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
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
  • G06N7/00—Computing arrangements based on specific mathematical models
  • G06N7/01—Probabilistic graphical models, e.g. probabilistic networks

Definitions

• the field of the invention is that of the maintenance of an equipment likely to go through a degraded state before failing.
• the control of the failure consists in a conventional way to:
• Manageable failures are breakdowns of wear and fatigue. They are present on the mechanical, hydraulic ... In the following we will take a tire or optronic equipment as examples of equipment.
• Corrective maintenance consists of repairing the equipment after the failure has occurred, which is not really optimal.
• Preventive maintenance consists in determining a fixed schedule whatever the equipment. For example, it will be advisable to check or change tires every 20,000 km.
• Predictive maintenance consists of triggering maintenance when a predetermined threshold is reached, such as a physical wear threshold.
• a predetermined threshold such as a physical wear threshold.
• the tire pressure is too low, it is recommended to inflate it to avoid breaking the tire and having to change it, which can happen well before the 20,000 km for a sporty driving tire and well after for a quiet driving tire.
• a maintenance date can be calculated based on a probability of degradation. This calculation describes in particular in the publication "Optimization of the maintenance of an optronic equipment" Camille Baysse et al. Congress of Risk Management and Dependability 1 6-18 October 2012, is carried out according to the estimated state of the equipment at a given moment, the history of these states and a desired performance.
• An optronic equipment for example, has a logbook which provides the following information at each start-up: the number of uses, the accumulated operating time of the equipment, its initial temperature and the cold-start time noted Tmf.
• This cold time is the time of passage of the equipment from an initial temperature (usually the ambient temperature) to a very low temperature necessary for the proper operation of the equipment.
• Tmf (t) which in fact reflects the state of the equipment at time t, makes it possible to detect as soon as possible a change of state of the equipment and to propose a maintenance action. whose main objective is to avoid the breakdown.
• the problem is broken down into two stages: detection of the state of the equipment, determination of the maintenance policy.
• the 1 st step uses a hidden Markov chain.
• the process is modeled by a Markov deterministic Markov Process, or "PDMP", which takes into account the transition to the state of failure since the steady state and since then. degraded state.
• PDMP Markov deterministic Markov Process
• Optimal stopping techniques are used and adapted to these processes to maximize a performance function that takes into account time spent in operation, maintenance, repair and downtime costs.
• the previous method can be improved as described in the publication "Maintenance optimization of optronic equipment” Camille Baysse et al. Chemical Engineering Transactions, Vol. 33, 2013.
• the determination of the maintenance date is based on the principle that the later the date, the better it will be in terms of performance. According to this method, two steps are also implemented: detection of the state of the equipment as in the previous case, determination of the maintenance policy.
• the step of determining the maintenance policy consists in particular in calculating the maximum of the mathematical expectation of a performance function, starting from a state detected at a date t and the date of transition to that state itself. determined according to the time of use.
• the object of the invention is to overcome these disadvantages. As a result, there remains to date a need for a maintenance process of equipment whose main objective is of course to avoid equipment failure by providing an optimal date of maintenance (neither too early nor too late ), as well as the risk of breaking down.
• the subject of the invention is a method for maintaining equipment capable of passing through at least one degraded state before it breaks down, this equipment being provided with a sensor. of data, connected to a recorder of these captured data, itself associated with a processing unit, which comprises the following steps of:
• the so-called optimum maintenance date is determined by minimizing the expected value of a time cost function and in addition being associated with each cell of the discretized quantization grid, a probability of to move from the state of said cell to each other possible state.
• it further comprises a step for determining, as a function of the discretized quantization grid, a probability of breaking down between a predetermined date t1 and a predetermined date t2, calculated at a time t with:
• This method makes it possible to develop a decision support tool that calculates the optimal date, which can be before, during or after the mission and thereby determine the risk of using equipment on a given mission.
• FIG. 1 already described illustrates the evolution of the maintenance processes of the state of the art
• FIG. 2 diagrammatically represents an exemplary operating model of a device
• FIG. 3 diagrammatically represents an example of functions implemented to carry out the maintenance method according to the invention.
• transitions between states are represented by the arrows. And for each transition is indicated the probability of passing from the state of origin to the other state, also called transition rate or failure rate, some transition rates A 0 (t), A 3 (t) depending on cumulative use time t of the equipment, other ⁇ - ⁇ , A 2 being independent of this time of use (therefore not dependent on wear).
• the proposed solution is to combine 2 types of data: Operational collected by one or more sensors capable of collecting data specific to the equipment 100, and stored in a file 21 1.
• processing unit 220 which has three subunits.
• a first sub-unit 221 for detecting states of the equipment makes it possible from the operational data to determine the probability that the equipment is in a degraded state, by detecting a break in the behavior of a physical variable representative of the device. state of health of the equipment. It is therefore :
• Yt Tmf (t).
• a second subunit 222 proposes according to the invention an optimum and dynamic maintenance date from the results provided by the first subunit 221, that is to say from the state of the equipment, therefore from the data collected by the sensor (s) and also from a reliability model 21 2. This date is
• a PDMP is a hybrid process with two components noted ⁇ ⁇ .
• the dynamic and optimal maintenance date is determined by minimizing the expectation of a time cost function g (m t , t), which amounts to solving an optimal stopping problem. This is equivalent to minimizing on the random instant T the equation ⁇ 0 ⁇ ( ⁇ ⁇ , T)], where E is the mathematical expectation, knowing that the random time T that minimizes this expectation is this dynamic and optimal maintenance date.
• This approach which differs from that adopted in the state of the art recalled in the preamble, does not follow clearly because E (1 / F) ⁇ 1 / E (F) where E represents the expectation and F a random variable.
• an initial optimum date T 0 predetermined stored for example in the reliability model 212; it is for example the first maintenance date proposed by the manufacturer of the equipment.
• ti2 being the date of transition from state 1 to state 2 (TI3, 3, S 3, P (3, t 3))
• ti3 being the transition date from state 1 to state 3, (t 2 3, 3, other S 3 , P (3, t 2 3))
• t 2 3 being the transition date of the state 2 to state 3
• each cell of this quantization grid is associated with a discretized time grid, which requires each cell to consider the remaining duration of use. If, for example, the accumulated use time of the equipment is 25,000 hours, then for a cell whose transition date from state i to state j is noted t, the remaining duration of use is equal to at :
• This duration is then itself discretized, all of these discretized durations forming for this cell the discrete grid of usage time.
• each cell of the discretized quantization grid is associated with the probability P of passing from the state of the cell to each other possible state, by using the transition rates resulting from the predictive reliability model.
• the optimal maintenance date T falls during a mission defined by its start dates t1 and end t2 (-> T between t1 and t2)
• the calculation of the probability of failure of the mission associated with the hourly cost equipment during the mission may prompt the equipment user to decide whether to perform maintenance before or after the mission.
• the probability of a failure (also designated calculation of the risk of failure) between two times t1 and t2 predetermined by the user is then calculated at time t by performing the following steps (we often have t1 ⁇ T ⁇ t2)
• This maintenance method may in particular be implemented from a computer program product, this computer program comprising code instructions for performing the steps of the processing sub-units 1, 2 and 3. It is recorded on a computer readable medium, such as the computer 200 connected to the sensors 1 10 of the equipment 100 and in which are stored the operational information files 21 1, the reliability model 212, and maintenance costs preventive 213 and corrective 214.
• the support may be electronic, magnetic, optical, electromagnetic or be an infrared type of diffusion medium.
• Such media are, for example, Random Access Memory RAMs (ROMs), floppy disks, magnetic disks or disks, or optical (Compact Disk - Read Only Memory (CD-ROM), Compact Disk - Read / Write (CD-R / W) and DVD).

