[go: up one dir, main page]

WO2017163561A1 - Dispositif de support d'exploitation et système d'énergie éolienne - Google Patents

Dispositif de support d'exploitation et système d'énergie éolienne Download PDF

Info

Publication number
WO2017163561A1
WO2017163561A1 PCT/JP2017/001621 JP2017001621W WO2017163561A1 WO 2017163561 A1 WO2017163561 A1 WO 2017163561A1 JP 2017001621 W JP2017001621 W JP 2017001621W WO 2017163561 A1 WO2017163561 A1 WO 2017163561A1
Authority
WO
WIPO (PCT)
Prior art keywords
failure
data
failure risk
maintenance
probability
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.)
Ceased
Application number
PCT/JP2017/001621
Other languages
English (en)
Japanese (ja)
Inventor
憲生 竹田
寛 新谷
和夫 武藤
智彬 山下
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hitachi Ltd
Original Assignee
Hitachi Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hitachi Ltd filed Critical Hitachi Ltd
Publication of WO2017163561A1 publication Critical patent/WO2017163561A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
    • F03D—WIND MOTORS
    • F03D17/00—Monitoring or testing of wind motors, e.g. diagnostics
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
    • F03D—WIND MOTORS
    • F03D80/00—Details, components or accessories not provided for in groups F03D1/00 - F03D17/00
    • F03D80/50—Maintenance or repair
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E10/00—Energy generation through renewable energy sources
    • Y02E10/70—Wind energy
    • Y02E10/72—Wind turbines with rotation axis in wind direction

