CN104573315A - Fault rate calculation method for electric transmission and transformation equipment on basis of state overhauling - Google Patents

Abstract

The invention discloses a fault rate calculation method for electric transmission and transformation equipment on the basis of state overhauling, and belongs to the technical field of the state overhauling of the electric transmission and transformation equipment of a power system. On the basis of a total-state integration method, an equivalent state scoring model is introduced in, the fault rate of the equipment is solved by an equivalent state scoring method, and an obtained result of the fault rate is completely the same with the result of the fault rate, which is obtained by the total-state integration method, of the equipment. According to the fault rate calculation method, the fault rates of the equipment under different health processes and states can be accurately calculated, and meanwhile, influence on the fault rate of the equipment by power-off overhauling can be correctly reflected.

Description

A kind of computing method of the power transmission and transforming equipment failure rate based on repair based on condition of component

Technical field

The invention belongs to electric system power transmission and transformation equipment state overhauling technical field, the present invention relates to a kind of computing method of the power transmission and transforming equipment failure rate based on repair based on condition of component more precisely.

Background technology

Repair based on condition of component, it is the status information of equipment that a kind of status monitoring according to advanced person and diagnostic techniques provide, the exception of judgment device, the fault of precognition equipment, the value-at-risk of comprehensive evaluation equipment, formulates the mode of overhauling it by the state evaluation score value (hereinafter referred to as condition grading) obtained.Wherein, condition grading to be patrolled and examined and the mode such as routine test, diagnostic test, on-line monitoring, live detection, family's defect diagonsis, bad condition diagnosis obtains the status information of equipment, comprise and its phenomenon intensity, value size and development trend are evaluated, draw the score value of concrete part of appliance, then the comprehensive grading value of equipment is tried to achieve, in this, as the condition grading of equipment.Estimate that possible equipment failure rate arranges turnaround plan with this according to the condition grading of equipment, the maintenance of facilities and equipments.Thus, ask for according to the condition grading of the equipment obtained the basis that the possible failure rate of equipment is repair based on condition of component.

Document one " the converting equipment life cycle failure rate based on Marquardt method parameter estimation is assessed " (protecting electrical power system and control in January, 2012 number the 4th volume the 1st phase 85-90 page) proposes a kind of computing method of time-based failure rate mathematical model parameter, and author utilizes Marquardt method to solve the failure rate model of the Weibull distribution of two parameter.Time-based probability of equipment failure is calculated according to the Weibull distribution of trying to achieve.

Document two " failure rate of electrical equipment in Decision-making of Condition-based Maintenance calculates " (Automation of Electric Systems in February, 2010 number the 30th volume the 2nd phase 91-94 page) proposes a kind of algorithm of the failure rate model parameter based on equipment state scoring.Article utilizes the method for inversion to try to achieve failure rate model parameter value by obtaining different conditions scoring lower equipment number of units to the statistics of historical data with the total number of units of equipment broken down.Thus the equipment failure rate can tried to achieve according to different equipment states under different conditions scoring.

But in document one, time-based failure rate only considered the impact of time on failure rate, does not consider the effect of condition grading, cannot the failure rate difference of characterization device when identical different conditions scoring working time.And only considered the impact of condition grading on failure rate based on the failure rate of equipment state scoring in document two, do not consider the impact of interruption maintenance.Even if equipment has identical condition grading failure rate after experienced by interruption maintenance be also different.

Document three " the failure rate model parameter study based on equipment complete health process " (east china electric power in September, 2012 number the 40th volume the 8th phase 1346-1349 page) proposes the total state Integration Method based on complete health process.For the equipment that experienced by maintenance of stopping transport, its complete health process refers to from equipment puts into operation again to the whole process occurring end mark.For without the equipment of maintenance with regard to putting into operation, its complete health process refers to from putting into operation to the whole process occurring end mark.

