DESCRIPTION

The present invention relates to a method for assessing health conditions of industrial equipment.

The method according to the invention is particularly suitable for performing a health assessment of a low, medium or high voltage electric apparatus or system, e.g. a switching device, a switchgear, an electric power distribution grid and the like.

For the purposes of the present invention:

- the term "low voltage" (LV) relates to operating voltages at electric power distribution level, which are lower than 1 kV AC and 1.5 kV DC;

- the term "medium voltage" (MV) relates to operating voltages at electric power distribution level, which are higher than 1 kV AC and 1.5 kV DC up to some tens of kV, e.g. up to 72 kV AC and 100 kV DC;

- the term "high voltage" (HV) relates to operating voltages at electric power transmission level, which are higher than some tens of kV, e.g. higher than 72 kV AC and 100 kV DC.

As is known, currently available methods for performing a health assessment of industrial equipment are basically of two types: time-based assessment methods and condition-based assessment methods.

Typically, time-based assessment methods calculate the conditional probability of failure of the equipment within a predefined time window, given the age thereof.

Such a conditional probability of failure, which is generally calculated for all the failure modes potentially leading to a failure of the equipment, represents a measure of the health conditions of said equipment.

Time -based assessment methods are generally based on the statistical analysis (life data analysis) of the equipment lifetime and do not take into account the actual usage and degradation history of the equipment.

It has been seen that this may lead to a large inaccuracy in the assessment activity, when the equipment usage and degradation history deviates significantly from the statistical average for the class of systems including the considered equipment.

Further, these methods do not often provide accurate indications on possible preventive maintenance activities to be carried out to avoid or reduce the occurrence of failures in the equipment.

Condition-based assessment methods typically assess health conditions of the equipment by processing monitoring data, which are related to the actual operating conditions of said equipment. Such an assessment is generally carried out by means of model-based and/or data- driven algorithms.

As they are based on the processing of monitoring data related to the actual usage and degradation history of the equipment, these methods are particularly suitable to implement condition-based maintenance policies.

Unfortunately, differently from the above mentioned time -based assessment methods, these methods generally analyze only few possible failure modes potentially leading to a failure of the equipment.

Further, they allow assessing health conditions of industrial equipment only on the basis of the operating status of components or groups of components for which said monitoring data are available.

It has been seen that both these critical aspects may lead to a large inaccuracy in the assessment activity.

In addition, current condition-based assessment methods typically provide no information on the accuracy of their assessment data and on how to improve this latter.

In the state of the art, it is therefore quite felt the need for technical solutions that are capable of overcoming the technical issues described above, thereby providing high levels of accuracy in assessing current or future health conditions of industrial equipment, identifying failures in said equipment and providing reliable information on possible preventive maintenance actions to avoid potential failures in said equipment.

In order to respond to this need, the present invention provides a method for assessing health conditions of industrial equipment, according to the following claim 1 and the related dependent claims.

In a general definition, the method, according to the invention, comprises the following steps:

- acquiring input data related to the equipment;

- calculating failure mode assessment data for each failure mode determined for said equipment, wherein the calculation of said failure mode assessment data comprises:

- if said failure mode has one or more determined failure causes:

- executing a first calculation procedure to calculate failure cause assessment data for each failure cause determined for said failure mode, said failure cause assessment data being calculated on the basis of said input data;

- calculating said failure mode assessment data on the basis of the failure cause assessment data calculated for each failure cause determined for said failure mode;

- if said failure mode has one or more undetermined failure causes, executing a second calculation procedure to calculate said failure mode assessment data, said failure mode assessment data being calculated on the basis of said input data;

- calculating system assessment data for said equipment, said system assessment data being calculated on the basis of said failure mode assessment data.

Further characteristics and advantages of the method, according to the present invention, will become more apparent from the detailed description of preferred embodiments illustrated only by way of non-limitative example in the accompanying drawings, in which:

Figures 1-7 are block diagrams that schematically show the method, according to the present invention.

Referring to the cited figures, the present invention is related to a method 1 for assessing health conditions of industrial equipment S.

In principle, the equipment S may be any industrial apparatus or system having components or group of components interacting one with another.