Definitions

  • the present invention particularly relates to an operation assistance device and a wind power generation system for an arbitrary product.
  • SCADA Supervision Control And Data Data Acquisition
  • CMS State Monitoring
  • SHM Structure Monitoring
  • CMS state monitoring
  • SHM structure monitoring
  • all or part of such control measurement, state monitoring, and structure monitoring are performed, and reliability is evaluated together with wind turbine control to realize stable operation of the wind turbine.
  • Patent Document 1 “a fatigue deterioration schedule in which a cumulative operating time of a windmill and an optimum fatigue deterioration degree of the windmill are associated with each other, a fatigue deterioration calculating means for calculating a current fatigue deterioration degree of the windmill, and the fatigue deterioration Driving a wind turbine comprising: an operation control unit that controls the operation of the wind turbine according to the relationship between the fatigue deterioration level of the wind speed calculated by the calculation unit and the current optimum fatigue deterioration level acquired from the fatigue deterioration schedule.
  • a control device “(Claim 1) discloses an operation control program and a wind turbine.
  • an operation control system for a wind farm having a plurality of wind turbines a remaining life prediction unit that predicts the remaining life of components for each wind turbine, and a power sale under a plurality of output restriction conditions for each wind turbine.
  • a power sales revenue prediction unit that predicts revenue
  • a maintenance cost prediction unit that predicts a maintenance cost under each output restriction condition based on the remaining life of each part for each wind turbine
  • a prediction for each output restriction condition for each wind turbine Output restriction condition selection unit for selecting an output restriction condition for each wind turbine that maximizes the profit obtained from the wind farm based on the power sale revenue and maintenance cost, and each operation command based on the selected output restriction condition.
  • a wind farm operation control system comprising an operation command section for sending to a windmill "(Claim 5) is disclosed.
  • Patent Document 3 “the maximum value obtained by multiplying the failure probability for each destruction phenomenon in the current operating environment of the equipment and each member constituting the plant by a weighting factor for each damage mode determined in advance for each member. And a plant risk value obtained by calculating a maximum value from a numerical value obtained by multiplying a failure probability for each failure phenomenon under an assumed operation condition of each member by a weighting factor for each damage mode.
  • Patent Documents 1 and 2 above the fatigue damage rate and the remaining life are used as the evaluation criteria for reliability.
  • the reliability of which part is determined. It is not disclosed whether plant operation control or maintenance plan should be implemented based on the property (fatigue damage rate and remaining life).
  • Patent Document 3 a numerical value obtained by multiplying a failure probability by a weighting factor is calculated for each of a plurality of members, and the maximum value is used as a plant risk estimated value to operate the plant. It is disclosed whether to operate the plant with attention and to plan maintenance.
  • Patent Document 3 assumes a thermal power plant such as a steam turbine, and is not considered to be used for wind turbines and construction machines that are exposed to severe outdoor environmental changes.
  • an object of the present invention is to enable a product operation and maintenance plan in consideration of highly accurate evaluation values of reliability of a plurality of parts constituting a product.
  • An auxiliary operation device A failure risk assessment department; Maintenance and operation scenario development department, With The failure risk evaluation unit Using the environmental data and operation data input from a plurality of sensors of the target product, and predetermined design data and material data, the destruction probability F (t1) of the target component p at time t1 is calculated, At time t1, the failure risk RS (t1, p) of the component p included in the product is determined for each failure probability F (t1) of the component p at the time t1 and for each predetermined component p when the component p is broken.
  • the maintenance / operation scenario formulation department The failure risk RS (t1, p) of the component p at time t1 sent from the failure risk evaluation unit, and the past from the past sent from the failure risk evaluation unit and stored in the failure risk database to time t1
  • a trend curve of failure risk is generated with a physical quantity x that affects the failure risk selected in advance from the environmental data and operation data input from the target product and time t as variables
  • Based on the failure risk trend curve obtain a predicted value of failure risk that has been advanced for a predetermined time from the present time,
  • the time t and the physical quantity x And a failure risk prediction model with the physical quantity y affecting the failure risk selected from the maintenance data and / or operation data as variables, and predicting the future failure risk of the component
  • a wind power generation system An operation assisting device as described above; A wind power generator as a target product having a plurality of sensors; A wind power generation system is provided.
  • the failure risk assessment / update unit Using the Bayes' theorem based on the probability density function of the life of the target component p breaking and the likelihood calculated from the failure data included in the failure database, the probability density function of the updated life considering the failure data is obtained. Seeking Based on the probability density function of the lifetime after the update, the environment data and operation data input from a plurality of sensors of the target product and the design data and material data determined in advance are used at the time t1 of the target component p.
  • a wind power generation system An operation assisting device as described above; A first wind power generator as a target product having a plurality of sensors; A second wind power generator having a plurality of sensors, the same type as the first wind power generator, or a similar wind power generator; A wind power generation system is provided.
  • FIG. 5 is a PSN diagram necessary for calculating a failure probability by evaluating the remaining life from the stress history of components included in the target product in the operation assistance system according to the first embodiment of the present invention. It is the figure which showed the method of calculating a damage degree from the stress frequency distribution and PSN diagram which arise in the components contained in the object product among the operation assistance systems by the 1st Embodiment of this invention.
  • the figure which showed the procedure from destruction probability calculation to maintenance / operation scenario formulation among the operation assistance systems by the 1st Embodiment of this invention The block diagram which roughly showed the main component of the operation assistance system by the 2nd Embodiment of this invention, the product which provides data to an operation assistance system, the same model machine, a similar machine, and a database.
  • the probability density function of the life is drawn with the equivalent stress amplitude, and it is Bayes
  • the probability density function of the failure life is updated from the equivalent stress amplitude calculation.
  • the probability density function of the failure life of the parts included in the product when the failure probability is calculated based on the probability density function of the failure life of the parts included in the product, the probability density function of the lifetime set in advance is based on Bayesian statistics. The figure which shows the example to update.
  • a failure life when a failure life is expressed by a multivariate probability density function including time and the failure probability is calculated, a preset probability density function is updated based on Bayesian statistics.
  • the figure which shows the example to do The block diagram which roughly showed the main component of the operation assistance system by the 3rd Embodiment of this invention, the product and database which provide the data utilized with an operation assistance system, and those relationships.
  • Block diagram The block diagram which roughly showed the main component of the operation assistance system by the 5th Embodiment of this invention, the product and database which provide the data utilized with an operation assistance system, and those relationships.
  • the main components of the maintenance / operation scenario formulation unit and the flow of data exchanged between the elements when creating a trend curve of the degree of damage are schematically shown. Block diagram.
  • the present embodiment includes a plurality of means for solving the above-mentioned problems, but if an example is given, it is an operation assisting system for an arbitrary product, It is the risk of failure of a plurality of parts constituting the product, in addition to information including at least one of environmental data, operation data, design data, material data of the product from the past to the present, the same type machine of the product, similar