Landscapes

• Engineering & Computer Science (AREA)
• Business, Economics & Management (AREA)
• Physics & Mathematics (AREA)
• Human Resources & Organizations (AREA)
• Theoretical Computer Science (AREA)
• General Physics & Mathematics (AREA)
• Operations Research (AREA)
• Quality & Reliability (AREA)
• Strategic Management (AREA)
• Marketing (AREA)
• Entrepreneurship & Innovation (AREA)
• Economics (AREA)
• Tourism & Hospitality (AREA)
• General Business, Economics & Management (AREA)
• Evolutionary Computation (AREA)
• Mathematical Analysis (AREA)
• Mathematical Physics (AREA)
• General Engineering & Computer Science (AREA)
• Computing Systems (AREA)
• Pure & Applied Mathematics (AREA)
• Mathematical Optimization (AREA)
• Software Systems (AREA)
• Data Mining & Analysis (AREA)
• Computational Mathematics (AREA)
• Artificial Intelligence (AREA)
• Algebra (AREA)
• Probability & Statistics with Applications (AREA)
• Management, Administration, Business Operations System, And Electronic Commerce (AREA)
• Time Recorders, Dirve Recorders, Access Control (AREA)

Applications Claiming Priority (2)

Publications (1)

Family

ID=50780512

Family Applications (1)

Country Status (5)

Families Citing this family (5)

Family Cites Families (3)

• 2013
  • 2013-12-20 FR FR1303026A patent/FR3015734B1/fr active Active
• 2014
  • 2014-12-18 US US15/104,945 patent/US20160314405A1/en not_active Abandoned
  • 2014-12-18 WO PCT/EP2014/078399 patent/WO2015091752A1/fr active Application Filing
  • 2014-12-18 EP EP14814873.7A patent/EP3084692A1/de not_active Withdrawn
• 2016
  • 2016-06-19 IL IL246322A patent/IL246322A0/en unknown

Non-Patent Citations (2)

Also Published As

Similar Documents

Legal Events