Preferably, the equipment S is a LV, MV or HV electric apparatus or system, such as a LV,

MV or HV switching device, switchgear, electric power distribution grid and the like.

In general, the method, according to the invention, is based on the following theoretical assumptions:

- any industrial equipment may be subject to a failure;

- one or more mutually exclusive "failure modes" concur to cause a failure of said equipment;

- one or more mutually exclusive failure causes concur to be at the origin of each failure mode.

Within the frame of the present invention, a "failure mode" of industrial equipment is defined as a determinable operating way, in which said equipment is subject to a failure.

As an example, in a MV circuit breaker, a failure mode may be the following:

"the electric poles of the circuit breaker are no more able to interrupt short-circuit currents when opening their electric contacts ".

According to another possible approach (which is complementary to the one mentioned above), a "failure mode" of industrial equipment is defined as a determinable component or group of components of said equipment, which may be subject to a failure.

As an example, in a MV circuit breaker, possible failure modes may be the following:

"the electric poles of the circuit breaker" or "the actuating mechanism of the movable electric contacts of the circuit breaker".

One or more failure modes may be determined for a given equipment. Such failure modes are assumed to be mutually exclusive and statistically independent in the sense that any failure of said equipment is assumed to be caused by only one of said failure modes, thereby excluding the others.

In practice, the failure modes of industrial equipment are assumed to "compete" one with another to cause the failure of said equipment.

Within the frame of the present invention, a "failure cause" for industrial equipment is defined as a determinable or undeterminable physical phenomenon or technical circumstance, which may be at the origin of a given failure mode determined for said equipment.

As an example, referring again a MV circuit breaker and to its possible failure mode mentioned above, possible failure causes at the origin of such a failure mode may be:

"wear phenomena involving the electric contacts"; or

"an unbalanced transmission of mechanical forced through the kinematic chain actuating the movable contacts" .

For some failure modes, which has been determined for industrial equipment, one or more failure causes may be determined.

In this case, the determined failure causes are assumed to be mutually exclusive in the sense that the corresponding failure mode is caused by only one of said failure causes, thereby excluding the others.

In practice, the failure causes referred to a given failure mode are assumed to "compete" one with another to be at the origin of a given failure mode.

Other failure modes, which has been determined for industrial equipment, may be however be originated by undeterminable failure causes that cannot be identified with certainty or somehow monitored.

In view the above, it can be easily understood that, for the equipment S, one or more failure modes are determinable while each of said one or more failure modes may have one or more determinable failure causes or one or more undeterminable failure causes.

According to the invention, the equipment S has one or more determined failure modes F _l5 F , each having in turn one or more determined failure causes FC _DI , · ··, FC _DK and/or one or more undetermined failure causes FCui, FC _UM (figure 2).

Preferably, the failure modes of the equipment S and, when possible, one or more failure causes FC _DI , · ··, FC _DK related to one or more failure modes are predetermined, e.g. established when the equipment S is installed on the field depending on the characteristics of the equipment itself.

Said predetermined failure modes and/or failure causes of the equipment S may be updated, when necessary, during the operating life of the equipment S.

According to the invention, the method 1 comprises a step 2 of acquiring input data Di _N related to the equipment S.

The input data D _I may be data related to the specific equipment S of interest. In addition or in alternative, the input data D _I may include data related to the class of equipment including the equipment S (e.g. data related to the family of MV circuit breakers including a specific circuit breaker of interest).

The input data D _IN may be provided by different sources and their nature may depend on the type of the equipment S, on the managing policies adopted for the equipment S, on the maintenance policies adopted for the equipment S, and the like.

As an example, the input data Di _N may include various subsets of data, e.g. customer data, manufacturer data, user data and external data.

Examples of information included in each subset of data, in the exemplary case in which the equipment S is a MV circuit breaker, are reported in the table of figure 3.

Of course, the input data Di _N preferably include information on the failure modes F _1; F _N and on possible failure causes FC _DI , ·· ·, FC _DK determined for the equipment S.

However, the input data D _IN may include different combinations and subsets of data with respect to those mentioned above.

According to the invention, the method 1 comprises a step 3 of calculating failure mode assessment data Rpi, RULpi, Api, RSKpi, POFpi for each failure mode Fj determined for the equipment S.