  • An arithmetic unit that performs failure risk evaluation and update based on machine failure data information and obtains failure risk estimates for a plurality of the parts that fluctuate when the maintenance / operation plan of the product is changed at present.
  • means for assigning operation of the product and maintenance time for the plurality of parts by referring to each of the estimated failure risk values of the plurality of parts. According to the present embodiment, it is possible to provide a product operation assistance system capable of formulating a highly reliable product operation and maintenance plan and performing stable operation.
  • FIGS. Operation assistance device and wind power generation system
  • a wind power plant is cited as the product 1, but the application of the present invention and / or the present embodiment is not limited to the wind power plant.
  • a product operation / maintenance plan is implemented by grouping multiple parts that make up a product, taking into account highly accurate reliability evaluation values of the multiple parts that make up the product Can be possible.
  • FIG. 1 is a block diagram schematically illustrating main components of a product operation support system according to the present embodiment, a product and database that provide data used in the operation support system, and a relationship between them.
  • An operation assistance system 100 shown in FIG. 1 includes a failure risk evaluation unit 2 and a maintenance / operation scenario formulation unit 3.
  • the maintenance / operation scenario formulation unit 3 includes a failure risk prediction unit 4 that targets a plurality of parts included in a product and predicts a failure risk of the parts that changes when a maintenance / operation plan is changed.
  • the operation assistance system 100 includes an input unit, a display unit, and an output unit to another device.
  • the product 1 is equipped with a number of sensors for measuring the use environment and the operating state.
  • Environmental data and operation data measured by these sensors are sent to the failure risk evaluation unit 2 of the operation assisting system 100 and used for evaluation.
  • the environmental data is data including data related to the environment to which the product is exposed.
  • wind condition data such as wind speed and direction of a windmill is included in the environmental data.
  • sea condition data such as wavelength and wave height is also a category of environmental data.
  • the operation data is data related to the operating state of the product, such as speed, acceleration, rotation speed, and rotation angle.
  • the amount of power generated by a windmill, the rotational speed of a generator, the azimuth angle, the nacelle angle, and the like are categories of operation data.
  • environmental data and operation data are often measured as control measurement (SCADA).
  • SCADA control measurement
  • CMS state monitoring
  • SHM structure monitoring
  • design data and material data are used in the failure risk evaluation unit 2.
  • the design data includes data relating to the product shape such as a product drawing.
  • the material data includes the characteristics of the materials constituting the product and the characteristics of structures such as bolt fastening and welded joints.
  • FIG. 2 is a block diagram schematically showing processing executed by the failure risk evaluation unit 2.
  • the failure risk evaluation unit 2 calculates a failure risk using at least one of environmental data, operation data, design data, and material data.
  • the failure risk RS (t1, p) of a certain part p included in the product is the destruction probability F (t1) of the target part p at the time t1 and the degree of influence C ( From p), it can be calculated by the following equation. Note that the data of the degree of influence C (p) for each part p is stored in the influence degree database 8 in advance.
  • the failure probability F (t1) can be calculated by remaining life evaluation or predictive diagnosis. First, the remaining life evaluation will be described. In order to obtain the failure probability F (t1) in the remaining life evaluation for the fatigue phenomenon caused by the repeated load acting on the product, first, from the past to the current t1 using the environmental data, operation data and design data. Calculate the history of stress that has occurred in the parts until now. Next, a frequency analysis method such as a rain flow method is applied to the stress history to create a stress frequency distribution that organizes how often a certain amount of stress is generated. Then, the fracture probability F (t1) is obtained using the stress frequency distribution and the material data related to the target part.
  • a frequency analysis method such as a rain flow method
  • FIG. 3 is a PSN diagram necessary for calculating the failure probability by the remaining life evaluation from the stress history of the parts included in the target product.
  • the material data used here is preferably the fatigue life curve of FIG. 3 called a PSN diagram, and is stored in the design / material database 5 in this embodiment.
  • the number of repetitions having a fracture probability P% is obtained from the probability density function 20 of fatigue life obtained by carrying out a fatigue test at each stress amplitude on the vertical axis, and these are connected and illustrated. (FIG. 3).
  • FIG. 4 is a diagram showing a method of calculating the damage degree from the stress frequency distribution generated in the parts included in the target product and the PSN diagram.
  • the fracture probability F (t1) is obtained by the following procedure.
  • n 1 , n 2 , and n m are respectively the number of repetitions of stress amplitudes S 1 , S 2 , and S m obtained by frequency analysis of stress history (m is an integer).
  • N 1 , N 2 , and N m are the number of repetitions of fracture at which fatigue failure occurs with a failure probability P% when stress amplitudes S 1 , S 2 , and S m are repeatedly applied.
  • the number of repetitions N (t1) for generating the damage degree D (t1) can be obtained by the following equation.
  • N (t1) D (t1) ⁇ Np (3)
  • Np is the advance number of repetitions of the failure probability P% defined, an input unit (not shown.)
  • S i e.g., average value, intermediate value, close to their values, etc.
  • the fracture probability F (t1) can be obtained from N (t1) and the probability density function 20 of fatigue life by the equation of FIG. That is, if the elapsed time from the start of operation of the product to the present is t1, the fracture probability F (t1) is a value obtained by integrating the density function f (N) from 0 to N (t1).
  • FIG. 5 is a diagram showing another method for calculating the fracture probability using the probability density function of the fracture life.
  • the failure probability F (t1) until the failure can be obtained by directly defining the probability density function f (t1) of the lifetime as shown in FIG. That is, assuming that the elapsed time from the start of operation of the product to the present time is t1, the fracture probability F (t1) is a value obtained by integrating the density function f (t) from 0 to t1.
  • the life T the difference between the current time t1 and the life time T can be considered as the remaining life, so there is a method for directly defining the probability density function of the life in this way. This can be interpreted as a remaining life evaluation.
  • the life probability density function f (t1) data can be stored in advance in an appropriate memory such as the internal memory of the failure risk evaluation unit 2 for each target product. Further, the data of the probability density function f (t1) of the lifetime can be determined in advance using environmental data, operation data, and the like.
  • FIG. 6 is a diagram illustrating a method of selecting a physical quantity related to destruction from measurement data of a product and appropriately converting it to obtain a probability density function of a lifetime.
  • the operation data when the target part is normal and abnormal are arranged in advance, and the current operation data is monitored to determine the failure of the target part.
  • a failure probability F (t1) at the current time t1 can be obtained by capturing the abnormal operation data group as a probability density function of the failure as shown in FIG. 6 and plotting the position of the current operation data in FIG. .
  • a physical quantity related to the destruction of the target part is selected in advance from the operation data at the time of abnormality and a probability density function using the selected physical quantity as a random variable is created, or related to the destruction of the target part as shown in FIG.
  • a physical quantity included in the operation data is converted into a format to be generated, and a probability density function is created using the converted physical quantities x1 ′ and x2 ′ as random variables.
  • the converted physical quantity includes, for example, a spectrum value of a specific frequency among acceleration spectra obtained by performing fast Fourier transform on acceleration data included in driving data.
  • the probability density function may store in advance an appropriate memory such as the internal memory of the failure risk evaluation unit 2 or the design / material database 5 for each target product.
  • the probability density function can be determined in advance using environmental data, operation data, or the like.
  • the influence degree database 8 stores data of the influence degree C (p) for each part p included in the equation (1). If costs such as all or a predetermined range required when an actual part p is broken are used as the degree of influence C (p), the failure risk RS (t1, p) is caused by a failure at the current time t1. It can be considered as an expected value of loss cost such as all or a predetermined range.
