monitoring;
the expected benefit of the condition overhaul decision corresponding to the degraded condition is the capital to implement the overhaul decision.
Preferably, determining a state frequency at which the converter valve based on the markov process stays in the different degradation state and an average duration in the different degradation state comprises:
describing the degradation process of the converter valve according to a Markov random process model, wherein the extreme state of the degradation state of the converter valve after n steps of rotation is the steady state of the degradation state of the converter valve, the probability of the steady state after the degradation state is a constant, and the probability of the steady state is irrelevant to the initial state of the converter valve;
the linear differential equation set of the Markov random process is as follows:
wherein λ isijFor the transfer rate, P (t) is a Markov universal equation, and t is time;
obtaining the steady state probability by solving the following system of linear equations:
when the deterioration process of the converter valve reaches the steady state, the average number of times of staying in the deteriorated state i in each unit time is the state frequency f of the converter valve in the state iiDuration T of state iiMeans the average duration of stay in state i, f, when the deterioration process of the converter valve reaches a steady stateijFor j frequencies of transition to state i, wherein
Preferably, the determining the available time of the converter valve according to the average duration of the different degradation states, establishing a state monitoring frequency model with the longest available time of the converter valve as a target, and determining an optimal state monitoring frequency includes:
the method comprises the following steps of establishing a state monitoring frequency model by taking the longest available time of the converter valve as a target, wherein an objective function is max { B }, wherein B is a function of monitoring frequency gamma, B is f (gamma), f (gamma) is a functional relation representing the detection frequency of the converter valve and the available time of the converter valve, the monitoring frequency is a variable, the available time is a dependent variable, and the optimal monitoring frequency gamma meets the following requirements:
γmin≤γ≤γmaxwherein γ ismax、γminRespectively, a maximum value and a minimum value of the monitoring frequency.
Preferably, the determining the expected profit sum of the different degradation states of the converter valves, the state monitoring corresponding to the degradation states and the state overhaul decision corresponding to the degradation states according to the average duration of the different degradation states, and the determining the optimal state overhaul decision with the target that the expected profit sum of the converter valves is the highest comprises:
and determining an objective function of an optimal condition maintenance decision model by taking the highest expected income sum of the converter valve as a target: max { G }, wherein G is a sum of expected benefits of the different degraded states, the state monitoring corresponding to the degraded states, and the state overhaul decisions corresponding to the degraded states for the converter valve over a life cycle.
Preferably, the time of the transition between said different degraded states of said converter valve follows an exponential distribution, the probability of the transition between said different degraded states being constant.
Based on another aspect of the invention, the invention provides a system for converter valve state monitoring and state maintenance decision based on Markov process, which comprises:
the device comprises an initialization unit, a state monitoring unit and a state maintenance decision unit, wherein the initialization unit is used for determining different degradation states describing the converter valve, and determining state monitoring corresponding to the degradation states and state maintenance decisions corresponding to the degradation states according to the different degradation states;
the acquisition unit acquires the transfer rate of different degradation states of the converter valve according to historical data, and determines the different degradation states, the state monitoring corresponding to the degradation states and the expected benefits of the state maintenance decision corresponding to the degradation states;
a first calculation unit for determining a state frequency at which the converter valve based on a Markov process stays in the different degradation state and an average duration in the different degradation state;
the second calculation unit is used for determining the available time of the converter valve according to the average duration time of the different degradation states, establishing a state monitoring frequency model by taking the longest available time of the converter valve as a target, and determining the optimal state monitoring frequency;
a third calculating unit, configured to determine, according to the average duration of the different degradation states, an expected revenue sum of the different degradation states of the converter valve, the state monitoring corresponding to the degradation states, and the state overhaul decision corresponding to the degradation states, and determine an optimal state overhaul decision with a target that the expected revenue sum of the converter valve is the highest.
Preferably, the expected revenue of the different degradation states of the converter valve is an operational revenue gained by maintaining the degradation states;
the expected benefit of the condition monitoring corresponding to the degraded condition is the fund for implementing the condition monitoring;
the expected benefit of the condition overhaul decision corresponding to the degraded condition is the capital to implement the overhaul decision.