Preferably, a set of said failure mode assessment data calculated for a i-th failure mode Fj comprises:

- a first failure mode assessment value Rpi indicative of the reliability of the failure mode Fj. The value Rpi is the probability that the failure mode Fj does not lead to a failure in the equipment S within a specified operation time window;

- a second failure mode assessment value RUL _Fi indicative of the remaining useful life of the equipment S in relation to the failure mode Fj. The value RULpi can be considered as an intervention time window to prevent the occurrence of a failure in relation to the failure mode Fj in the equipment S. In this case, said intervention time window can be defined as the time interval within which to execute an intervention action on the equipment S to prevent a failure in relation to the failure mode Fj;

- a third failure mode assessment value Api indicative of a recommended intervention action on the equipment S to prevent the occurrence of a failure in relation to the failure mode Fj in the equipment S;

- a fourth failure mode assessment value POF _F i indicative of a probability that the failure mode Fj leads to a failure in the equipment S within the above operation time window;

- a fifth failure mode assessment value RSKpi indicative of a risk of failure for the equipment S in relation to the failure mode Fj.

The calculation of the failure mode assessment data R _Fi , RUL _Fi , A _Fi , POF _Fi , RSK _Fi for each failure mode F _; depends on whether the failure mode F _; has determined failure causes FC _DI ,

FC _DK or undetermined failure causes FCui, ···, FCU _M (figure 4).

If a i-th failure mode Fj has one or more determined failure causes FC _DI , ·· ·, FC _DK , the method 1 comprises a step 31 of executing a first calculation procedure 11 to calculate failure cause assessment data R _F q, RULpq, Apq for each failure cause FCo _j determined for the failure

The failure cause assessment data R _F q, RULpq, Apq are calculated for each determined failure cause on the basis of the input data D _I .

Preferably, a set of failure cause assessment data calculated for the j-th determined failure cause FCo _j comprises:

- a first failure cause assessment value Rpq indicative of a probability that the failure cause FCo _j does not lead to a failure in the equipment S within the above operation time window;

- a second failure mode assessment value RULpq indicative of an intervention time window T _MA to prevent the occurrence of the failure cause FCo _j in the equipment S. In this case, said intervention time window can be defined as the time interval within which to execute an intervention action to prevent the occurrence of the failure cause FCo _j in the equipment S;

- a third failure mode assessment value Apq indicative of a recommended intervention action on the equipment S to prevent the occurrence of the failure cause FCo _j in the equipment S.

Following the execution of the calculation procedure 11 for each failure cause FCo _j determined for a i-th failure mode Fj, the method 1 comprises a step 32 of calculating the failure mode assessment data Rpj, RULpj, Apj, RSKpj, POFpj for the failure mode Fj on the basis of the failure cause assessment data Rpq, RULpq, Apq calculated for each failure cause FCoj determined for said failure mode.

Preferably, each of the failure mode assessment values Rpj, RULpj, Apj, RSKpj, POFpj are calculated on the basis of the failure cause assessment values Rpq, RULpq, Apq calculated for each determined failure cause FCo _j related to the failure mode Fj. Preferably, the first failure mode assessment value R _F j for the failure mode Fj is calculated as:

R.Fi = RFCI * RFC2 * ··· *R-FCk

where RFCI, RFC2, · ·· , RFO _C are the first failure cause assessment values Rpc calculated for each failure cause FCDI, FCD ₂ , ·· ·, FCDK determined for the failure mode Fj, according to the first calculation procedure 11.

If some failure causes are not statistically independent, correlation factors modeling the statistically dependency of the failure causes must be considered in the above relation.

Preferably, the second failure mode assessment value RULpi for the failure mode Fj is calculated as:

RUL _Fi = min [RUL _FC1 , RUL _FC2 , ... , RUL _FCk ]

where RULp _C i, RUL _FC 2, ... RUL _FCk are the second failure cause assessment values RUL _FC calculated for each determined failure cause FCDI, FCD ₂ , · ··, FCDK determined for the failure mode Fj, according to the first calculation procedure 11.