  • the cost required when a part is broken includes the cost of the new part itself, the replacement cost of the part, the transportation cost of the part, and the loss cost of the power generation opportunity due to the shutdown of the product.
  • the failure risk RS (t1, p) can be relatively compared between the components, and it can be determined which component should be considered for reliability.
  • the failure risk RS (t1, p) calculated in this way is input to the maintenance / operation scenario formulation unit 3 together with the maintenance / operation data, environmental data, operation data, design data, and material data.
  • FIG. 7 is a block diagram schematically showing processing executed by the maintenance / operation scenario formulation unit.
  • the maintenance / operation scenario formulation unit 3 includes a failure risk prediction unit 4.
  • the failure risk prediction unit 4 includes a failure risk RS (t 1, p) at time t 1 sent from the failure risk evaluation unit 2 and a failure.
  • a plurality of failures at each time tn (n 0, ⁇ 1, ⁇ 2, ⁇ 3,%) From the past to time t1 that have already been sent from the risk evaluation unit 2 and stored in advance in the failure risk database 30
  • the fluctuation trend of the failure risk is analyzed from the risk RS (tn, p), the environmental data, and the operation data.
  • FIG. 8 is a diagram showing an example of a failure risk trend curve created by failure risk trend analysis of the failure risk prediction unit.
  • a failure risk trend curve created by failure risk trend analysis of the failure risk prediction unit.
  • wind tends to be strong in winter, so that depending on the part, a trend curve can be drawn with time on the horizontal axis and failure risk on the vertical axis as shown in FIG.
  • a regular trend curve as shown in FIG. 8 may not be obtained depending on the PSN diagram and probability density function distribution shape used for the calculation of failure probability. Not a few.
  • the fluctuation of the failure risk RS (t1, p) due to the short-term wind turbulence may be greater than or equal to the seasonal fluctuation
  • the failure risk prediction unit 4 selects the following equations (4), (5), etc. according to the product and the parts constituting the product, where t is the time and x is the selected physical quantity.
  • the curve RS is determined.
  • Equations (4) and (5) are the trend curves of failure risk when the time t and the physical quantity x are variables, but the autoregressive moving average model including the physical quantity and error term from the past to the present time, It is also possible to adopt a neural network learned by inputting environmental data and driving data as a trend curve.
  • RS (t, x, p) ⁇ (p) ⁇ t + ⁇ (p) ⁇ x + ⁇ (p) ( ⁇ , ⁇ , and ⁇ are constants, but depend on parts) However, it is not limited to this.
  • the failure risk prediction unit 4 executes the failure risk prediction 11 based on the failure risk trend curve, maintenance / operation data, and the maintenance / operation scenario defined in the maintenance / operation scenario formulation 12.
  • the maintenance / operation data is past data, and includes, for example, information on periodic inspection of products, control change information for making operations different, and information on inspection execution due to failure.
  • the maintenance / operation scenario refers to, for example, a plan such as how to replace which part in what year and month, how to operate, and the like.
  • the maintenance / operation scenario formulation 12 may automatically create a maintenance / operation scenario by a predetermined method, or may manually input a maintenance / operation scenario from an input unit.
  • FIG. 9 is a diagram illustrating an example of risk prediction processing when the maintenance / operation scenario executed by the maintenance / operation scenario formulation unit is different.
  • the periodic inspection and operation method (product control method, etc.) currently employed in the future are continued, the future failure risk (predicted risk) a advanced by a predetermined time ⁇ T from the present time is shown in FIG.
  • forecast risk a it will be along the trend curve so far.
  • the maintenance / operation scenario formulation unit 3 performs the maintenance / operation scenario formulation 12 using the predicted risk. That is, based on the predicted risk value, the maintenance of the product and the change of the operation method are examined manually by the input from the input unit by the maintenance operator or automatically by a predetermined process. For example, when the predicted risk is higher than a predetermined threshold, a more gentle change to the product operation method is proposed in the maintenance / operation scenario formulation 12, and when the predicted risk is lower than the predetermined threshold, it is more severe Product operation methods and frequent maintenance inspections are proposed.
  • the proposed maintenance / operation scenario is sent again to the failure risk prediction 11, and the future risk b or c is predicted as shown in FIG. 9 according to the maintenance / operation scenario.
  • failure risk from maintenance / operation data in addition to physical quantity x (selected from environmental data and operation data) that affects failure risk at time t The physical quantity y that affects the physical quantity y is selected, and the failure risk prediction model g3 or g4 of the following equation using the physical quantities x and y as variables can be used.
  • RS (t + ⁇ T, y, p) g3 (t + ⁇ T, y, p) (6)
  • RS (t + ⁇ T, x, y, p) g4 (t + ⁇ T, x, y, p) ...
  • Equations (6) and (7) are trend curves in which the time t + ⁇ T and the future physical quantities x and y are variables, and the autoregressive moving average model including the physical quantities and error terms from the past going back to a certain time to the time t + ⁇ T
  • a neural network learned by inputting time, environmental data, operation data, and maintenance / operation data can also be used as a prediction model.
  • FIG. 10 is a diagram illustrating an example in which main components of the product are interpreted as parts, arranged and displayed in descending order of predicted value of failure risk, and grouped.
  • the maintenance / operation scenario formulation 12 shown in FIG. 7 is performed in consideration of grouping. For example, a plan may be established such as which group will be maintained after how many years.
  • the failure risk prediction in FIG. 9 can be performed on all parts constituting the product. However, by predicting failure risk only for parts with high failure risk (or prediction risk or future risk), the man-hours required for prediction can be reduced, and maintenance and operation can be efficiently planned. For example, as shown in FIG.
  • the maintenance / operation scenario formulation unit 3 has a failure risk (or a prediction risk or a future risk) together with a component having the maximum prediction value and the maximum value as shown in FIG. Are arranged in descending order of the predicted values and displayed on the display unit.
  • the group and the maintenance time can be assigned according to the upper and lower thresholds for each group so that the failure risk of all parts does not exceed a certain threshold. For example, such an assignment is possible by a genetic algorithm.
  • the maintenance / operation scenario formulation unit 12 of the maintenance / operation scenario formulation unit 3 refers to FIG. 10, which is manually input by the maintenance operator or the like from the input unit or automatically by a predetermined process for each group. Plan maintenance operations such as when to maintain parts.
  • a maintenance / operation scenario such as maintenance operation contents may be manually input from the input unit.
  • the planned maintenance operation content is stored in an appropriate storage unit and / or displayed on the display unit. If the predicted value of the failure risk (predicted risk or future risk, etc.) is arranged according to the magnitude, for example, as shown in FIG. 10, group A, group B, group C are determined according to a plurality of predetermined thresholds.
  • maintenance can be planned by grouping parts into groups. According to such a maintenance plan based on grouping, for example, replacement of parts belonging to group A is performed at the maintenance time that was most recently planned. It is possible to replace parts with high prediction risk or future risk).
  • Such grouping for assigning maintenance times can be performed using a combinational optimization method such as a genetic algorithm or a branch and bound method.
  • FIG. 11 is a diagram in which predicted values of failure risk are calculated for each part belonging to main components of the product, and the parts are arranged and displayed in descending order of the predicted value.
  • FIG. 10 described above interprets the components included in the product as parts, and arranges the relationship between the parts and the predicted value of the failure risk. For example, a file that defines which component is included in which component is stored in advance in an appropriate storage unit, and the maintenance / operation scenario formulation unit 3 displays the component, component, and failure risk by referring to the file. Can be displayed.
  • the predicted value (predicted risk or future risk) of the failure risk for the component it is possible to clarify the component that affects the product.
  • FIG. 10 described above interprets the components included in the product as parts, and arranges the relationship between the parts and the predicted value of the failure risk.