Preferably, the first computing unit is further configured to:
describing the degradation process of the converter valve according to a Markov random process model, wherein the extreme state of the degradation state of the converter valve after n steps of rotation is the steady state of the degradation state of the converter valve, the probability of the steady state after the degradation state is a constant, and the probability of the steady state is irrelevant to the initial state of the converter valve;
the linear differential equation set of the Markov random process is as follows:
wherein λ isijFor the transfer rate, P (t) is a Markov universal equation, and t is time;
obtaining the steady state probability by solving the following system of linear equations:
when the deterioration process of the converter valve reaches the steady state, the average number of times of staying in the deteriorated state i in each unit time is that of the converter valve in the state iFrequency of state fiDuration T of state iiMeans the average duration of stay in state i, f, when the deterioration process of the converter valve reaches a steady stateijFor j frequencies of transition to state i, wherein
Preferably, the second computing unit is further configured to:
the method comprises the following steps of establishing a state monitoring frequency model by taking the longest available time of the converter valve as a target, wherein an objective function is max { B }, wherein B is a function of monitoring frequency gamma, B is f (gamma), f (gamma) is a functional relation representing the detection frequency of the converter valve and the available time of the converter valve, the monitoring frequency is a variable, the available time is a dependent variable, and the optimal monitoring frequency gamma meets the following requirements:
γmin≤γ≤γmaxwherein γ ismax、γminRespectively, a maximum value and a minimum value of the monitoring frequency.
Preferably, the third computing unit is further configured to:
and determining an objective function of an optimal condition maintenance decision model by taking the highest expected income sum of the converter valve as a target: max { G }, wherein G is a sum of expected benefits of the different degraded states, the state monitoring corresponding to the degraded states, and the state overhaul decisions corresponding to the degraded states for the converter valve over a life cycle.
Preferably, the time of the transition between said different degraded states of said converter valve follows an exponential distribution, the probability of the transition between said different degraded states being constant.
The technical scheme of the invention takes the converter valve as an object, and provides a model and a solving method for state monitoring and state maintenance of the converter valve. The technical scheme of the invention provides a converter valve state transfer model considering the maintenance strategy and based on the Markov random process under the guidance of the whole life cycle management of the converter valve, the equipment state maintenance technology and related concepts, and calculates related index parameters based on the Markov process equation, so that the reliability of the converter valve can be conveniently and quantitatively analyzed, and the converter valve has strong operability.
Drawings
A more complete understanding of exemplary embodiments of the present invention may be had by reference to the following drawings in which:
FIG. 1 is a flow chart of a method for Markov process based converter valve condition monitoring and maintenance decision-making according to one embodiment of the present invention;
FIG. 2 is a diagram of a model of a state transition of a converter valve based on a Markov process in accordance with one embodiment of the present disclosure; and
fig. 3 is a system configuration diagram of a state monitoring and maintenance decision of a converter valve based on a markov process according to an embodiment of the present invention.
Detailed Description
The exemplary embodiments of the present invention will now be described with reference to the accompanying drawings, however, the present invention may be embodied in many different forms and is not limited to the embodiments described herein, which are provided for complete and complete disclosure of the present invention and to fully convey the scope of the present invention to those skilled in the art. The terminology used in the exemplary embodiments illustrated in the accompanying drawings is not intended to be limiting of the invention. In the drawings, the same units/elements are denoted by the same reference numerals.
Unless otherwise defined, terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Further, it will be understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense.
Fig. 1 is a flow chart of a method for monitoring the state of a converter valve and making a maintenance decision based on a markov process according to an embodiment of the present invention. The embodiment of the invention takes the converter valve as an object, and provides a model and a solving method for state monitoring and state maintenance decision of the converter valve. The method and the device provided by the embodiment of the invention aim at the object of the converter valve, provide a converter valve state transition model based on the Markov random process under the guidance of the whole life cycle management, the state monitoring, the state maintenance decision and related concepts of the converter valve, and calculate related index parameters based on the Markov process equation, so that the reliability of the converter valve can be quantitatively analyzed, and the method and the device have strong operability. The embodiment of the invention provides a converter valve state monitoring and state maintenance decision method based on a Markov process, which comprises the following steps:
determining different degradation states describing the converter valve, and determining state monitoring corresponding to the degradation states and state maintenance decisions corresponding to the degradation states according to the different degradation states.
In the reliability analysis, the degradation process of the converter valve can be expressed as continuous degradation loss of the converter valve until finally a fault occurs. The embodiments of the present invention describe different stages in the converter valve deterioration process in a deteriorated state 1, a deteriorated state 2 and a deteriorated state 3, and if no maintenance is performed, the converter valve will progress from the deteriorated state 1 to the deteriorated state 3 and finally to the faulty state. After the converter valve is in failure, the converter valve can be restored to the degradation state 1 after replacement or maintenance. In each degradation state, corresponding state monitoring is carried out, and the maintainers can determine different maintenance decisions according to the state monitoring condition, namely adopt different maintenance means. After servicing, the equipment will also revert to a different state.