Fig. 1 is the example of equipment complete health process, wherein the condition grading of S indication equipment, the enlistment age (time, generally in units of year or the moon) of t indication equipment.0 moment equipment investment runs.T as shown in Figure 1 ₁, t ₂, t ₃and t ₅in the moment, the existing defects of status monitoring display device, arrangement hotline maintenance or voluntarily defect elimination make the condition grading of equipment be restored to 100 points.Equipment should arrange interruption maintenance when being in abnormality in good time, should arrange interruption maintenance as early as possible, the t shown in Fig. 1 when equipment is in severe conditions ₇moment has carried out interruption maintenance.And t ₄moment is equipment failure.

The complete health process of equipment has two kinds of end marks.

1, equipment failure: to put into operation from new equipment, interruption maintenance or eliminate again to put into operation to after fault and break down.As A point device breaks down in Fig. 1, arrange interruption maintenance immediately, life-span T ₁it is a healthy process.

2, equipment is in exception or severe conditions: put into operation from new equipment, interruption maintenance or eliminate the equipment state scoring that again to put into operation after fault and be in exception or severe conditions, and requires to allow to extend a period of time t according to maintenance _d(t _d=t ₈-t ₆).T as corresponding in B point in Fig. 1 ₆in the moment, status monitoring display device defectiveness, condition grading is in abnormality, at the moment t that C point is corresponding ₇arrange maintenance.In this case, although do not carry out interruption maintenance immediately, equipment has been in equipment failure process, and from enlistment age t corresponding to real inefficacy moment F point ₈very near, relative to longer whole complete health process, t ₈-t ₇can ignore during this period of time.As shown in Figure 1, T ₂it is a complete health process.

When equipment is in exception or severe conditions, interruption maintenance should be arranged in good time.Thus the end mark of complete health process is interruption maintenance in essence.

In practical engineering application, there is certain time interval between adjacent twice condition grading, the measurement data in the complete health process obtained is discrete value.The complete health process of Fig. 2 example has 7 discrete condition grading values.The duration of equal state being marked is added, and just can obtain this figure by the size inverted order arrangement of condition grading.Making p>=1, is integer, and the sequence number of the complete health process that indication equipment experiences, makes i>=1, is integer, represents that equal state scoring is by the sequence number after descending sort, then S _p,icondition grading when indication equipment is in complete health process p during descending sort sequence number i, making N>=1, is integer, represents S in process p _p,inumber.S in Fig. 2 _p,ithe corresponding duration is T _p,i, T _p,ican directly obtain from on-the-spot tracking data.I, p, S of below mentioning _p,i, T _p.iabove-mentioned implication is with N.

Total state Integration Method utilizes whole status informations of equipment complete health process, unified for the duration of each condition grading conversion to the expected time on time shaft, calculate the equipment failure rate model parameter that complete health process is corresponding, then the failure rate parameter based on complete health process that all samples obtain is studied.Specifically, document three proposes a kind of improved model that the condition grading in different complete health process (hereinafter referred to as process) affects equipment failure rate of considering:

${\mathrm{\λ}}_{p,i}=K_p\·e^{{-C}_p\·S_{p,i}}---\left(1\right)$

Wherein:

K _pthe failure rate model scale parameter of the total state Integration Method of the process p that indication equipment experiences.

C _pthe failure rate model curvature parameters of the total state Integration Method of the process p that indication equipment experiences.

λ _p,iexpression condition grading is S _p,itime equipment failure rate.

The K below mentioned _p, C _p, λ _p,ibe above-mentioned implication.

The fault type of equipment can be divided into certainty fault and chance failure according to the difference of fault inducement.Certainty fault is caused by internal cause, relevant to the condition grading situation of change of equipment; And chance failure is caused by external cause (as personnel's maloperation, protection incorrect operation, repair quality are bad).The chance failure rate of equipment can be obtained by statistics, is constant λ ₀.For arbitrary process, when equipment is that full marks 100 timesharing is broken down at condition grading, can think to be exactly that chance failure causes.Therefore, the relation between the scale-up factor of the total state Integration Method that process p is corresponding and coefficient of curvature meets:

$K_p\·e^{-100\·C_p}={\mathrm{\λ}}_0---\left(2\right)$

For continuing the failure rate parameter of the total state Integration Method when equipment that solves is in process p, document three describes following concept and definition.