  • a file that defines which component is included in which component is stored in advance in an appropriate storage unit and the maintenance / operation scenario formulation unit 3 displays the component, component, and failure risk by referring to the file. Can be
  • the predicted value of the failure risk can be arranged for each part included in the component.
  • the operation assistance system of this embodiment is provided with the apparatus which arranges and displays the predicted value of a failure risk like FIG.
  • the failure risk evaluation in the failure risk evaluation unit 2 of the operation assistance system 100, the failure risk prediction in the maintenance / operation scenario formulation unit 3, and the maintenance / operation scenario formulation are executed at certain time intervals. This time interval may be the same as or different from the time interval at which the environmental data and operation data are measured.
  • SCADA control measurement
  • environmental data and operational data statistical values are calculated at 10-minute intervals, for example, and the statistical values are PCs or the like. Is stored in the server configured with.
  • the operation assistance system can include components included in the wind power plant.
  • the wind turbine can be operated stably while sufficiently predicting the failure of the wind turbine.
  • FIG. 12 shows a flowchart of the operation assistance system according to the first embodiment.
  • the failure risk evaluation unit 2 calculates the failure probability F (t1) and the failure risk RS (t1, p) at the time t1 of the target component p (S11, S12).
  • the failure risk prediction unit 4 calculates a failure risk trend curve RS (t, x, p) and a future failure risk RS (t + ⁇ T, x, y, p) (S13, S14). Whether maintenance / operation is to be changed automatically based on the calculated future failure risk, for example by comparing with a predetermined threshold, or by manual input from the maintenance / operator etc. Judging.
  • the physical quantity / condition to be changed to the maintenance / operation scenario formulation 12 automatically by a predetermined process or manually by an input from an input unit by a maintenance / operator or the like.
  • the maintenance / operation scenario formulation unit 3 formulates a maintenance / operation scenario according to the setting / input (S15, S16, S14). As described above, such a procedure is repeated at intervals of 10 minutes, for example, to assist the stable operation of the product.
  • the failure risk evaluation / update unit 14 executes failure risk evaluation and update using the failure data information.
  • Failure data of a plurality of parts constituting the product 1 and its similar machines and similar machines 13 is stored in the failure database 15.
  • the failure data includes, for example, the operation time from the operation start to the failure, the environmental data from the operation start to the time of the failure, the operation data, and the maintenance / operation data.
  • the same type machine includes products of the same type as that of the product 1, and the similar machine includes a product of a different type from the product 1.
  • FIG. 14 is a diagram showing an example in which the probability density function of the life is drawn with the equivalent stress amplitude and updated based on Bayesian statistics when the failure probability is calculated by the remaining life evaluation from the stress history of the parts included in the product. is there.
  • the failure risk evaluation / update unit 14 targets fatigue phenomena that occur due to repeated loads on the product and obtains the failure probability F (t1) in the remaining life evaluation, the stress generated in the part from the past to the current t1
  • a stress frequency distribution is obtained by applying a frequency analysis method such as a rainflow method to the history (see FIG. 4).
  • the stress frequency distribution 21 is used to express the stress, but the equivalent stress amplitude when only a certain magnitude of stress is assumed to be generated.
  • Seq can be calculated from the stress frequency distribution using, for example, the formula in FIG. If the equivalent stress amplitude Seq is obtained, the probability density function 16 of the lifetime at which the part breaks at the equivalent stress amplitude Seq can be drawn.
  • FIG. 15 shows a flowchart for updating the density function of such a fracture life.
  • the failure risk evaluation / update unit 14 uses the equation shown in FIG. 14 for the equivalent stress amplitude Seq (p) from the stress frequency distribution obtained by frequency analysis of the stress history from the past to the current time t1 of the target component p. To calculate (S21). Then, the failure risk evaluation / update unit 14 refers to the material data (PSN diagram) of the target part p stored in advance in the design / material database 5 or the like, and uses the equivalent stress amplitude Seq (p). A probability density function f (N) of the fracture life is obtained (S22).
  • the failure risk evaluation / update unit 14 obtains the stress history and stress frequency distribution until failure from the environmental data, operation data, maintenance / operation data and design / material data from the start of operation to the failure of these parts,
  • the fracture life N f j at the equivalent stress amplitude Seq (p j ) is calculated according to the expression on the right side in FIG. 15 (S23).
  • the failure risk evaluation / updating unit 14 can calculate the likelihood L from the k pieces of the fracture life N f j thus obtained according to the expression on the left side in FIG.
  • the failure risk evaluating / updating unit 14 can obtain the updated probability life function f (N) ′ of the fracture life from the probability density function f (N) and the likelihood L in advance according to the equation (8). Yes (S25).
  • the probability density function in advance and the probability distribution shape of likelihood are Weibull distribution or lognormal distribution.
  • the degree of damage D (t1) with respect to fatigue failure P% can be calculated by the following equation.
  • D (t1) (n 1 / N 1 ) + (n 2 / N 2 ) +... + (N m / N m ) ...
  • n 1 , n 2 , and n m are respectively the number of repetitions of stress amplitudes S 1 , S 2 , and S m obtained by frequency analysis of stress history (m is an integer).
  • N 1 , N 2 , and N m are the number of repetitions of fracture at which fatigue failure occurs with a failure probability P% when stress amplitudes S 1 , S 2 , and S m are repeatedly applied.
  • the number of repetitions N (t1) for generating the damage degree D (t1) is obtained by the following equation.
  • N (t1) D (t1) ⁇ Np (3)
  • Np is the number of repetitions with a predetermined failure probability P%, and by setting the stress amplitude to a predetermined stress amplitude S i or Seq (p), as shown in FIG. 4, N (t1) From the probability density function of the fatigue life, the fracture probability F (t1) ′ can be obtained by the equation of FIG.
  • the fracture probability F (t1) ′ is a value obtained by integrating the density function f (N) ′ from 0 to N (t1).
  • the updated probability is updated by using the updated PSN diagram, and the updated probability according to the formula (1) is obtained by multiplying the updated probability F (t1) ′ and the influence degree C (p).
  • the failure risk RS (t1, p) ′ can be calculated.
  • Such a method of updating the failure risk from the stress frequency distribution, PSN diagram, and failure data is an example.
  • the updated failure probability may be calculated without using the equivalent stress amplitude.
  • the update function of the density function of the fracture life based on the Bayes' theorem may be changed according to the use environment of the same type machine as the target product 1, the operating environment of the similar machine 13, the operating situation, the structural similarity, and the like.
  • FIG. 16 is a diagram illustrating an example of updating a preset probability density function of a lifetime based on Bayesian statistics when calculating a fracture probability based on a probability density function of a fracture life of a part included in a product. .
  • the representative Bayes theorem can be used. That is, the lifetime density function and the failure data are substituted into the equation (8) to update the lifetime density function, and the probability probability function after the update is used to determine the fracture probability F (t1) ′ (FIG. 16), which has an effect on it.
  • the failure risk can be updated by multiplying the degree.
  • the breakdown life of the parts of the product 1 and the parts of the same type machine and similar machine 13 will be greatly different even if the operating time is the same. It is done.
  • a physical quantity that affects the fracture life is selected from environmental data and operation data as variables, and a multi-variable life probability density function is created.
  • the data of the probability density function f (t1) of the lifetime can be stored in advance in an appropriate memory such as the internal memory of the failure risk evaluation unit 2 or the design / material database 5 for each target product. Further, the data of the probability density function f (t1) of the lifetime can be determined in advance using environmental data, operation data, and the like.
  • FIG. 17 is a diagram illustrating an example in which the failure life is expressed by a multivariate probability density function including time and the probability density function set in advance is updated based on Bayesian statistics when the failure probability is calculated.