And acquiring the transition rate of different degradation states of the converter valve according to the historical data, and determining the different degradation states, state monitoring corresponding to the degradation states and expected benefits of state maintenance decisions corresponding to the degradation states. The expected revenue for different degradation states of the converter valve is the operational revenue gained by maintaining the degradation state; the expected benefit of condition monitoring corresponding to the degraded condition is the capital to implement the condition monitoring; the expected revenue for a condition overhaul decision corresponding to the degraded condition is the capital to implement the overhaul decision.
The transition rate between different degradation states of the converter valve is derived from historical statistical data in units of/year. The expected benefit of different degradation states refers to the operation benefit that the converter valve can obtain for maintaining the degradation state, the expected benefit of degradation state monitoring refers to the capital consumed for implementing degradation state monitoring, and the expected benefit of overhaul refers to the capital consumed for implementing different overhaul means.
The state frequency and the average duration of the state in which the markov process based converter valve stays in different degradation states are determined. Preferably, determining the state frequency and the average duration of the state of the markov process based converter valve staying in the different degradation states comprises:
describing a degradation process of the converter valve according to a Markov random process model, wherein the degradation state of the converter valve enters a limit state after n steps of conversion, the limit state is a steady state of the degradation state of the converter valve, the probability of the steady state after the degradation state is a constant, and the probability of the steady state is irrelevant to the initial state of the converter valve;
the linear differential equation set of the Markov random process is as follows:
wherein λ isijFor the transfer rate, P (t) is a Markov universal equation, and t is time;
the steady state probability is obtained by solving the following system of linear equations:
when the deterioration process of the converter valve reaches a steady state, the average number of times of staying in the deteriorated state i per unit time is the state frequency f of the converter valve in the state iiDuration T of state iiMeans the average duration of stay in state i, f, when the deterioration process of the converter valve reaches a steady stateijFor j frequencies of transition to state i, wherein
And determining the available time of the converter valve according to the average duration of different degradation states, establishing a state monitoring frequency model by taking the longest available time of the converter valve as a target, and determining the optimal state monitoring frequency. Preferably, determining the available time of the converter valve according to the average duration of different degradation states, establishing a state monitoring frequency model with the longest available time of the converter valve as a target, and determining the optimal state monitoring frequency comprises:
the method comprises the following steps of establishing a state monitoring frequency model by taking the longest available time of the converter valve as a target, wherein an objective function is max { B }, wherein B is a function of monitoring frequency gamma, B is f (gamma), f (gamma) is a functional relation representing the detection frequency of the converter valve and the available time of the converter valve, the monitoring frequency is a variable, the available time is a dependent variable, and the optimal monitoring frequency gamma meets the following requirements:
γmin≤γ≤γmaxwherein γ ismax、γminRespectively, a maximum value and a minimum value of the monitoring frequency.
And determining different degradation states of the converter valve, state monitoring corresponding to the degradation states and expected income sum of state maintenance decisions corresponding to the degradation states according to the average duration time of the different degradation states, establishing a state maintenance decision model by taking the highest expected income sum of the converter valve as a target, and determining an optimal state maintenance decision. Preferably, determining the different degradation states of the converter valves, the sum of expected gains of the state monitoring corresponding to the degradation states and the state overhaul decision corresponding to the degradation states according to the average duration of the different degradation states, and determining the optimal state overhaul decision by taking the sum of expected gains of the converter valves as the highest target, comprises:
and determining an objective function of the optimal state maintenance decision model by taking the highest expected income sum of the converter valve as a target: max { G }, where G is a sum of expected benefits of different degraded states, state monitoring corresponding to the degraded states, and state overhaul decisions corresponding to the degraded states of the converter valve over the life cycle.
The function solving steps are as follows: (1) firstly, giving an initial decision of iteration, and giving an initial maintenance strategy according to the principle that the expected maximum benefit of each degradation state is maximum in order to reduce the iteration times and simplify the iteration process; (2) changing a maintenance strategy and calculating expected benefits of the whole life cycle; (3) and (3) judging whether the expected yield of the whole life cycle of the converter valve in the degradation state process is the maximum after the maintenance strategy is changed, repeating the step (2) if the expected yield is not the maximum, and if the strategies before and after two times in the iteration process are the same, indicating that the current strategy is the optimal maintenance strategy and stopping the iteration. And the maintenance strategy obtained at the moment is the optimal strategy for maintenance of the deteriorated state of the converter valve.