Preferably, the third failure mode assessment value A _F j is given by the recommended action Apq associated to the determined failure cause F _C D which has the minimum second failure cause assessment value RULpq among all the determined failure causes FCDI, FCD2, · ··, FCDK related to the failure mode Fj.

Preferably, the action Apq is included in the input data DI .

Preferably, the fourth failure mode assessment value POF _Fi for the failure mode F _; is calculated as:

POF _Fi = 1 - Rpi

where R _F j is the above first failure mode assessment value related to the failure mode Fj. Preferably, the fifth failure mode assessment value RSK _Fi for the failure mode F _; is calculated as:

RSK _Fi = POF _Fi * C _Fi

where POF _F j is the above fourth failure mode assessment value and CR is a cost value associated to the failure mode F _; , which is indicative of the overall cost of a failure of the equipment S when said failure is due to the failure mode Fj.

Preferably, the cost value CR is included in the input data DI .

If a i-th failure mode Fj has undetermined failure causes FCui, ·· ·, FCUM, the method 1 comprises the step 33 of executing a second calculation procedure 12 to calculate the failure mode assessment data RR, RULR, AR, RSKR, POFR for the failure mode Fj.

In this case, the failure mode assessment data RR, RULR, AR, RSKR, POFR are calculated on the basis of said input data DIN. According to the invention, following the calculation of the failure mode assessment data Rpi, RULpi, A _F i, POFpi, RSKpi for each failure mode F _; , the method 1 comprises a step 4 of calculating system assessment data Rs, RULs, As, RSKs, POFs, His for the equipment S. Said set of system assessment data is calculated on the basis of the failure mode assessment data Rpi, RULpi, Ακ, RSKpi, POFpi calculated for each failure mode Fj.

Preferably, said set of system assessment data calculated comprises:

- a first system assessment value R _s indicative of the reliability of the equipment S. The value Rs is the probability that a failure does not occur in the equipment S within a specified operation time window;

- a second system assessment value RULs indicative of the remaining useful life of the equipment S. The value RUL _S can be considered as an intervention time window to prevent the occurrence of a failure in the equipment S. In this case, said intervention time window can be defined as the time interval within which to execute an intervention action to prevent the occurrence of a failure in the equipment S;

- a third system assessment value A _s indicative of a recommended intervention action on the equipment S to prevent the occurrence a failure in the equipment S;

- a fourth failure mode assessment value POFs indicative of a probability that a failure occurs within the above operation time window;

- a fifth system assessment value RSK _S indicative of a risk of failure for the equipment S;

- a sixth system assessment value His indicative of the health level of the equipment S.

Preferably, each of the system assessment values Rs, RULs, As, RSKs, POFs, His are calculated on the basis of the failure mode assessment values Rpi, RULpi, Apj, RSKpi, POFpi calculated for each failure mode F _l5 F _N .

Preferably, the first system assessment value Rs for the equipment S is calculated as:

Rs = RFI * RF2 * ··· * RFN

where Rpi, Rp ₂ , ... _, R _FN are the first failure mode assessment values calculated for each failure mode of the equipment S.

If some failure modes are not statistically independent, correlation factors modeling the statistically dependency of the failure causes must be considered in the above relation.

Preferably, the second system mode assessment value RULs for the equipment S is calculated as:

RULs = min [RUL _F1 , RUL _F2 , ... , RUL _FN ]

where RULpi, RULp ₂ , ... _, RUL _F are the second failure mode assessment values calculated for each failure mode of the equipment S. Preferably, the third system assessment value As for the equipment S is given by the recommended action associated to the failure mode F _; , which has the minimum second failure mode assessment value RULpi among all the determined failure modes Fi, F of the equipment S.

Preferably, the action code ACTs is included in the input data D _I .

Preferably, the fourth system assessment value POF _s for the equipment S is calculated as:

POF _s = l - Rs

where Rs is the above first system assessment value related to the equipment S.

Preferably, the fifth system assessment value RSKs for the equipment S is calculated as:

RSKs = POF _s * Cs

where POF _s is the above fourth system assessment value and Cs is a cost value associated a failure of the equipment S.

Preferably, the cost value Cs is included in the input data D _I .

Preferably, the sixth system assessment value His for the equipment S is calculated by comparing the above fourth system assessment value POF _s with predefined first threshold values and assigning predefined numbers or codes to the sixth system assessment value His on the basis of said comparison.

As an example:

if POFs < 3% , then HI _S = 1;

if 3% <= POFs <= 7% , then HI _S = 2;

if POFs > 7% , then HI _S = 3.

Preferably, the control procedure 11 executed for each determined failure cause FCo _j related to a i-th failure mode F _; comprises different calculation sessions depending on whether monitoring data CMD, NCMD related to the operating status of the equipment S are available or not.

As reported in figure 3, possible monitoring data CMD, NCMD processed during the control procedure 11 are preferably comprised, if available, in the input data Di _N acquired at the acquisition step 2 described above.

If monitoring data CMD, NCMD related to the operating life of the equipment S are not available, the control procedure 11 comprises the step 111 of selecting first reliability statistics data REL1 related to the operating life of the equipment S.

As it may be derived from the table of figure 3, the first reliability statistics data REL1 are preferably comprised in the input data D _IN acquired at the acquisition step 2 described above. Examples of reliability statistics data are reported in the table of figure 3 as well. Then, the control procedure 11 comprises the step 112 of executing a first calculation session 21 to calculate the failure cause assessment data R _F q, RUL _F q, A _F q for the j-th failure cause FCo _j on the basis of the first reliability statistics data RELl.

Preferably, the first calculation session 21 comprises a step 221 of calculating a first conditional reliability function CRF1 for the j-th failure cause FCo _j under consideration on the basis of the first reliability statistics data RELl and the step 222 of calculating the failure cause assessment data R _F q, RUL _F q, A _F q on the basis of said first conditional reliability function CRF1.

As an example, the first conditional reliability function CRF1 may be derived from a Weibull function of the type:

(where a is the scale parameter and β is the shape parameter) by applying a formula for the determination of a conditional reliability function CRF as given by:

P T > t + τ) _ R (t + τ)

CRF(_r I t) = P T > t + r \ T > t) =

P(T > t) ^~ R (t)

where T is the current age of the equipment and τ is a time with τ > T.

The conditional reliability function for the j-th failure cause values FCo _j , R _FQ is given by the value of the function CRF1 for a time equal to the above operation time window. The value RUL _F q is calculated as the mean of the conditional probability density function obtained by deriving the conditional reliability function CRF1.

Preferably, the value A _F q is instead included in the input data D _I and can be selected on the base of the other failure cause assessment values.

If monitoring data CMD, NCMD related to the operating life of the equipment S are available, the control procedure 11 preferably comprises different calculation sessions depending on the available type of monitoring data.

The available monitoring data, in fact, may comprise continuous monitoring data CMD or non-continuous monitoring data NCMD related to the operating status of the equipment S. Advantageously, the continuous monitoring data CMD are data related to the operating status of the equipment S, which are provided by a detection arrangement (e.g. including suitable sensors or detection devices) capable of collecting data related to the operating status of the equipment S in a continuous manner (e.g. cyclically at given detection instants or as a consequence of an event) during the operation of the equipment S, whereas the non- continuous monitoring data NCMD are data related to the operating status of the equipment S, which are provided by a detection arrangement capable of collecting data related to the operating status of the equipment S only when it is activated to perform a test on the operating status of the equipment S.

If continuous monitoring data CMD related to the operating life of the equipment S are not available, the calculation procedure 11 comprises the step 113 of selecting second reliability statistics data REL2 related to the operating life of the equipment S.

The second reliability statistics data REL2 are preferably comprised in the input data Di _N acquired at the acquisition step 2 described above.

Then, the control procedure 11 comprises the step 114 of executing a second calculation session 22 to calculate the failure cause assessment data R _FQ , RULpq, Apq for the j-th failure cause FCo _j under consideration on the basis of the second reliability statistics data REL2 and on the basis of the non-continuous monitoring data NCMD comprised in the available monitoring data.

Preferably, the second calculation session 22 comprises the step 221 of calculating first status data SD1 related to the operating status of the equipment S of the base of the non-continuous monitoring data NCMD.

A first diagnostic algorithm comparing the non-continuous monitoring data NCMD with predefined second threshold values may be advantageously adopted to calculate the status data SD1.

The calculation of the first status data SD1 is aimed at obtaining a quantitative information on the actual status of the equipment S, thereby taking into account possible degradation phenomena of the performances provided by the equipment S.

Preferably, the second calculation session 22 comprises the step 222 of calculating a second conditional reliability function CRF2 for the j-th failure cause FCo _j under consideration on the basis of the second reliability statistics data REL2.

Also the second conditional reliability function CRF2 may be a Weibull function of the type illustrated above.

Preferably, the second calculation session 22 comprises the step 223 of calculating a third conditional reliability function CRF3 for the j-th failure cause FCo _j under consideration.

Also the third conditional reliability function CRF3 may be a Weibull function of the type illustrated above.

Preferably, the third conditional reliability function CRF3 is calculated by modifying said second conditional reliability function CRF2 on the basis of the first status data SD1.

In practice, the second conditional reliability function CRF2 is translated (or more generally transformed) using the first status data SD1 in order to take into consideration the actual operating conditions of the equipment S.

Preferably, the second calculation session 22 comprises the step 224 of calculating the failure cause assessment data Rpq, RULpq, Apq for the failure cause FCo _j under consideration on the basis of said third conditional reliability function CRF3.

The failure cause assessment data Rpq, RULpq, Apq may be obtained from the third conditional reliability function CRF3 in a way similar to the example described above with reference to the first conditional reliability function CRF1.

If continuous monitoring data CMD related to the operating life of said equipment are available, the calculation procedure 11 comprises the step 115 of executing a third calculation session 23 to calculate the failure cause assessment data Rpq, RULpq, Apq for the j-th failure cause FCo _j under consideration on the basis of said continuous monitoring data.

Preferably, the third calculation session 23 comprises the step 231 of calculating second status data SD2 related to the operating status of the equipment S of the base of the non-continuous monitoring data CMD.

A second diagnostic algorithm comparing the continuous monitoring data CMD with predefined third threshold values may be advantageously adopted to calculate the status data SD2.

Also in this case, the calculation of the status data SD2 is aimed at obtaining a quantitative information on the actual status of the equipment S, thereby taking into account possible degradation phenomena of the performances provided by the equipment S.

Preferably, the third calculation session 23 comprises the step 232 of calculating the failure cause assessment data Rpq, RULpq, Apq for the j-th failure cause FCo _j under consideration on the basis of the second status data SD2.

A first prognostic algorithm using Monte-Carlo simulations and/or descriptive statistics calculations may be advantageously adopted to calculate the failure cause assessment data Rpq, RULpq, Apq.

Preferably, the control procedure 12 comprises different calculation sessions depending on whether monitoring data CMD, NCMD related to the operating status of the equipment S are available or not.

If monitoring data CMD, NCMD related to the operating life of the equipment S are not available, the control procedure 12 comprises the step 121 of selecting third reliability statistics data REL3 related to the operating life of the equipment S.

The third reliability statistics data REL3 are preferably comprised in the input data D _I acquired at the acquisition step 2 described above. Then, the control procedure 12 comprises the step 122 of executing a fourth calculation session 24 to calculate the failure mode assessment data R _Fi , RUL _Fi , A _Fi , RSK _Fi , POF _Fi for the i-th failure node Fj under consideration on the basis of the third reliability statistics data REL3.

Preferably, the fourth calculation session 24 comprises a step 241 of calculating a fourth conditional reliability function CRF4 for the i-th failure mode F _; under consideration on the basis of the third reliability statistics data REL3 and the step 242 of calculating the failure mode assessment data R _F j, RULpi, Ακ, RSKpi, POFpj on the basis of said fourth conditional reliability function.

Also the fourth conditional reliability function CRF4 may be a Weibull function of the type illustrated above.

The failure mode assessment data Rpi, RULpi, Apj, RSKpi, POFpj may be obtained from the fourth conditional reliability function CRF4 in a way similar to the example described above with reference to the first conditional reliability function CRF1.

If said monitoring data CMD, NCMD related to the operating life of the equipment S are available, the control procedure 12 preferably comprises different calculation sessions depending on the available type of monitoring data.

If continuous monitoring data CMD related to the operating life of said equipment are not available, the calculation procedure 12 comprises the step 123 of selecting fourth reliability statistics data REL4 related to the operating life of the equipment S.

The fourth reliability statistics data REL4 are preferably comprised in the input data D _I acquired at the acquisition step 2 described above.

Then, the control procedure 12 comprises the step 124 of executing a fifth calculation session 25 to calculate the failure mode assessment data Rpi, RULpi, Ακ, RSKpi, POFpj for the i-th failure node Fj under consideration on the basis of the fourth reliability statistics data REL4 and on the basis of the non-continuous monitoring data NCMD comprised in the available monitoring data.

Preferably, the fifth calculation session 25 comprises the step 251 of calculating third status data SD3 related to the operating status of the equipment S of the base of the non-continuous monitoring data NCMD.

A third diagnostic algorithm comparing the non-continuous monitoring data NCMD with predefined fourth threshold values may be advantageously adopted to calculate the status data SD3.

The calculation of the first status data SD3 is aimed at obtaining a quantitative information on the actual status of the equipment S, thereby taking into account possible degradation phenomena of the performances provided by the equipment S.

Preferably, the fifth calculation session 25 comprises the step 252 of calculating a fifth conditional reliability function CRF5 for the i-th failure mode Fj under consideration on the basis of the fourth reliability statistics data REL4.

Also the fifth conditional reliability function CRF5 may be a Weibull function of the type illustrated above.

Preferably, the fifth calculation session 25 comprises the step 253 of calculating a sixth conditional reliability function CRF6 for the i-th failure mode under consideration.

Also the sixth conditional reliability function CRF6 may be a Weibull function of the type illustrated above.

Preferably, the sixth conditional reliability function CRF6 is calculated by modifying said fifth conditional reliability function CRF5 on the basis of the status data SD3.

In practice, the fifth conditional reliability function CRF5 is translated (or more generally transformed) using the third status data SD3 in order to take into consideration the actual operating conditions of the equipment S.

Preferably, the fifth calculation session 25 comprises the step 254 of calculating the failure mode assessment data Rpi, RULpi, Ακ, RSKpi, POFpj for the i-th failure node Fj under consideration on the basis of said sixth conditional reliability function CRF6.

The failure mode assessment data Rpi, RULpi, Apj, RSKpi, POFpj may be obtained from the sixth conditional reliability function CRF6 in a way similar to the example described above with reference to the first conditional reliability function CRF1.

If continuous monitoring data CMD related to the operating life of said equipment are available, the calculation procedure 12 comprises the step 125 of executing a sixth calculation session 26 to calculate the failure mode assessment data Rpi, RULpi, Apj, RSKpi, POFpj for the i-th failure node Fj under consideration on the basis of said continuous monitoring data.

Preferably, the sixth calculation session 26 comprises the step 261 of calculating fourth status data SD4 related to the operating status of the equipment S of the base of the non-continuous monitoring data CMD.

A fourth diagnostic algorithm comparing the continuous monitoring data CMD with predefined fifth threshold values may be advantageously adopted to calculate the status data SD4.

Also in this case, the calculation of the status data SD4 is aimed at obtaining a quantitative information on the actual status of the equipment S, thereby taking into account possible degradation phenomena of the performances provided by the equipment S.

Preferably, the third calculation session 26 comprises the step 262 of calculating the failure mode assessment data R _F j, RULpi, Ακ, RSKpi, POFpj for the i-th failure node Fj under consideration on the basis of the fourth status data SD4.

A second prognostic algorithm using Monte-Carlo simulations and/or descriptive statistics calculations may be advantageously adopted to calculate the failure mode assessment data R _Fi , RUL _Fi , A _Fi , RSKpi, POFpi.

In a preferred embodiment, the method 1 comprises also the step 5 of calculating accuracy data Aces indicative of the accuracy of the system assessment data Rs, RULs, As, RSKs, POFs, His described above.

Preferably, the accuracy data Acc _s are calculated only when monitoring data NCMD, CMD (non-continuous or continuous) related to the operating life of the equipment S are available. The calculation of the accuracy data Accs is based on the calculation of accuracy data Accpq indicative of the accuracy of said failure cause assessment data R _F q, RULpq, Apq and on the calculation of accuracy data Acc _Fi indicative of the accuracy of said failure mode assessment data R _Fi , RUL _Fi , A _Fi , RSK _Fi , POF _Fi .

The calculation of the accuracy data Accpq _, Accpi is in turn based on the accuracy of the diagnostic algorithms adopted to calculate the status data SDl, SD2, SD3, SD4 described above. The accuracy of such diagnostic algorithms is known per se and it is an information advantageously included in the input data DI .

The step 5 of calculating the accuracy data Accs comprises a step 51 of calculating an accuracy value Acc _F q for each determined failure cause FCo _j taken into consideration during the calculation of the system assessment data R _s , RUL _S , A _s , RSK _S , POF _s , HI _S described above.

The accuracy value Acc _F q for each failure cause FCo _j may be calculated as:

Accpcj = DA _FCj * POFpcj

where POFpq =(1- R _F q) is a value indicative of a probability that said determined failure cause FCo _j occurs within the above operation time window and DA _F q is the accuracy of the diagnostic algorithms used to calculate the failure cause assessment data R _F q, RUL _F q, A _F q related to the determined failure cause FCo _j .

The step 5 of calculating the accuracy data Acc _s comprises a step 52 of calculating an accuracy value Accpi for each failure mode Fj taken in to consideration during the calculation of the system assessment data Rs, RULs, As, RSKs, POFs, His described above.

The accuracy value Accpi for each failure mode Fj may be calculated as: Accn = DA _f i * POF _f i

where POF _F i is the fourth failure mode assessment value calculated for the failure mode F _; and DApi is the accuracy value of the diagnostic algorithms used to calculate the failure mode assessment data Rpi, RULpi, Ακ, RSKpj, POFpj related to the failure mode Fj.

The step 5 of calculating data accuracy Aces comprises a step 53 of calculating an accuracy value Acc _s indicative of the accuracy of the system assessment data R _s , RUL _S , A _s , RSK _S , POF _s , HI _S described above on the basis of the accuracy value Acc _Fi , Acc _F q respectively calculated for each failure mode Fj and each failure cause FCo _j taken into consideration during the calculation of said system assessment data.

The accuracy value Accs may be calculated as the sum of the accuracy values Accpi, Accpq calculated for each failure cause F and for each failure mode F _; taken into consideration during the calculation of the system assessment data Rs, RULs, As, RSKs, POFs, His described above.

The method, according to the invention, is particularly suitable to be implemented by a computerised arrangement.

In a further aspect, the present invention thus relates to a computer program comprising software instructions to carry out the method, according to the invention.

Said computer program is stored or storable in a storage medium, e.g. in a memory of a computerised device.

In a further aspect, the present invention further relates also to a computerised device or platform comprising computerised resources (e.g. one or more microprocessors) configured to execute software instructions to carry out the method, according to the invention.

Such a computerised device or platform may have various data processing architectures, according to the needs, e.g. centralised, multi-level or distributed data processing architectures.

Such a computerised device or platform may be installed on the field or be in a remote location with respect to the equipment S or be distributed on different field or remote devices. The method, according to the invention, allows obtaining diagnostic information on the actual health conditions of the equipment S and identifying possible faults in the equipment S.

The method, according to the invention, allows obtaining prognostic information on the future health conditions of the equipment S, thereby allowing predicting the performances and health conditions of the equipment S over time.

The method, according to the invention, allows identifying and prescribing the most appropriate maintenance actions to avoid or mitigate potential failures (i.e. prescriptive analytics).

The method, according to the invention, allows obtaining a quantitative evaluation of the accuracy of the calculations carried out to assess the health conditions of the equipment S, thereby providing a feedback on the overall quality of the assessment activity.

The method, according to the present invention, is particularly adapted to be implemented by various computerised architectures and it is easily scalable and tunable in accordance to the characteristics of the equipment S and on the available input data Di _N .

The method, according to the invention, can thus be easily implemented at industrial level, at competitive costs with the currently available technical solutions.