  • FIG. 17 shows an example of a probability density function of a bivariate lifetime.
  • a value x ′ obtained by converting a physical quantity x selected from environmental data or operation data is used as a random variable with time. This conversion includes conversion of a physical quantity that changes over time into a statistical quantity (average value, maximum value, etc. at a certain time interval), conversion into an equivalent physical quantity based on frequency analysis, and the like.
  • the life density function can be updated according to Bayes' theorem as in the case of univariate, as shown in FIG.
  • a physical quantity that greatly affects the fracture life or a function r (t, x ') including a quantity x ′ and a time t converted from the physical quantity is generated, and the probability density of the fracture life is interpreted by interpreting the function as a random variable.
  • the failure risk evaluation / update unit 14 calculates the failure probability by predictive diagnosis
  • the failure probability density function may be updated using the Bayes' theorem represented by the equation (8).
  • a failure probability is obtained from the updated probability density, and the failure risk may be updated by multiplying it by the degree of influence.
  • the probability density function of the bivariate lifetime can store, for example, an appropriate memory such as the internal memory of the failure risk evaluation unit 2 or the design / material database 5 for each target product.
  • the probability density function of the bivariate lifetime can be determined in advance using environmental data, operation data, or the like.
  • failure risk evaluation / update unit 14 and the maintenance / operation scenario formulation unit 3 use the updated failure risk RS (t1, p), as in the first embodiment shown in FIG.
  • Each process of trend analysis (S13), future risk prediction (S14), maintenance / operation change (S15), and maintenance / operation scenario formulation (S16) is executed.
  • the calculation of the failure probability can be made highly accurate, so that the failure risk evaluation and prediction of the parts included in the product 1 Can be carried out with higher accuracy.
  • the time interval of failure risk update may be the same as or different from the failure risk evaluation interval.
  • FIG. 18 is a block diagram schematically showing main components of the operation assistance system according to the third embodiment of the present invention, products and databases that provide data used in the operation assistance system, and their relationship.
  • the product 1 Information from the external database 17 that does not depend on is used.
  • the external data included in the external database 17 includes, for example, weather calculated by a large computer, sea state future prediction data, resource supply prediction data, resource reserve prediction data, and the like. Such external data is not affected by the operating state of the product 1, and therefore the external data does not depend on the product 1.
  • FIG. 19 is a block diagram schematically showing the main components of the maintenance / operation scenario formulation unit of the operation assistance system according to the third embodiment of the present invention and the flow of data exchanged between the components.
  • the failure risk prediction 11 of the maintenance / operation scenario formulation unit 3 calculates a future failure risk using the failure risk trend curve, the maintenance / operation scenario, and external data. That is, the predicted value of failure risk can be calculated from the physical quantity x included in the environmental data or the operation data, the physical quantity y included in the maintenance / operation data, and the physical quantity z included in the external data by the following equation.
  • FIG. 20 is a block diagram schematically showing main components of an operation assistance system according to the fourth embodiment of the present invention, products and databases that provide data used in the operation assistance system, and their relationships.
  • FIG. 21 is a block diagram schematically showing the main components of the maintenance / operation scenario formulation unit of the operation assistance system according to the fourth embodiment of the present invention and the flow of data exchanged between the components.
  • the environmental data and operation data measured by the product 1 are input to the failure probability evaluation unit 18, where the failure probability F of the parts included in the product 1 is calculated.
  • the calculated failure probability F becomes an input to the maintenance / operation scenario formulation unit 3, where a failure risk trend curve is created by multiplying the failure probability and the influence degree.
  • FIG. 22 is a block diagram schematically showing the main components of the maintenance / operation scenario formulation unit and the flow of data exchanged between the elements when creating a trend curve of the failure probability. Therefore, in the present embodiment, as shown in FIG. 21, the impact database 8 is arranged in the maintenance / operation scenario formulation unit 3. On the other hand, as shown in FIG. 22, a form of creating a trend curve of failure probability instead of a trend curve of failure risk is also conceivable. In such a case, a trend curve of the failure probability F as shown in the following equation can be created for the part p from the time t and the physical quantity x that affects the failure probability or the failure probability.
  • F (t + ⁇ T, y, p) h3 (t + ⁇ T, y, p) (13)
  • F (t + ⁇ T, x, y, p) h4 (t + ⁇ T, x, y, p) (14)
  • y is a physical quantity that affects the failure risk RS in the maintenance / operation data.
  • the future failure risk RS can be determined by multiplying this future failure probability or failure probability by the degree of influence C (p) as shown in the following equation.
  • FIG. 23 is a block diagram schematically showing main components of an operation assistance system according to the fifth embodiment of the present invention, products and databases that provide data used in the operation assistance system, and their relationships.
  • the operation assistance system by 5th Embodiment of this embodiment is demonstrated using FIG.
  • environmental data and operation data measured by the product 1 are input to the damage degree evaluation unit 19 where the damage degree of the parts included in the product 1 is calculated.
  • the degree of damage can be obtained by selecting a fatigue life curve with a certain probability of failure in the PSN diagram of FIG. it can.
  • the calculated damage level is input to the maintenance / operation scenario formulation unit 3, where the failure probability is calculated by referring to the material data, and a failure risk trend curve is created by multiplying the damage probability. Therefore, in the present embodiment, as shown in FIG. 21, the impact database 8 is arranged in the maintenance / operation scenario formulation unit 3. On the other hand, as shown in FIG. 22, not only the failure risk trend curve but also a failure probability trend curve may be considered.
  • FIG. 24 is a block diagram schematically showing the main components of the maintenance / operation scenario formulation unit when creating a trend curve of the degree of damage and the flow of data exchanged between the elements.
  • a form of creating a trend curve of the degree of damage is also conceivable. That is, the failure risk trend curve can also be determined as shown in the following equation by creating the future damage levels d3 and d4 and multiplying by the influence level C (p).
  • K (p) is a conversion constant necessary for calculating the fracture probability from the degree of damage.
  • K (p) may be stored in a damage degree database, an influence degree database, or another database.
  • the failure probability and the damage degree are independently calculated once, there is an advantage that the failure probability and the damage degree can be easily visualized (displayed by the display unit).
  • the failure risk calculated according to the equation (1) is a failure risk that considers all damage accumulated in the product during the period from the start of operation of the product to time t1.
  • a failure risk considering only the damage accumulated in the product and use it as a predicted value of the failure risk.
  • a probability P that a certain part included in the product is broken is calculated by the following equation during a period from the current time t1 to a certain future time point t1 + ⁇ T.
  • F (t1, t1 + ⁇ T) (F (t1 + ⁇ T) ⁇ F (t1)) / (1 ⁇ F (t1)) ... (21)
  • F (t1) and F (t1 + ⁇ T) are destruction probabilities at times t1 and t1 + ⁇ T, respectively.
  • the failure risk during the period from the current time t1 to a future time point t1 + ⁇ T can be calculated from the influence degree C (p) and the failure probability P (t1, t1 + ⁇ T) of the target component by the following equation.
  • Each of the above-described configurations, functions, and the like may be realized by software by interpreting and executing a program that realizes each function by the processor.
  • Information such as programs, tables, and files for realizing each function can be stored in a memory, a hard disk, a recording device such as an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, or a DVD.
  • the control lines and information lines indicate what is considered necessary for the explanation, and not all the control lines and information lines on the product are necessarily shown. Actually, it may be considered that almost all the components are connected to each other.