Preferably, the time of the transition between different degraded states of the converter valve follows an exponential distribution, the probability of the transition between different degraded states being constant.
The following further illustrates embodiments of the invention:
fig. 1 is a flowchart of a method for monitoring a converter valve state and making a maintenance decision based on a markov process according to an embodiment of the present invention. The method and the device for simulating the state transition of the converter valve in the embodiment of the invention equate the deterioration state process of the converter valve to a random process, and establish a state transition model of the deterioration process of the converter valve as shown in figure 2. The model comprises different degradation stages and fault states of the degradation process of the converter valve, degradation state monitoring and maintenance decision in the life cycle of the converter valve. Assuming that the converter valve may transition from the current degraded state to another state and that the time for the transition between the respective degraded states satisfies an exponential distribution, i.e. the probability of a degraded state transition is constant, the next degraded state of the converter valve is only relevant for the current state. An exponential distribution is a continuous probability distribution in probability theory and statistics, which can be used to represent the time interval of occurrence of independent random events, and an important characteristic of which is memoryless.
For example, the state transition probability of the degraded state 2 progressing to the degraded state 3 is constant, and the degraded state 3 is related to the degraded state 2 only and is not related to the degraded state 1.
The deterioration process of the converter valves is illustrated in fig. 2 in three separate deterioration states, the converter valves eventually entering a faulty state, after which a new converter valve can be replaced or repaired back to the oldest deterioration state. Each degradation state corresponds to state monitoring, a maintenance strategy can be determined through monitoring, and the converter valve can be restored to the degradation state of the previous stage after maintenance.
And the state of the converter valve 2 is monitored according to the state monitoring 2, after monitoring, a maintainer determines to adopt an inspection means 2, and the state of the converter valve is restored from the state of the converter valve 2 to the state of the converter valve 1 through the implementation of inspection.
λn、δn、γnAnd munRespectively representing transition probabilities between different states, whereinnRefers to the transition probability of the converter valve progressing from the degraded state n to the degraded state n + 1.
δnThe method refers to the transition probability of implementing the maintenance means n after the converter valve is subjected to state monitoring n.
γnRefers to a transition probability of performing state monitoring n on the degraded state n of the converter valve.
μnThe transition probability of the converter valve recovering the degradation state n-1 after the maintenance means n is implemented on the converter valve is shown.
After the converter valves in different degradation states are monitored, different maintenance decisions are made, and different maintenance means are adopted to have different influences on the running economy of the converter valves, so expected benefits are given to different states to describe the maintenance economy of the converter valves.
Taking the degradation state 2 as an example, if the converter valve does not adopt the state monitoring 2 in the degradation state 2, the maintainer cannot find that the converter valve is in the degradation state 2, and the costs required for monitoring in different states are different. Similarly, if the state monitoring 2 is adopted, the converter valve is found to be in the degradation state 2, but the maintenance personnel do not decide to implement the maintenance means 2, the converter valve cannot be restored to the degradation state 1, and the cost required by different maintenance means is different. Ultimately resulting in different desired benefits.
And establishing a state transition model of the converter valve in the degradation process, and determining expected benefits of each state. And the transition rate between the degradation states, the steady state probability of the converter valve degradation process and the average duration of the degradation states can be obtained. After the steady state probability representation converter valve degradation process enters the steady state, even if state transition occurs again, the transition probability of each state in the converter valve degradation process does not change any more.
After the parameters are obtained, namely the expected income of each degradation state, the average duration of each degradation state and the degradation state transition probability, an optimization model of routine state monitoring frequency of the converter valve can be established, the longest available time of the converter valve is taken as an objective equation, and an optimization method is applied to obtain the optimal routine state monitoring frequency. After the routine inspection frequency is determined, a converter valve state maintenance optimization decision model can be established, and an optimal maintenance strategy is obtained by applying an iteration method, wherein the method comprises the following steps: (1) firstly, giving an initial decision of iteration, and giving an initial maintenance strategy according to the principle that the expected maximum benefit of each degradation state is maximum in order to reduce the iteration times and simplify the iteration process; (2) changing a maintenance strategy and calculating expected benefits of the whole life cycle; (3) and (3) judging whether the expected yield of the whole life cycle of the converter valve in the degradation process is maximum after the maintenance strategy is changed, repeating the step (2) if the expected yield is not maximum, and if the strategies before and after two times in the iteration process are the same, indicating that the current strategy is the optimal maintenance strategy and stopping the iteration. The obtained maintenance strategy is the optimal strategy for the state maintenance of the converter valve.
According to the converter valve state maintenance optimization decision method, under the background of power equipment life cycle management, the degradation process of the converter valve is described by the Markov process, so that the reliability of the converter valve can be accurately and quantitatively evaluated, and the operation reliability and economy of the converter valve are comprehensively considered when maintenance strategy optimization is carried out, so that the maintenance effect is better.
Fig. 3 is a system configuration diagram of a state monitoring and maintenance decision of a converter valve based on a markov process according to an embodiment of the present invention. As shown in fig. 3, the system 300 includes:
the initialization unit 301 is used for determining different degradation states of the converter valve, and determining state monitoring corresponding to the degradation states and state maintenance decisions corresponding to the degradation states according to the different degradation states;
the obtaining unit 302 is used for obtaining the transfer rates of different degradation states of the converter valve according to the historical data, and determining the expected benefits of the different degradation states, state monitoring corresponding to the degradation states and state maintenance decisions corresponding to the degradation states;
a first calculation unit 303, the first calculation unit 303 being configured to determine a state frequency at which the markov process-based converter valve stays in the different degradation states and an average duration in the different degradation states;
the second calculating unit 304, the second calculating unit 304 is configured to determine the available time of the converter valve according to the average duration of different degradation states, establish a state monitoring frequency model with the longest available time of the converter valve as a target, and determine an optimal state monitoring frequency;
a third calculating unit 305, wherein the third calculating unit 305 is configured to determine different degradation states of the converter valves according to the average duration of the different degradation states, the expected revenue sum of the state monitoring corresponding to the degradation states and the state overhaul decision corresponding to the degradation states, and determine an optimal state overhaul decision by taking the expected revenue sum of the converter valves as a target with the highest expected revenue sum.
Preferably, the expected revenue for different degradation states of the converter valves of the system 300 is the operational revenue gained for maintaining the degradation state;
the expected benefit of condition monitoring corresponding to the degraded condition is the capital to implement the condition monitoring;
the expected revenue for a condition overhaul decision corresponding to the degraded condition is the capital to implement the overhaul decision.
Preferably, the first calculation unit 303 is further configured to:
describing a degradation process of the converter valve according to a Markov random process model, wherein the degradation state of the converter valve enters a limit state after n steps of conversion, the limit state is a steady state of the degradation state of the converter valve, the probability of the steady state after the degradation state is a constant, and the probability of the steady state is irrelevant to the initial state of the converter valve;
the linear differential equation set of the Markov random process is as follows:
wherein λ isijFor the transfer rate, P (t) is a Markov universal equation, and t is time;
the steady state probability is obtained by solving the following system of linear equations:
when the deterioration process of the converter valve reaches a steady state, the average number of times of staying in the deteriorated state i per unit time is the state frequency f of the converter valve in the state iiDuration T of state iiMeans the average duration of stay in state i, f, when the deterioration process of the converter valve reaches a steady stateijFor j frequencies of transition to state i, wherein
Preferably, the second calculation unit 304 is further configured to:
the method comprises the following steps of establishing a state monitoring frequency model by taking the longest available time of the converter valve as a target, wherein an objective function is max { B }, wherein B is a function of monitoring frequency gamma, B is f (gamma), f (gamma) is a functional relation representing the detection frequency of the converter valve and the available time of the converter valve, the monitoring frequency is a variable, the available time is a dependent variable, and the optimal monitoring frequency gamma meets the following requirements:
γmin≤γ≤γmaxwherein γ ismax、γminRespectively, a maximum value and a minimum value of the monitoring frequency.
Preferably, the third calculation unit 305 is further configured to:
and determining an objective function of the optimal state maintenance decision model by taking the highest expected income sum of the converter valve as a target: max { G }, where G is a sum of expected benefits of different degraded states, state monitoring corresponding to the degraded states, and state overhaul decisions corresponding to the degraded states of the converter valve over the life cycle.
Preferably, the time of transition between different degraded states of the converter valve of the system 300 follows an exponential distribution, with a constant probability of transition between different degraded states.
The system 300 for monitoring and maintaining the state of the converter valve based on the markov process in one embodiment of the present invention corresponds to the method 100 for monitoring and maintaining the state of the converter valve based on the markov process in another embodiment of the present invention, and will not be described herein again.
The invention has been described with reference to a few embodiments. However, other embodiments of the invention than the one disclosed above are equally possible within the scope of the invention, as would be apparent to a person skilled in the art from the appended patent claims.
Generally, all terms used in the claims are to be interpreted according to their ordinary meaning in the technical field, unless explicitly defined otherwise herein. All references to "a/an/the [ device, component, etc ]" are to be interpreted openly as referring to at least one instance of said device, component, etc., unless explicitly stated otherwise. The steps of any method disclosed herein do not have to be performed in the exact order disclosed, unless explicitly stated.
Claims (10)
1. A method of converter valve condition monitoring and condition overhaul decision based on a markov process, the method comprising:
determining different degradation states describing the converter valve, and determining state monitoring corresponding to the degradation states and state maintenance decisions corresponding to the degradation states according to the different degradation states;
according to historical data, obtaining the transfer rate of different degradation states of the converter valve, and determining the expected benefits of the different degradation states, the state monitoring corresponding to the degradation states and the state maintenance decision corresponding to the degradation states;
determining a state frequency at which the converter valve based on the Markov process stays in the different degradation state and an average duration in the different degradation state;
determining the available time of the converter valve according to the average duration time of the different degradation states, establishing a state monitoring frequency model by taking the longest available time of the converter valve as a target, and determining the optimal state monitoring frequency;
determining the different degradation states of the converter valve, the state monitoring corresponding to the degradation states and the expected income sum of the state maintenance decision corresponding to the degradation states according to the average duration time of the different degradation states, establishing a state maintenance decision model by taking the expected income sum of the converter valve as a target to determine an optimal state maintenance decision; the method comprises the following steps: and determining an objective function of an optimal condition maintenance decision model by taking the highest expected income sum of the converter valve as a target: max { G }, wherein G is a sum of expected benefits of the different degraded states, the state monitoring corresponding to the degraded states, and the state overhaul decisions corresponding to the degraded states for the converter valve over a life cycle.
2. The method of claim 1, the expected revenue for the different degradation states of the converter valve being operational revenue gained for maintaining the degradation states;
the expected benefit of the condition monitoring corresponding to the degraded condition is the fund for implementing the condition monitoring;
the expected benefit of the condition overhaul decision corresponding to the degraded condition is the capital to implement the overhaul decision.
3. The method of claim 1, determining a state frequency at which the Markov process based converter valve stays in the different degradation state and an average duration in the different degradation state, comprising:
describing the degradation process of the converter valve according to a Markov random process model, wherein the extreme state of the degradation state of the converter valve after n steps of rotation is the steady state of the degradation state of the converter valve, the probability of the steady state after the degradation state is a constant, and the probability of the steady state is irrelevant to the initial state of the converter valve;
the linear differential equation set of the Markov random process is as follows:
wherein λ isijFor the transfer rate, P (t) is a Markov universal equation, and t is time;
obtaining the steady state probability by solving the following system of linear equations:
when the deterioration process of the converter valve reaches the steady state, the converter valve stays in the steady state in unit timeThe average number of times in the degradation state i is the state frequency f of the converter valve in the state iiDuration T of state iiMeans the average duration of stay in state i, f, when the deterioration process of the converter valve reaches a steady stateijFor j frequencies of transition to state i, wherein
4. The method of claim 1, said determining an availability time for said converter valves based on an average duration of said different degradation states, establishing a state monitoring frequency model with a goal of said converter valve availability time being longest, determining an optimal state monitoring frequency, comprising:
the method comprises the following steps of establishing a state monitoring frequency model by taking the longest available time of the converter valve as a target, wherein an objective function is max { B }, wherein B is a function of monitoring frequency gamma, B is f (gamma), f (gamma) is a functional relation representing the detection frequency of the converter valve and the available time of the converter valve, the monitoring frequency is a variable, the available time is a dependent variable, and the optimal monitoring frequency gamma meets the following requirements:
γmin≤γ≤γmaxwherein γ ismax、γminRespectively, a maximum value and a minimum value of the monitoring frequency.
5. The method of claim 1, wherein a time for a transition between the different degraded states of the converter valve follows an exponential distribution, and a probability of a transition between the different degraded states is constant.
6. A system for markov process based converter valve condition monitoring and condition overhaul decisions, the system comprising:
the device comprises an initialization unit, a state monitoring unit and a state maintenance decision unit, wherein the initialization unit is used for determining different degradation states describing the converter valve, and determining state monitoring corresponding to the degradation states and state maintenance decisions corresponding to the degradation states according to the different degradation states;
the acquisition unit acquires the transfer rate of different degradation states of the converter valve according to historical data, and determines the different degradation states, the state monitoring corresponding to the degradation states and the expected benefits of the state maintenance decision corresponding to the degradation states;
a first calculation unit for determining a state frequency at which the converter valve based on a Markov process stays in the different degradation state and an average duration in the different degradation state;
the second calculation unit is used for determining the available time of the converter valve according to the average duration time of the different degradation states, establishing a state monitoring frequency model by taking the longest available time of the converter valve as a target, and determining the optimal state monitoring frequency;
a third calculating unit, configured to determine, according to an average duration of the different degradation states, an expected revenue sum of the different degradation states of the converter valve, the state monitoring corresponding to the degradation states, and the state overhaul decision corresponding to the degradation states, and determine an optimal state overhaul decision with a target that the expected revenue sum of the converter valve is the highest; the third computing unit is further configured to:
and determining an objective function of an optimal condition maintenance decision model by taking the highest expected income sum of the converter valve as a target: max { G }, wherein G is a sum of expected benefits of the different degraded states, the state monitoring corresponding to the degraded states, and the state overhaul decisions corresponding to the degraded states for the converter valve over a life cycle.
7. The system of claim 6, the expected revenue for the different degradation states of the converter valve being operational revenue gained to maintain the degradation states;
the expected benefit of the condition monitoring corresponding to the degraded condition is the fund for implementing the condition monitoring;
the expected benefit of the condition overhaul decision corresponding to the degraded condition is the capital to implement the overhaul decision.
8. The system of claim 6, the first computing unit to further:
describing the degradation process of the converter valve according to a Markov random process model, wherein the extreme state of the degradation state of the converter valve after n steps of rotation is the steady state of the degradation state of the converter valve, the probability of the steady state after the degradation state is a constant, and the probability of the steady state is irrelevant to the initial state of the converter valve;
the linear differential equation set of the Markov random process is as follows:
wherein λ isijFor the transfer rate, P (t) is a Markov universal equation, and t is time;
obtaining the steady state probability by solving the following system of linear equations:
when the deterioration process of the converter valve reaches the steady state, the average number of times of staying in the deteriorated state i in each unit time is the state frequency f of the converter valve in the state iiDuration T of state iiMeans that said converter valve deterioration process is reachedAverage duration of stay in state i, f, at steady stateijFor j frequencies of transition to state i, wherein
9. The system of claim 6, the second computing unit to further:
the method comprises the following steps of establishing a state monitoring frequency model by taking the longest available time of the converter valve as a target, wherein an objective function is max { B }, wherein B is a function of monitoring frequency gamma, B is f (gamma), f (gamma) is a functional relation representing the detection frequency of the converter valve and the available time of the converter valve, the monitoring frequency is a variable, the available time is a dependent variable, and the optimal monitoring frequency gamma meets the following requirements:
γmin≤γ≤γmaxwherein γ ismax、γminRespectively, a maximum value and a minimum value of the monitoring frequency.
10. The system of claim 6, a time of a transition between the different degraded states of the converter valve follows an exponential distribution, a probability of the transition between the different degraded states being constant.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710338753.7A CN107220712B (en) | 2017-05-15 | 2017-05-15 | Converter valve state monitoring and maintenance decision method and system based on Markov process |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710338753.7A CN107220712B (en) | 2017-05-15 | 2017-05-15 | Converter valve state monitoring and maintenance decision method and system based on Markov process |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107220712A CN107220712A (en) | 2017-09-29 |
| CN107220712B true CN107220712B (en) | 2021-11-26 |
Family
ID=59945550
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710338753.7A Active CN107220712B (en) | 2017-05-15 | 2017-05-15 | Converter valve state monitoring and maintenance decision method and system based on Markov process |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107220712B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108320048A (en) * | 2018-01-05 | 2018-07-24 | 中车青岛四方机车车辆股份有限公司 | A kind of wheel Xuan repaiies policy optimization method and device |
| CN108845202B (en) * | 2018-06-15 | 2020-11-17 | 国网山东省电力公司经济技术研究院 | Transformer state overhauling method considering opportunistic maintenance among multiple fault modes |
| CN110018371B (en) * | 2019-03-28 | 2021-08-24 | 国网山东省电力公司青岛供电公司 | Converter station main equipment state maintenance system and implementation method thereof |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102222170A (en) * | 2011-06-21 | 2011-10-19 | 重庆大学 | Markov state space diagram method for evaluating reliability of converter transformer system |
| CN105182749A (en) * | 2015-09-02 | 2015-12-23 | 湘潭大学 | Converter control strategy unification method based on Markov decision process |
| CN106447201A (en) * | 2016-09-29 | 2017-02-22 | 国网山东省电力公司电力科学研究院 | Optimal maintenance decision method of power transmission and transformation equipment based on Markov decision process |
| CN106647263A (en) * | 2016-12-01 | 2017-05-10 | 贵州电网有限责任公司电力科学研究院 | Power equipment maintenance decision-making method utilizing equal degradation theory and equipment risks |
-
2017
- 2017-05-15 CN CN201710338753.7A patent/CN107220712B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102222170A (en) * | 2011-06-21 | 2011-10-19 | 重庆大学 | Markov state space diagram method for evaluating reliability of converter transformer system |
| CN105182749A (en) * | 2015-09-02 | 2015-12-23 | 湘潭大学 | Converter control strategy unification method based on Markov decision process |
| CN106447201A (en) * | 2016-09-29 | 2017-02-22 | 国网山东省电力公司电力科学研究院 | Optimal maintenance decision method of power transmission and transformation equipment based on Markov decision process |
| CN106647263A (en) * | 2016-12-01 | 2017-05-10 | 贵州电网有限责任公司电力科学研究院 | Power equipment maintenance decision-making method utilizing equal degradation theory and equipment risks |
Non-Patent Citations (1)
| Title |
|---|
| 基于Markov过程的电网资产全寿命周期管理的理论研究;吕翔;《工程科技Ⅱ辑 中国优秀硕士论文全文数据库》;20130228;正文第7-53页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107220712A (en) | 2017-09-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | Joint optimization of condition-based opportunistic maintenance and spare parts provisioning policy in multiunit systems | |
| CN105096053B (en) | A kind of health control decision-making technique suitable for complicated technology system | |
| CN107220712B (en) | Converter valve state monitoring and maintenance decision method and system based on Markov process | |
| CN111539515A (en) | Complex equipment maintenance decision method based on fault prediction | |
| CN104390657A (en) | Generator set operating parameter measuring sensor fault diagnosis method and system | |
| CN111652461A (en) | Aero-engine continuous health state evaluation method based on SAE-HMM | |
| CN105825279A (en) | Multi-component system group maintenance decision method and multi-component system group maintenance decision device based on prediction | |
| US11120350B2 (en) | Multilevel pattern monitoring method for industry processes | |
| CN112598144B (en) | CNN-LSTM burst fault early warning method based on correlation analysis | |
| CN105740975A (en) | Data association relationship-based equipment defect assessment and prediction method | |
| CN103679280A (en) | Optimal maintaining method for equipment with performance slow degradation | |
| CN106682159A (en) | Threshold configuration method | |
| Li et al. | Preventive maintenance decision model of urban transportation system equipment based on multi-control units | |
| CN106649755A (en) | Threshold self-adaption setting abnormity detection method for multi-dimensional real-time power transformation device data | |
| CN104899646A (en) | Predictive maintenance method for multi-device hybrid system | |
| CN112287624A (en) | Relay protection system reliability evaluation method based on survival analysis plus GO method | |
| CN104218571A (en) | Running state evaluation method for wind power generation equipment | |
| CN105404940B (en) | Maintenance resource prediction method for ship use stage | |
| CN117354171A (en) | Platform health condition early warning method and system based on Internet of things platform | |
| CN103985059A (en) | Power grid transformer operational risk assessment method based on fuzzy fault rate | |
| Moslemi et al. | A machine learning based demand charge management solution | |
| CN117154945A (en) | Intelligent monitoring method for transformer and distribution station based on cloud computing technology | |
| CN111598328A (en) | Power load prediction method considering epidemic situation events | |
| CN111861181A (en) | Intelligent production system and production line formulation method for digital energy products | |
| CN110909763A (en) | Equipment fault diagnosis method based on fault knowledge base and Bayesian network |
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 |