1, time standard axle T ₀

It refers to that equipment expects life-span at the equipment that condition grading is full marks 100 timesharing, namely supposes that equipment state change does not occur and maintains the operation of full marks state until the time experienced occurs chance failure always, meets:

$T_0=\frac{1}{{\mathrm{\λ}}_0}=\frac{1}{K_p}{e}^{100\·C_p}---\left(3\right)$

Therefore, time standard axle T ₀size determined by chance failure rate.

2, equipment is S at condition grading _p,itime expectation state life-span T (S _p,i)

It refers to that hypothesis equipment is with condition grading S _p,iput into operation and keep this state to run until break down the time of stopping transport and expecting, middle without going through any other state, expectation state life-span T (S _p,i) meet:

$T\left(S_{p,i}\right)=\frac{1}{K_p}{e}^{C_p\·S_{p,i}}---\left(4\right)$

With T ₀difference, T (S _p,i) both can be subject to the impact of chance failure factor, also can be subject to and condition grading S _p,ithe impact of relevant certainty failure factor.

3, based on the conversion factor m (S of time standard axle _p,i)

It refers to that equipment state scoring is the expectation life-span T of 100 timesharing ₀be S with condition grading _p,itime expectation state life-span T (S _p,i) ratio, namely

$m\left(S_{p,i}\right)=\frac{T_0}{T\left(S_{p,i}\right)}=e^{(100-S_{p,i})\·C_p}---\left(5\right)$

Because process p in formula (5) is confirmable.So based on the conversion factor m (S of time standard axle _p,i) only with condition grading S _p,irelevant, only with S _p,ichange.The object introduced based on the conversion factor of time standard axle to comprise the expectation state life-span of conversion for only having chance failure factor to affect in expectation state life-span of certainty failure factor.

4, equipment actual sustain condition grading S _p,iexpectation state duration T on standard time axle _p0i

As shown in Figure 3, it is by condition grading S _p,ithe time standard axle that obtains through the conversion of time standard axle of actual duration on the expectation state duration.Its value is:

T _p0i＝T _p,i·m(S _p,i) (6)

If each condition grading actual duration experienced in each process is carried out the conversion based on time standard axle, the expectation state duration affected by " chance failure factor " that each condition grading on standard time axle is corresponding just can be obtained.In the process life-span that the expectation state duration that all considerations " chance failure factor " corresponding to process affect adds up, the time standard axle T determined by chance failure rate should be equaled ₀:

$\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{T}_{p0i}=T_0---\left(7\right)$

Can be obtained by formula (5), formula (6) and formula (7):

$\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{T}_{p,i}\·e^{(100-S_{p,i})\·C_p}=T_0---\left(8\right)$

The model coefficient of curvature C of total state Integration Method when this equipment is in process p just can be solved by formula (8) _p, then result is substituted into formula (2), just can solve the model scale COEFFICIENT K of total state Integration Method when this equipment is in process p _p.And then can in the hope of the equipment failure rate of different conditions scoring correspondence.

Total state Integration Method result of calculation is accurate, but, also there are some inherent shortcomings:

Computation process is loaded down with trivial details: application total state Integration Method, must by the K corresponding to each process _p, C _pcalculate, required calculated amount is larger.

Physical significance is indefinite: after equipment experienced by interruption maintenance, even if having identical condition grading, corresponding equipment failure rate also should be different, utilizes total state Integration Method to calculate failure rate and can not embody the impact of interruption maintenance on equipment failure rate.

Summary of the invention

The object of the invention is: the deficiency existed for total state Integration Method, proposes a kind of computing method of the power transmission and transforming equipment failure rate based on repair based on condition of component newly.The method introduces equivalent state Rating Model on the basis of total state Integration Method, and utilizes equivalent state point system to solve the failure rate of equipment, and its result is identical with the result of total state Integration Method computing equipment failure rate.

Specifically, the present invention adopts following technical scheme to realize, and comprises the following steps:

1) the random failure rate λ of equipment is added up ₀, all virgin states scoring of equipment when being in each complete health process and the duration of correspondence, use S _p,iwhen indication equipment is in p complete health process, after all virgin state scoring descending sorts, sequence number is the virgin state scoring of i, T _p,irepresent S _p,ithe corresponding duration, p>=1;

Described complete health process, for the equipment that experienced by maintenance of stopping transport, referring to the whole process occurring end mark from equipment puts into operation again, for without overhauling the equipment just put into operation, referring to from putting into operation to the whole process occurring end mark;

2) total state Integration Method is utilized to calculate the failure rate model parameter K of first complete health process that equipment experiences ₁, C ₁, the failure rate model parameter of the equivalent state point system of all processes making equipment experience all and K ₁, C ₁identical, namely meet K' _p=K ₁, C' _p=C ₁, wherein, K' _pthe failure rate model scale parameter of the equivalent state point system of p the complete health process that indication equipment experiences, C' _pthe failure rate model curvature parameters of the equivalent state point system of p the complete health process that indication equipment experiences;

3) first, equivalent state scoring when each complete health process is in for equipment is solved by following formula:

S' _p,i＝a _p·S _p,i+b _p

Wherein, a _pbe the equivalent state Rating Model scale-up factor of p complete health process, b _pbe the constant term of p complete health process, S' _p,iwhen being p complete health process, after all virgin state scoring descending sorts, sequence number is the equivalent state scoring of the virgin state scoring correspondence of i;

Above-mentioned a _pand b _pdetermine by the following method: as p=1, a ₁=1, b ₁=0; As p>1, try to achieve a by two formulas below simultaneous _pand b _p:

$\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{T}_{p,i}{e}^{(100-a_p\·S_{p,i}-b_p)\·C_p^{\′}}=\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{T}_{p,i}{e}^{(100-a_p\·S_{p,i}-b_p)\·C_1}=T_0$

100·a _p+b _p＝100

In above-mentioned formula, N represents S in p complete health process _p,inumber; T ₀the equipment that indication equipment is full marks 100 timesharing at condition grading is expected life-span, is tried to achieve according to following formula:

$T_0=\frac{1}{{\mathrm{\λ}}_0}=\frac{1}{K_p^{\′}}{e}^{100\·C_p^{\′}}=\frac{1}{K_1}{e}^{100\·C_1}$

After equivalent state scoring when the equipment of trying to achieve is in each complete health process, try to achieve the equipment failure rate based on equivalent state scoring by following formula:

${\mathrm{\λ}}_{p,i}^{\′}=K_1\·e^{{-C}_1\·{S^{\′}}_{p,i}}=K_1\·e^{{-C}_1\·(a_p\·S_{p,i}+b_p)}$

Wherein, λ ' _p,irepresent that equivalent state scoring is for S' _p,itime equipment failure rate.

Technique scheme is further characterized in that, described complete health process end mark comprise equipment failure and equipment is in exception or severe conditions.

Beneficial effect of the present invention is as follows: the present invention introduces equivalent state Rating Model on the basis of total state Integration Method, and utilizes equivalent state point system to solve the failure rate of equipment, and its result is identical with the result of total state Integration Method computing equipment failure rate.The present invention correctly can calculate the failure rate under the different healthy process different conditions of equipment, correctly can reflect the impact of interruption maintenance on equipment failure rate simultaneously.

Accompanying drawing explanation

Fig. 1 is equipment complete health process schematic.

Fig. 2 is a complete health process having 7 different conditions scorings.

Fig. 3 is the expectation state duration of actual sustain condition grading on standard time axle.

Fig. 4 is equivalent state point system process flow diagram.

Embodiment

With reference to the accompanying drawings and in conjunction with example, the present invention is described in further detail.

The inventive method proposes on the basis of total state Integration Method, comprises following basic assumption:

1, equipment is in the random failure rate of virgin state scoring to be failure rate corresponding to 100 timesharing be equipment.

2, equipment is operated in stage running-in period being in equipment when putting into operation for the first time, and all the other processes are then operated in the random failure rate stage.

3, put into operation for the first time before not through interruption maintenance, thus using the virgin state scoring in first process (i.e. process 1) experienced after equipment puts into operation for the first time as normal condition scoring (in this method, the condition grading without equivalent process obtained is called that virgin state is marked, the condition grading after equivalent process is then called that equivalent state is marked).The virgin state scoring of process 1 is marked equal with its equivalent state, and the equivalent state scoring of all the other processes and normal condition exist linear relationship between marking.

4, equipment experienced by altogether M (M >=1 and for integer) individual process.

The step of this method as shown in Figure 4, mainly comprises the following steps:

1) the random failure rate of equipment, the virgin state scoring of each process and the duration of correspondence is added up.And use S _p,iwhen indication equipment is in p complete health process, after all virgin state scoring descending sorts, sequence number is the virgin state scoring of i, T _p,irepresent S _p,ithe corresponding duration, p>=1.

2) total state Integration Method is utilized to calculate the failure rate model parameter K of the total state Integration Method of process 1 ₁, C ₁, the failure rate model parameter K' of the equivalent state point system of all processes _p, C' _p(p>=1) all with the failure rate model parameter K of the total state Integration Method of process 1 ₁, C ₁identical, namely meet K' _p=K ₁, C' _p=C ₁.

3) the equivalent state Rating Model solved for each process also utilizes the equivalent state Rating Model of trying to achieve to obtain the failure rate of equipment.

First, equivalent state Rating Model is solved:

In order to embody the impact of interruption maintenance on equipment, demand goes out equipment and marks in the equivalent state of the identical virgin state scoring of various process using process 1 condition grading as benchmark.Equivalent state Rating Model:

S' _p,i＝a _p·S _p,i+b _p(9)

Wherein a _pfor the equivalent state Rating Model scale-up factor of process p, b _pfor the constant term of process p, S' _p,ifor the equivalent state of process p is marked.(a mentioned below _p, b _p, S' _p,ibe above-mentioned implication)

The virgin state scoring of process 1 equals equivalent state scoring, and namely the equivalent state Rating Model of process 1 meets a ₁=1, b ₁=0; For all the other processes, following methods is adopted to solve equivalent state Rating Model:

$\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{T}_{p,i}{e}^{(100-a_p\·S_{p,i}-b_p)\·C_p^{\′}}=\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{T}_{p,i}{e}^{(100-a_p\·S_{p,i}-b_p)\·C_1}=T_0---\left(10\right)$

From document three, equipment state scoring is failure rate corresponding to 100 timesharing is random failure rate, is fixed value.Thus each process virgin state scoring is the equivalent state scoring of 100 points is still 100 points, then have:

100·a _p+b _p＝100 (11)

Simultaneous (10), formula (11) can solve equivalent state Rating Model.

Then, the failure rate of equipment is solved:

By the total state Integration Method failure rate parameter K of process 1 correspondence tried to achieve ₁, C ₁the equivalent state Rating Model of trying to achieve with step 3 brings the equipment failure rate that following formula can try to achieve based on equivalent state scoring into:

${\mathrm{\λ}}_{p,i}^{\′}=K_1\·e^{{-C}_1\·{S^{\′}}_{p,i}}=K_1\·e^{{-C}_1\·(a_p\·S_{p,i}+b_p)}---\left(12\right)$

Wherein λ ' _p,ifor the equipment failure rate utilizing equivalent state point system to try to achieve, represent that equivalent state scoring is for S' _p,itime equipment failure rate.

The following describes coming to the same thing of equivalent state point system and total state Integration Method computing equipment failure rate:

As previously mentioned, as p=1, both result of calculation is identical;

As p>1, to remaining process with the K of process 1 ₁, C ₁based on carry out the equivalence of virgin state scoring, bringing formula (11) into formula (9) can obtain

S' _p,i＝100-(100-S _p,i)·a _p(13)

Bring formula (13) into formula (8) can obtain:

$T_0=\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{T}_{p,i}\·e^{C_1\·(100-S_{p,i}^{\′})}=\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{T}_{p,i}\·e^{C_1\·(100-S_{p,j})\·a_p}---\left(14\right)$

Contrast (8) and formula (14) can find:

C _p＝C ₁·a _p(15)

From formula (2):

$K_1\·e^{-100\·C_1}=K_p\·e^{-100\·C_p}={\mathrm{\λ}}_0---\left(16\right)$

Thus can be obtained by formula (16):

$K_p=K_1\·e^{100\·(a_p-1)\·C_1}---\left(17\right)$

Formula (15) and formula (17) are brought in the failure rate model of formula (1) and arrange and can obtain:

${\mathrm{\λ}}_{p,i}=K_1\·e^{{-C}_1\·(a_p\·S_{p,i}+100-100\·a_p)}---\left(18\right)$

Formula (11) is substituted into:

${\mathrm{\λ}}_{p,i}=K_1\·e^{{-C}_1\·(a_p\·S_{p,i}+b_p)}=K_1\·e^{{-C}_1\·S_{p,i}^{\′}}={\mathrm{\λ}}_{p,i}^{\′}---\left(19\right)$

The equipment failure rate λ ' utilizing equivalent state point system to try to achieve can be derived thus _p,iwith the equipment failure rate λ that total state Integration Method is tried to achieve _p,iidentical.

Although the present invention with preferred embodiment openly as above, embodiment is not of the present invention for limiting.Without departing from the spirit and scope of the invention, any equivalence change done or retouching, belong to the protection domain of the present invention equally.Therefore the content that protection scope of the present invention should define with the claim of the application is standard.

Claims (2)

1., based on computing method for the power transmission and transforming equipment failure rate of repair based on condition of component, it is characterized in that, comprise the steps:

1) the random failure rate λ of equipment is added up ₀, all virgin states scoring of equipment when being in each complete health process and the duration of correspondence, use S _p,iwhen indication equipment is in p complete health process, after all virgin state scoring descending sorts, sequence number is the virgin state scoring of i, T _p,irepresent S _p,ithe corresponding duration, p>=1;

Described complete health process, for the equipment that experienced by maintenance of stopping transport, referring to the whole process occurring end mark from equipment puts into operation again, for without overhauling the equipment just put into operation, referring to from putting into operation to the whole process occurring end mark;

2) total state Integration Method is utilized to calculate the failure rate model parameter K of first complete health process that equipment experiences ₁, C ₁, the failure rate model parameter of all processes making equipment experience all and K ₁, C ₁identical, namely meet K' _p=K ₁, C' _p=C ₁, wherein, K' _pthe failure rate model scale parameter of the equivalent state point system of p the complete health process that indication equipment experiences, C' _pthe failure rate model curvature parameters of the equivalent state point system of p the complete health process that indication equipment experiences;

3) first, equivalent state scoring when each complete health process is in for equipment is solved by following formula:

S' _p,i＝a _p·S _p,i+b _p

Wherein, a _pbe the equivalent state Rating Model scale-up factor of p complete health process, b _pbe the constant term of p complete health process, S' _p,iwhen being p complete health process, after all virgin state scoring descending sorts, sequence number is the equivalent state scoring of the virgin state scoring correspondence of i;

Above-mentioned a _pand b _pdetermine by the following method: as p=1, a ₁=1, b ₁=0; As p>1, try to achieve a by two formulas below simultaneous _pand b _p:

$\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{T}_{p,i}{e}^{(100-a_p\·S_{p,i}-b_p)\·C_p^{\′}}=\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{T}_{p,i}{e}^{(100-a_p\·S_{p,i}-b_p)\·C_1}=T_0$

100·a _p+b _p＝100

In above-mentioned formula, N represents S in p complete health process _p,inumber; T ₀the equipment that indication equipment is full marks 100 timesharing at condition grading is expected life-span, is tried to achieve according to following formula:

$T_0=\frac{1}{{\mathrm{\λ}}_0}=\frac{1}{K_p^{\′}}{e}^{100C_p^{\′}}=\frac{1}{K_1}{e}^{100C_1}$

After equivalent state scoring when the equipment of trying to achieve is in each complete health process, try to achieve the equipment failure rate based on equivalent state scoring by following formula:

${\mathrm{\λ}}_{p,i}^{\′}=K_1\·e^{-C_1\·{S^{\′}}_{p,i}}=K_1\·e^{-C_1\·(a_p\·S_{p,i}+b_p)}$

Wherein, λ ' _p,irepresent that equivalent state scoring is for S' _p,itime equipment failure rate.

2. the computing method of the power transmission and transforming equipment failure rate based on repair based on condition of component according to claim 1, is characterized in that: described complete health process end mark comprise equipment failure and equipment is in exception or severe conditions.