Landscapes

  • Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Sustainable Development (AREA)
  • Sustainable Energy (AREA)
  • Chemical & Material Sciences (AREA)
  • Combustion & Propulsion (AREA)
  • Mechanical Engineering (AREA)
  • General Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Tourism & Hospitality (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Wind Motors (AREA)
  • Testing And Monitoring For Control Systems (AREA)

Abstract

La présente invention concerne un système de support d'exploitation qui est capable de se référer à des valeurs d'évaluation de haute précision de fiabilité d'une pluralité de composants qui configurent un produit, et de mettre en œuvre un plan de gestion et de maintenance du produit. L'invention porte sur un dispositif de support d'exploitation 100 d'un produit arbitraire, comprenant une unité d'évaluation de risque de défaut 2 qui : dérive un risque de défaut d'une pluralité de composants qui configurent le produit, ledit risque de défaut étant calculé sur la base d'informations qui comprennent au moins l'un parmi des données environnementales, des données de fonctionnement, des données de conception et/ou des données de matériaux, du produit, du passé au présent ; et dérive des valeurs d'estimation de risque de défaut de la pluralité de composants qui fluctuent si un plan de gestion et de maintenance actuel du produit est modifié. Le dispositif de support d'exploitation 100 comprend en outre une unité de création de scénario de maintenance/gestion 3 qui sert à se référer à chacune des valeurs d'estimation de risque de défaut calculées de la pluralité de composants, et à affecter la gestion du produit et les temps de maintenance de la pluralité de composants.
PCT/JP2017/001621 2016-03-25 2017-01-19 Dispositif de support d'exploitation et système d'énergie éolienne Ceased WO2017163561A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2016-061725 2016-03-25
JP2016061725A JP2019113883A (ja) 2016-03-25 2016-03-25 稼働補助装置及び風力発電システム

Publications (1)

Publication Number Publication Date
WO2017163561A1 true WO2017163561A1 (fr) 2017-09-28

Family

ID=59899924

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2017/001621 Ceased WO2017163561A1 (fr) 2016-03-25 2017-01-19 Dispositif de support d'exploitation et système d'énergie éolienne

Country Status (3)

Country Link
JP (1) JP2019113883A (fr)
TW (1) TWI607328B (fr)
WO (1) WO2017163561A1 (fr)

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6526362B1 (ja) * 2018-03-13 2019-06-05 三菱電機株式会社 故障度算出装置及び故障度算出システム
CN109858689A (zh) * 2019-01-21 2019-06-07 北京航空航天大学 一种可靠性导向的产品装配系统健康风险分析方法
CN110084465A (zh) * 2019-03-11 2019-08-02 杭州电子科技大学 基于储能的风力发电系统成本/供电可靠性评估方法
JP2019160128A (ja) * 2018-03-16 2019-09-19 株式会社日立製作所 故障確率評価システム及び方法
JP2020042515A (ja) * 2018-09-10 2020-03-19 日本電気株式会社 モデル生成装置、モデル生成方法、及びプログラム
WO2020129898A1 (fr) * 2018-12-21 2020-06-25 日立建機株式会社 Système de sélection de composant de maintenance d'une machine de travail
WO2021005943A1 (fr) * 2019-07-05 2021-01-14 株式会社日立製作所 Système d'évaluation de probabilité de défaillance et procédé correspondant
CN114183314A (zh) * 2021-11-25 2022-03-15 吉林省电力科学研究院有限公司 一种基于可靠度的风电机组机会维护方法
CN115045803A (zh) * 2021-03-08 2022-09-13 北京金风科创风电设备有限公司 风力发电机组的故障触发调节方法和装置
CN115130702A (zh) * 2022-09-02 2022-09-30 山东汇泓纺织科技有限公司 一种基于大数据分析的纺织机故障预测系统
CN115860212A (zh) * 2022-11-29 2023-03-28 国网福建省电力有限公司经济技术研究院 一种配电网的风险预测方法与终端
JP2023069397A (ja) * 2021-11-05 2023-05-18 株式会社東芝 設備点検計画作成システム、および、設備点検計画作成方法
US11746756B2 (en) 2018-08-07 2023-09-05 Vestas Wind Systems A/S Operating a wind turbine during service
EP4198487A4 (fr) * 2020-08-17 2024-08-28 Hitachi, Ltd. Dispositif d'évaluation de progression de fissure et programme d'évaluation de progression de fissure

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6917288B2 (ja) * 2017-12-12 2021-08-11 株式会社日立製作所 保守計画生成システム
JP7364382B2 (ja) * 2019-07-19 2023-10-18 三菱重工業株式会社 運転比較装置、運転比較方法およびプログラム
JP7475978B2 (ja) * 2020-06-12 2024-04-30 株式会社日立製作所 保全改善支援装置及び保全改善支援方法
JP7490470B2 (ja) * 2020-06-29 2024-05-27 日揮株式会社 保全管理システム
JP7759394B2 (ja) * 2021-08-26 2025-10-23 株式会社カネカ 監視システム
JP2023116047A (ja) * 2022-02-09 2023-08-22 コベルコ建機株式会社 検出システム、検出方法、および、検出プログラム
JP2024006371A (ja) * 2022-07-01 2024-01-17 三菱電機株式会社 設備交換計画策定装置、設備交換計画策定システム、設備交換計画策定方法および設備交換計画策定プログラム
JP2024035858A (ja) * 2022-09-05 2024-03-15 株式会社日立製作所 リスクデータ管理装置及びリスクデータ管理方法
JP7450830B1 (ja) * 2023-03-17 2024-03-15 三菱電機株式会社 プログラム、部品選定支援装置、部品選定支援システムおよび部品選定支援方法
KR102655555B1 (ko) * 2023-11-20 2024-04-11 한국전력기술 주식회사 심층 강화학습을 활용한 해상풍력단지 정비 플래닝 시스템 및 방법

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005221265A (ja) * 2004-02-03 2005-08-18 Toshiba Corp 原子力発電プラント機器検査方法および原子力発電プラント機器検査システム
JP2009217718A (ja) * 2008-03-12 2009-09-24 Toshiba Corp 保守計画支援システム
US20120143565A1 (en) * 2011-09-20 2012-06-07 Graham Iii Fred Gorum System and method for predicting wind turbine component failures

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102637289B (zh) * 2012-04-16 2016-02-24 天津大学 含大规模风电的电力系统规划方案的安全性价值评估方法
US10861090B2 (en) * 2013-11-27 2020-12-08 Apple Inc. Provisioning of credentials on an electronic device using passwords communicated over verified channels
CN103914791A (zh) * 2014-03-17 2014-07-09 中国南方电网有限责任公司超高压输电公司检修试验中心 一种电力设备状态检修系统

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005221265A (ja) * 2004-02-03 2005-08-18 Toshiba Corp 原子力発電プラント機器検査方法および原子力発電プラント機器検査システム
JP2009217718A (ja) * 2008-03-12 2009-09-24 Toshiba Corp 保守計画支援システム
US20120143565A1 (en) * 2011-09-20 2012-06-07 Graham Iii Fred Gorum System and method for predicting wind turbine component failures

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
YASUHIRO TOYAMA: "Failure Risk Inference for Equipment Maintenance", PROCEEDINGS OF THE 2016 IEICE GENERAL CONFERENCE JOHO SYSTEM 1, 1 March 2016 (2016-03-01), pages 28 *

Cited By (27)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6526362B1 (ja) * 2018-03-13 2019-06-05 三菱電機株式会社 故障度算出装置及び故障度算出システム
WO2019175984A1 (fr) * 2018-03-13 2019-09-19 三菱電機株式会社 Dispositif de calcul de niveau de défaillance et système de calcul de niveau de défaillance
US12406190B2 (en) 2018-03-16 2025-09-02 Hitachi, Ltd. Failure probability evaluation system and method
JP2019160128A (ja) * 2018-03-16 2019-09-19 株式会社日立製作所 故障確率評価システム及び方法
WO2019176315A1 (fr) * 2018-03-16 2019-09-19 株式会社日立製作所 Système et procédé d'évaluation de probabilité de dysfonctionnement
US11746756B2 (en) 2018-08-07 2023-09-05 Vestas Wind Systems A/S Operating a wind turbine during service
JP2020042515A (ja) * 2018-09-10 2020-03-19 日本電気株式会社 モデル生成装置、モデル生成方法、及びプログラム
US11494247B2 (en) 2018-09-10 2022-11-08 Nec Corporation Model generation apparatus, model generation method, and non-transitory storage medium
WO2020129898A1 (fr) * 2018-12-21 2020-06-25 日立建機株式会社 Système de sélection de composant de maintenance d'une machine de travail
JP2020102009A (ja) * 2018-12-21 2020-07-02 日立建機株式会社 作業機械の保守部品の選定システム
JP7248877B2 (ja) 2018-12-21 2023-03-30 日立建機株式会社 作業機械の保守部品の選定システム
CN109858689A (zh) * 2019-01-21 2019-06-07 北京航空航天大学 一种可靠性导向的产品装配系统健康风险分析方法
CN110084465B (zh) * 2019-03-11 2022-03-18 杭州电子科技大学 基于储能的风力发电系统成本/供电可靠性评估方法
CN110084465A (zh) * 2019-03-11 2019-08-02 杭州电子科技大学 基于储能的风力发电系统成本/供电可靠性评估方法
AU2020309879B2 (en) * 2019-07-05 2023-11-02 Hitachi, Ltd. Failure probability assessment system and method therefor
JP7145821B2 (ja) 2019-07-05 2022-10-03 株式会社日立製作所 故障確率評価システムおよびその方法
JP2021012504A (ja) * 2019-07-05 2021-02-04 株式会社日立製作所 故障確率評価システムおよびその方法
WO2021005943A1 (fr) * 2019-07-05 2021-01-14 株式会社日立製作所 Système d'évaluation de probabilité de défaillance et procédé correspondant
US11983002B2 (en) 2019-07-05 2024-05-14 Hitachi, Ltd. Failure probability assessment system and method therefor
EP4198487A4 (fr) * 2020-08-17 2024-08-28 Hitachi, Ltd. Dispositif d'évaluation de progression de fissure et programme d'évaluation de progression de fissure
CN115045803A (zh) * 2021-03-08 2022-09-13 北京金风科创风电设备有限公司 风力发电机组的故障触发调节方法和装置
CN115045803B (zh) * 2021-03-08 2025-08-12 北京金风科创风电设备有限公司 风力发电机组的故障触发调节方法和装置
JP2023069397A (ja) * 2021-11-05 2023-05-18 株式会社東芝 設備点検計画作成システム、および、設備点検計画作成方法
CN114183314A (zh) * 2021-11-25 2022-03-15 吉林省电力科学研究院有限公司 一种基于可靠度的风电机组机会维护方法
CN115130702B (zh) * 2022-09-02 2022-12-02 山东汇泓纺织科技有限公司 一种基于大数据分析的纺织机故障预测系统
CN115130702A (zh) * 2022-09-02 2022-09-30 山东汇泓纺织科技有限公司 一种基于大数据分析的纺织机故障预测系统
CN115860212A (zh) * 2022-11-29 2023-03-28 国网福建省电力有限公司经济技术研究院 一种配电网的风险预测方法与终端

Also Published As

Publication number Publication date
JP2019113883A (ja) 2019-07-11
TW201734870A (zh) 2017-10-01
TWI607328B (zh) 2017-12-01

Similar Documents

Publication Publication Date Title
WO2017163561A1 (fr) Dispositif de support d'exploitation et système d'énergie éolienne
He et al. Condition-based maintenance optimization for multi-component systems considering prognostic information and degraded working efficiency
Nielsen et al. A case study on risk-based maintenance of wind turbine blades with structural health monitoring
KR102633236B1 (ko) 시뮬레이션 방법 및 시스템
Van Horenbeek et al. Quantifying the added value of an imperfectly performing condition monitoring system—Application to a wind turbine gearbox
CN106404403B (zh) 用于涡轮机的分析的方法和系统
Rinaldi et al. A decision support model to optimise the operation and maintenance strategies of an offshore renewable energy farm
Al‐Najjar On establishing cost‐effective condition‐based maintenance: Exemplified for vibration‐based maintenance in case companies
US9188021B2 (en) Steam turbine blade vibration monitor backpressure limiting system and method
TWI546762B (zh) Wind power generation equipment of the stress estimation device and wind power equipment, the stress estimation method, wind power generation system
JP2018060387A (ja) 予兆診断装置及びそれを有する発電装置制御システム
Dawid et al. Review of Markov models for maintenance optimization in the context of offshore wind
Zhang et al. Condition based maintenance and operation of wind turbines
WO2023286659A1 (fr) Dispositif de prédiction de défaillance, procédé de prédiction de défaillance et programme
JP2024057316A (ja) 故障確率評価システムおよび故障確率評価方法
Li et al. Advances in intelligent reliability and maintainability of energy infrastructure assets
US20240280084A1 (en) Systems and methods for monitoring wind turbines using wind turbine component vibration data
EP4325314A2 (fr) Systèmes et procédés d'estimation du risque futur de défaillance d'un composant d'éolienne à l'aide d'un apprentissage automatique
JP2024094823A (ja) 故障評価システム、故障評価方法、及び故障評価プログラム
Dinh Opportunistic predictive maintenance for multi-component systems with multiple dependences
Bore Analysis of management methods and application to maintenance of geothermal power plants
Wang Degradation Assessment and Maintenance Decision-making for Mechanical and Electrical Assets Driven by Condition Data
Vootukuru DMARC: A framework for the integration of DMAIC and DMADV
Carlucci et al. Data-centric Maintenance and Operational Knowledge and Its Impact on Classification
Taşkıner Incipient Fault Detection in Wind Turbines

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17769608

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 17769608

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP