EP4260119A1 - Prédiction de maintenance pour modules d'un microscope - Google Patents

Prédiction de maintenance pour modules d'un microscope

Info

Publication number
EP4260119A1
EP4260119A1 EP21819726.7A EP21819726A EP4260119A1 EP 4260119 A1 EP4260119 A1 EP 4260119A1 EP 21819726 A EP21819726 A EP 21819726A EP 4260119 A1 EP4260119 A1 EP 4260119A1
Authority
EP
European Patent Office
Prior art keywords
evaluation unit
control
assembly
operating state
microscope
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.)
Pending
Application number
EP21819726.7A
Other languages
German (de)
English (en)
Inventor
Markus Weber
Stefan Bamberger
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.)
Leica Microsystems CMS GmbH
Original Assignee
Leica Microsystems CMS GmbH
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 Leica Microsystems CMS GmbH filed Critical Leica Microsystems CMS GmbH
Publication of EP4260119A1 publication Critical patent/EP4260119A1/fr
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G02—OPTICS
    • G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
    • G02B21/00—Microscopes
    • G02B21/36—Microscopes arranged for photographic purposes or projection purposes or digital imaging or video purposes including associated control and data processing arrangements
    • G02B21/365—Control or image processing arrangements for digital or video microscopes
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/008—Reliability or availability analysis
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
    • G06F11/3409—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
    • G06F11/3452—Performance evaluation by statistical analysis
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
    • G06F11/3013—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system is an embedded system, i.e. a combination of hardware and software dedicated to perform a certain function in mobile devices, printers, automotive or aircraft systems
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2201/00—Indexing scheme relating to error detection, to error correction, and to monitoring
    • G06F2201/81—Threshold

Definitions

  • the invention relates to an assembly for a microscope and a method and a computer program for determining operating states of an assembly of a microscope.
  • a system is known from document US Pat. No. 9,599,804 B2, in which tests on a microscope are carried out remotely from a central control server. The tests are evaluated on the central control server.
  • the document DE 10 2018 133 196 A1 discloses a method for determining the need for maintenance of a microscope using an image recorded with an image sensor of the microscope. This detects errors in the microscope that affect the image. It is not possible to detect possible errors before they affect the image or errors that do not affect the image at all. It is therefore usually only possible to determine particularly serious errors.
  • a further disadvantage of the known methods is that error detection requires the installation of additional cost-intensive and space-consuming hardware, in particular in the form of sensors or data transmission elements.
  • the assembly for a microscope includes at least one actuator and/or at least one sensor and a control and evaluation unit.
  • the control and evaluation unit is designed to determine current parameter values of parameters of the sensor and/or the actuator and to analyze the determined parameter values using a statistical model stored in the control and evaluation unit. Furthermore, the control and evaluation unit is designed to determine an operating state value based on this analysis, which characterizes an operating state of the assembly, and to determine the operating state as an abnormal operating state if the operating state value meets a predetermined condition in relation to a threshold value.
  • the parameters of the sensor or actuator are in particular technical operating variables of the sensor or actuator.
  • the actuator is an electric motor, for example.
  • the detected parameter of the actuator which characterizes the operation, can then be, for example, the power consumption of the electric motor or its temperature or its running time over a specified travel range.
  • the sensor is, for example, a contact sensor that assigned to an assembly that can be positioned and which supplies a contact signal as a parameter when the assembly has reached a specific position. If several contact sensors are present, several contact signals can also be detected and, for example, their time interval can be evaluated in order to derive a travel speed as a parameter.
  • the parameter values serve as input data for the statistical model.
  • Corresponding output data of the statistical model include the operating state value.
  • the threshold value is stored in the control and evaluation unit. This enables the operating state to be determined particularly quickly and efficiently.
  • the threshold value is stored in a computer network.
  • the computer network can in particular be a so-called cloud in which the threshold value is stored decentrally.
  • the threshold value can be adjusted in a particularly simple manner, and a particularly flexible determination of the operating state is made possible.
  • the threshold value can be updated via a computer network, in particular in such a way that the threshold value stored in the control and evaluation unit can be overwritten with a new, additional threshold value. This enables a particularly flexible determination of the operating state.
  • control and evaluation unit includes a microcontroller.
  • the microcontroller enables the operating state of the assembly to be determined particularly efficiently and quickly, in particular a determination of the operating state in real time. Furthermore, the microcontroller enables a particularly space-saving design.
  • the statistical model includes a multivariate distribution function.
  • a multivariate distribution function describes the probability that a random variable will take on a value less than or equal to a number x.
  • the value of the random variable depends on a large number of parameters.
  • Such a random variable is also called a random vector or an input vector.
  • This enables not only individual errors to be detected, but also more complex changes, such as interactions between parameters that can lead to an error.
  • a limit value is considered for the entirety of the parameters. The aim is therefore to determine deviations from the normal case in order to determine maintenance or repair measures before a defect occurs, i.e. in a proactive manner. This means that the operating status of the assembly is determined particularly reliably and efficiently.
  • the statistical model is created using machine learning.
  • input data is evaluated (“labeled”) by assigning the possible value range of the parameter value, for example, to a faultless or impaired function of the assembly or to a non-function of the assembly. This makes it possible to analyze a particularly large number of parameters and so on determine the operating status with particular certainty.
  • the control and evaluation unit is designed to output a service message in the event of an abnormal operating state, with the service message being able to be output locally to the microscope and/or to a remote device via a computer network. This makes it possible to react particularly quickly to an abnormal operating state and to take preventive maintenance measures particularly efficiently.
  • control and evaluation unit is designed to continuously determine the operating state of the assembly during operation. This makes it possible to ensure uninterrupted operation of the microscope.
  • control and evaluation unit can be updated via a computer network. This enables the operating state of the assembly to be determined particularly quickly and efficiently.
  • control and evaluation unit is designed to store the parameter values in the control and evaluation unit each with a time stamp and to analyze the parameter values of the respective parameter over time. This makes it possible to detect gradual changes in the parameter values when determining the operating state in a particularly simple manner.
  • control and evaluation unit is designed to determine the operating state without image data analysis. This enables the operating status to be determined efficiently and quickly, especially for assemblies that do not affect image data.
  • Another aspect of the invention relates to a method for determining a
  • Evaluation unit comprises at least one actuator and / or at least one sensor. That The method includes the following steps: determining current parameter values of parameters of the sensor and/or the actuator; Analyzing the determined parameter values using a statistical model stored in the control and evaluation unit and, based on this analysis, determining an operating state value which characterizes an operating state of the assembly; and determining the operating condition as an abnormal operating condition when the operating condition value satisfies a predetermined condition related to a threshold value.
  • a further aspect of the invention relates to a computer program with a program code for executing the method for determining an operating state.
  • a further aspect of the invention relates to a microcontroller with a program code for executing the method for determining an operating state.
  • FIG. 1 shows a schematic structure of an assembly of a microscope
  • FIG. 2 shows a flow chart of a method for creating a statistical model
  • FIG. 3 shows a flow chart of a method for determining operating states of the assembly according to FIG. 1, and
  • Figure 4 shows a system 400 comprising a microscope 410 and a computer system 420.
  • FIG. 1 shows a schematic structure of an assembly 100 of a microscope.
  • the microscope can in particular be a light microscope, for example a confocal microscope or a wide-field microscope such as a light sheet microscope.
  • the assembly 100 includes a sensor 102 and an actuator 104, which are each connected to a control and evaluation unit 106 and are controlled by this.
  • the microscope includes a large number of different assemblies of the type shown in FIG. 1.
  • the assemblies monitor and/or carry out various functions of the microscope.
  • the assembly 100 is a mirror unit which selectively moves a mirror into a position in a beam path of the microscope or into a position outside the beam path.
  • the actuator 104 of the mirror unit includes, for example, a motor that moves the mirror.
  • the sensor 102 can be used to determine whether the mirror has reached one of the two end positions as a result of the motor-driven movement.
  • the sensor 102 is a contact sensor, for example.
  • the actuator 104 includes a motor output stage and a motor encoder.
  • Control and evaluation unit 106 may be connected.
  • the assembly 100 additionally include other contact sensors that determine other positions of the mirror.
  • At least one sensor 102 or at least one actuator 104 can be connected to the control and evaluation unit 106 in each case.
  • assemblies of the microscope are: a focus unit that adjusts a motorized focus drive of an objective, a stage unit that shifts a motorized stage, and a nosepiece unit that changes objectives of the microscope in a motorized manner.
  • actuators are motors, for example for belt drives, and lifting magnets.
  • sensors are light barriers, Hall sensors, microswitches, temperature sensors, humidity sensors and rotary encoders.
  • the control and evaluation unit 106 is in particular a microcontroller. Of the
  • Microcontrollers include at least a processor and a memory element. Such a microcontroller is also referred to as "system-on-a-chip".
  • the control and evaluation unit 106 can comprise a microprocessor or an integrated circuit such as a field programmable gate array.
  • the control and evaluation unit 106 is connected, for example, to a internal bus system 108 of the microscope.
  • the control and evaluation unit 106 receives control commands from an internal system computer of the microscope as a function of an operating program of the system computer via the internal bus system 108.
  • the control and evaluation unit 106 also implements these control commands Using the actuator 104 and the sensor 102 to.
  • the control and evaluation unit 106 calculates motor control signals in the form of a pulse width modulation value (so-called PWM value) and controls the motor via the motor output stage.
  • PWM value pulse width modulation value
  • the position of the motor is continuously determined by the control and evaluation unit 106 with the aid of the motor encoder and a finite state machine.
  • a finite state machine is an electronic circuit that, based on input data, such as a motor encoder signal, determines a limited number of output data, such as motor encoder positions.
  • sensor 102 determines when one of the two end positions of the mirror has been reached.
  • Parameters of the sensor 102 and of the actuator 104 are monitored during the operation of the microscope.
  • the parameters are read out by the control and evaluation unit 106 and the determined parameter values are analyzed by the control and evaluation unit 106 .
  • Parameters that are assigned to an electric motor can be, for example, torque, motor current and/or speed, control commands, and errors that have occurred.
  • Parameters that are assigned to a contact sensor can be, for example, the switching state of the contact, current consumption, electrical resistance on contact and/or switching times, and errors that have occurred.
  • the corresponding parameter values can also be generated by the control and evaluation unit 106 itself, for example when driving the motor using the motor control signals and the motor output stage, as described above. Furthermore, the control and evaluation unit 106 can generate the corresponding parameter values with the aid of the motor coder. It is also possible for the control and evaluation unit 106 to determine further secondary parameter values from primarily determined parameter values.
  • parameter values can be provided with a time stamp. This time stamp allows each parameter value to be assigned to the point in time and/or time segment at which the parameter value is determined by the control and evaluation unit 106 .
  • parameter values determined from microscope images by the assembly 100 are not image data that are determined using an image sensor of the microscope.
  • the parameter values are not image data that are determined using an image sensor of the microscope.
  • the control and evaluation unit 106 is connected to a computer network 110 via the bus system 108 .
  • the computer network 110 is, for example, a computer at the location of the microscope or a server with which the control and evaluation unit 106 can exchange data locally.
  • the computer network 110 can be a decentralized cloud, which in particular consists of a number of servers connected to one another via a network connection.
  • the control and evaluation unit 106 can thus exchange data with remote, external computers.
  • control and evaluation unit 106 works with a statistical model 112.
  • the statistical model 112 is stored in the control and evaluation unit 106, for example in a memory element of the microcontroller.
  • the parameter values ascertained with the aid of the control and evaluation unit 106 are analyzed with the aid of the statistical model 112 .
  • an operating state is determined using a method that is explained in more detail below with reference to FIG.
  • a service message 114 can be output by the control and evaluation unit 106 based on the determined operating state.
  • the statistical model 112 is represented by a multivariate distribution function, for example.
  • a multivariate distribution function describes the probability that a random variable will take on a value less than or equal to a number x.
  • the value of the random variable depends on a large number of parameters.
  • Such a random variable is also called a random vector or an input vector.
  • the input vector includes a large number of parameter values.
  • FIG. 2 shows a flow chart of a method for creating the statistical model 112, in particular the multivariate distribution function.
  • the method starts in a step S210.
  • training data of the assembly 100 are recorded.
  • These data are in particular special parameter values of the parameters assigned to the actuator 104 or sensor 102 .
  • the parameters for the assembly 10 are, for example, control commands corresponding to a target position to which the mirror is to be pivoted by the actuator 104, a motor encoder position corresponding to a position of a motor axis, a contact sensor position corresponding to one of the two end positions of the mirror and a PWM value for controlling the actuator 104.
  • These parameter values are determined during operation of the assembly 100 at short time intervals.
  • the sampling interval is 100ms, for example.
  • the parameter values are preferably determined for a plurality of assemblies 100, so that for each of the assemblies 100 a data set of parameter values is determined and is available. Furthermore, it must be ensured that the data record includes both assemblies with normal operating states and assemblies with abnormal operating states.
  • Table 1 shows the parameter values determined in this way for the assembly 100 from the time 0 ms with a control command for the position change issued at the time 100 ms.
  • step S214 the data records recorded in step S212 are given a rating.
  • the evaluation of the data sets is also called "la at n". This evaluation is carried out, for example, by a development engineer and assesses whether the parameter values indicate an error at a specific point in time. This evaluation corresponds to an evaluation of the operating state of the assembly 100, in particular the actuator 104 and sensor 102.
  • the evaluation is described here as a binary variable, for example Alternatively, the evaluation can be a discrete or continuous variable.
  • determined data for the assembly 100 are provided with ratings, the rating "0" identifying an abnormal operating state of the assembly 100, in particular an error in the assembly 100.
  • a step S216 the collected and evaluated training data are then used in the form of parameter values as input vectors for training the multivariate distribution function.
  • the multivariate distribution function can be, for example, a neural network consisting of a large number of interconnected nodes.
  • the nodes of the neural network process the input vectors and calculate an output vector using a weight and an activation function.
  • the weightings of the individual nodes are varied iteratively in such a way that the error between the output vector and the rating allocated in step S214 is minimized.
  • step S2108 the multivariate distribution function learned in step S216 is exported into executable program code.
  • Input and output vectors are implemented as transfer parameters.
  • the program code also includes an array that includes the weightings determined in step S216.
  • step S220 the multivariate distribution function is integrated into the control and evaluation unit 106 in the form of the program code exported in step S218.
  • the multivariate distribution function is stored in the storage element of the control and evaluation unit 106 .
  • the storage provided in step S220 can take place during the assembly of the microscope, so that the microscope is delivered with the multivariate distribution function stored in the assembly 100 .
  • the multivariate distribution function can be subsequently stored in the storage element, for example during a functional upgrade of the microscope.
  • the multivariate distribution function can be updated, in particular via the computer network 110. For example, an update from a decentralized cloud is possible.
  • the method ends in a step S222.
  • the method according to FIG. 2 is normally carried out on a powerful computer. This calculator does not have to be part of the microscope.
  • FIG. 3 shows a flowchart of a method for determining an operating state of assembly 100 of the microscope.
  • the method starts in a step S310.
  • parameter values of the actuator 104 and the sensor 102 of the assembly 100 are recorded by the control and evaluation unit 106 at a point in time t.
  • the Provide parameter values with a time stamp in order to enable a chronological assignment of the recorded parameter values.
  • the timestamp can be a time indication that is measured relative to an event.
  • the event can be a function activation of the assembly 100 in particular.
  • the time specification can be absolute, for example the time with or without a date specification.
  • the parameters read out by the control and evaluation unit 106 are, in particular, control commands corresponding to a target position into which the mirror is to be pivoted by the actuator, a motor encoder position corresponding to a position of a motor axis, a contact sensor position corresponding to one of the two end positions of the mirror and a PWM value to control the actuator.
  • step S314 the parameter values determined in step S312 are then bundled into an input vector.
  • Table 3 shows, for example, parameter values of such an input vector.
  • step S316 the input vector is analyzed by the control and evaluation unit 106 using the multivariate distribution function.
  • the multivariate distribution function outputs an operating state value.
  • This operating state value represents the output vector.
  • the parameters used in steps S312 and S314 correspond to the parameters used to learn the multivariate distribution function (see method according to FIG. 2, in particular step S212).
  • the acquisition and processing of the parameter values in steps S312 and S314 and the analysis in step S316 take place continuously in real time. This means that the parameter values are determined at short time intervals during operation of the assembly 100 . For example, this can be at intervals in a range from 10 ms to 200 ms, in particular occur at intervals of 100ms.
  • the time range of less than 10ms and greater than 100ms is also suitable for use.
  • the acquisition of the parameters essentially depends on the conditions of the assembly.
  • the analysis then takes place immediately after the determination of the parameter values. Processing in real time is made possible, for example, by the fact that all processing and analysis of the parameter values and determination of the operating state takes place within the assembly 100, in particular in the control and evaluation unit 106. A time-consuming and delaying transfer of parameter values or other data to determine the operating status does not take place.
  • step S318 an operating state of the assembly 100 is then determined by the control and evaluation unit 106 based on the output vector.
  • the output vector is compared to a threshold value. In the event that the output vector falls below the threshold, an abnormal operating condition is determined.
  • the operating state of assembly 100 is thus determined based on the parameter values of the assembly, in particular parameters of assembly 100 assigned to actuator 104 or sensor 102 .
  • the threshold may be transmitted over the computer network 110 as described for the multivariate distribution function. In particular, an update from a decentralized cloud is possible.
  • the abnormal operating state of assembly 100 is thus determined in particular when at least actuator 104 or sensor 102 is in an abnormal operating state.
  • This abnormal operating state can be caused, among other things, by intrinsic factors, such as an intrinsically faulty actuator 104, or by extrinsic factors, such as a gradual deterioration in the function of the actuator 104 caused by wear and tear or dirt.
  • the goal is to detect a possible failure of the actuator 104 before the actual failure of the actuator 104, in particular to detect at a time when the actuator 104 has not yet failed and the proper functioning of the actuator 104 is still guaranteed. In the result is so that the proper function of the corresponding assembly 100 continues to be guaranteed.
  • the abnormal operating state of the actuator 104 indicates a failure of the actuator 104 in the near future.
  • step S320 the control and evaluation unit 106 outputs the service message 114 about the determined operating state of the microscope.
  • This service message 114 is output in particular when an abnormal operating state has been determined.
  • the service message 114 is output on the microscope, in particular on a display unit of the microscope, or on a local computer. Alternatively or additionally, the service message 114 is output via the computer network 110 to a remote service computer. In this way, for example, a service technician can be informed about necessary maintenance.
  • the service message 114 can contain a list of the actuators 104 and sensors 102 that have an abnormal operating state.
  • corresponding spare parts can preferably be ordered automatically based on the list, or a service appointment can be scheduled for the corresponding microscope.
  • the method ends in a step S322.
  • the operating state is not only determined on the basis of individual parameters, but also taking into account a large number of parameters, in particular their interaction. This allows a possible failure of the assembly 100 to be determined particularly efficiently before the assembly 100 actually fails.
  • FIG. 4 shows a schematic representation of a system 400 designed to carry out a method described herein.
  • the system 400 includes a microscope 410 and a computer system 420.
  • the microscope 410 is designed to record images and is connected to the computer system 420.
  • Computer system 420 is configured to perform at least part of a method described herein.
  • Computer system 420 may be configured to execute a machine learning algorithm.
  • the computer system 420 and the microscope 410 can be separate units, but can also be integrated together in a common housing.
  • the computer system 420 could be part of a central processing system of the microscope 410 and/or the computer system 420 could be part of a sub-component of the microscope 410, such as a sensor, an actuator, a camera or an illumination unit, etc. of the microscope 410.
  • Computing system 420 may be a local computing device (e.g., personal computer, laptop, tablet computer, or cellular phone) with one or more processors and one or more storage devices, or may be a distributed computing system (e.g., a cloud computing system with a or multiple processors, or one or more storage devices distributed at different locations, for example at a local client and/or one or more remote server farms and/or data centers).
  • Computer system 420 may include any circuit or combination of circuits.
  • computer system 420 may include one or more processors, which may be of any type.
  • processor can mean any type of computing circuitry, such as for example, but not limited to, a microprocessor, a microcontroller, a complex instruction set (CISC) microprocessor, a reduced instruction set (RISC) microprocessor, a Very Long Instruction Word (VLIW) microprocessor, a graphics processor , a digital signal processor (DSP), a multi-core processor, a field programmable gate array (FPGA), eg of a microscope or microscope component (eg camera) or any other type of processor or processing circuitry.
  • CISC complex instruction set
  • RISC reduced instruction set
  • VLIW Very Long Instruction Word
  • DSP digital signal processor
  • FPGA field programmable gate array
  • Computer system 420 may be custom built circuitry, an application specific integrated circuit (ASIC), or the like, such as one or more circuits (e.g., a communications circuit) for use in wireless devices such as e.g. B. mobile phones, tablet computers, laptop computers, two-way radios and similar electronic systems.
  • Computer system 420 may include one or more storage devices, which may include one or more storage elements appropriate for the particular application, such as main memory in the form of random access memory (RAM), one or more hard drives, and/or one or more drives handling removable media such as CDs, flash memory cards, DVD, and the like.
  • RAM random access memory
  • HDD hard drives
  • drives handling removable media such as CDs, flash memory cards, DVD, and the like.
  • Computer system 420 may also include a display device, one or more speakers, and a keyboard and/or controller, which may include a mouse, trackball, touch screen, voice recognition device, or any other device that allows a system user to enter information into computer system 420 and to receive information from the same.
  • a display device one or more speakers
  • a keyboard and/or controller which may include a mouse, trackball, touch screen, voice recognition device, or any other device that allows a system user to enter information into computer system 420 and to receive information from the same.
  • Some or all of the method steps may be performed by (or using) a hardware device, such as a processor, microprocessor, programmable computer, or electronic circuit. In some embodiments, one or more of the main method steps can be performed by such a device. Depending on particular implementation requirements, embodiments of the invention can be implemented in hardware or software. The implementation can be performed with a non-volatile storage medium such as a digital storage medium such as a floppy disk, a DVD, a Blu-Ray, a CD, a ROM, a PROM and EPROM, an EEPROM or a FLASH memory in which electronically readable control signals are stored which interact (or can interact) with a programmable computer system in such a way that the respective method is carried out. Therefore, the digital storage medium can be computer-readable.
  • a non-volatile storage medium such as a digital storage medium such as a floppy disk, a DVD, a Blu-Ray, a CD, a ROM, a PROM and EPROM, an
  • Some exemplary embodiments according to the invention comprise a data carrier with electronically readable control signals which can interact with a programmable computer system so that one of the methods described herein is carried out.
  • embodiments of the present invention may be implemented as a computer program product having program code, where the program code is operable to perform one of the methods when the computer program product is run on a computer.
  • the program code can be stored on a machine-readable carrier, for example.
  • an embodiment of the present invention is therefore a computer program with a program code for performing one of the methods described herein when the computer program runs on a computer.
  • a further embodiment of the present invention is therefore a storage medium (or a data carrier or a computer-readable medium) which comprises a computer program stored thereon for carrying out one of the methods described herein, when run by a processor.
  • the data carrier, the digital storage medium or the recorded medium is usually tangible and/or not seamless.
  • Another embodiment of the present invention is an apparatus as described herein, including a processor and the storage medium.
  • a further exemplary embodiment of the invention is therefore a data stream or a signal sequence which represents the computer program for carrying out one of the methods described herein.
  • the data stream or burst may be configured to be transmitted over a data communications link, such as the Internet.
  • Another embodiment includes a processing means, for example a computer or a programmable logic device, configured or adapted to perform any of the methods described herein.
  • a processing means for example a computer or a programmable logic device, configured or adapted to perform any of the methods described herein.
  • Another embodiment includes a computer on which the computer program for executing one of the methods described herein is installed.
  • Another embodiment according to the invention includes an apparatus or system configured to transmit (e.g. electronically or optically) a computer program for performing any of the methods described herein to a recipient.
  • the recipient may be a computer, mobile device, storage device, or the like.
  • the device or system may include a file server for transmitting the computer program to the recipient.
  • a programmable logic device eg, a field programmable gate array, FPGA
  • FPGA field programmable gate array
  • a field programmable gate array having a Microprocessor working together to perform any of the methods described herein. In general, the methods are preferably performed by any hardware device.
  • Embodiments may be based on using a machine learning model or machine learning algorithm.
  • Machine learning can refer to algorithms and statistical models that computer systems can use to perform a specific task without using explicit instructions, rather than relying on models and inference.
  • a transformation of data that can be derived from an analysis of historical and/or training data can be used.
  • the content of images can be analyzed using a machine learning model or using a machine learning algorithm.
  • the machine learning model can be trained using training images as input and training content information as output.
  • the machine learning model By training the machine learning model with a large number of training images and/or training sequences (e.g., words or phrases) and associated training content information (e.g., labels or annotations), the machine learning model "learns" the content of the images recognize, so the content of images not included in the training data can be recognized using the machine learning model.
  • training a machine learning model using training sensor data and For a desired output, the machine learning model "learns" a conversion between the sensor data and the output, which can be used to provide an output based on non-training sensor data provided to the machine learning model.
  • the provided data e.g.
  • Machine learning models can be trained using training input data.
  • the above examples use a training method called "supervised learning".
  • supervised learning the machine learning model is trained using a plurality of training samples, each sample being a plurality of input data values and a plurality of desired output values, ie each training sample associated with a desired output value
  • the machine learning model "learns" which output value to provide based on an input sample that is similar to the samples provided during training.
  • semi-supervised learning can also be used. In semi-supervised learning, some of the training samples lack a desired output value.
  • Supervised learning can be based on a supervised learning algorithm (e.g. a classification algorithm, a regression algorithm or a similarity learning algorithm).
  • Classification algorithms can be used when the outputs are constrained to a limited set of values (categorical variables), i.e. the input is classified as one from the limited set of values.
  • Regression algorithms can be used when the outputs show any numerical value (within a range). Similarity learning algorithms can be similar to both classification and regression algorithms, but are based on learning from examples using a similarity function that measures how similar or related two objects are.
  • unsupervised learning can be used to train the machine learning model.
  • input data may (only) be provided and an unsupervised learning algorithm may be used to find structure in the input data (e.g. by grouping or clustering the input data, finding commonalities in the data).
  • Clustering is the assignment of input data, comprising a plurality of input values, into subsets (clusters) such that input values within the same cluster are divided according to one or more (Predefined) similarity criteria are similar while dissimilar to input values included in other clusters.
  • Reinforcement learning is a third group of machine learning algorithms.
  • reinforcement learning can be used to train the machine learning model.
  • one or more software actors are trained to take actions in an environment. Based on the actions taken, a reward is computed.
  • Reinforcement learning is based on training the one or more software agents to choose actions such that the cumulative reward is increased, resulting in software agents getting better at the task they are given (as by increasing rewards proven).
  • feature learning can be used.
  • the machine learning model may be trained at least in part using feature learning and/or the machine learning algorithm may include a feature learning component.
  • Feature learning algorithms called representation learning algorithms, can preserve the information in their input but transform it in a way that makes it useful, often as a pre-processing stage before performing classification or prediction.
  • feature learning can be based on principal component analysis or cluster analysis.
  • anomaly detection ie, outlier detection
  • the machine learning model may be trained at least in part using anomaly detection and/or the machine learning algorithm may include an anomaly detection component.
  • the machine learning algorithm can use a decision tree as a prediction model.
  • the machine learning model can be based on a decision tree.
  • the observations about an item e.g., a set of input values
  • an output value corresponding to the item may be represented by the leaves of the decision tree.
  • Decision trees can support both discrete and continuous values as output values. If discrete values are used, the decision tree can be called a classification tree, if continuous values are used, the decision tree can be called a regression tree.
  • Association rules are another technique that can be used in machine learning algorithms.
  • the machine learning model can be based on one or more association rules.
  • Association rules are created by identifying relationships between variables in large data sets.
  • the machine learning algorithm may identify and/or utilize one or more relationship rules that represent knowledge derived from the data.
  • the rules can e.g. B. be used to store, manipulate or apply the knowledge.
  • Machine learning algorithms are usually based on a machine learning model.
  • the term “machine learning algorithm” can denote a set of instructions that can be used to create, train, or use a machine learning model.
  • the term “machine learning model” can mean a data structure and/or a denote the set of rules that represents the learned knowledge (e.g., based on the training performed by the machine learning algorithm).
  • the use of a machine learning algorithm may imply the use of an underlying machine learning model (or a plurality of underlying machine learning models).
  • the use of a machine learning model may imply that the machine learning model and/or the data structure/set of rules that is/are the machine learning model is trained by a machine learning algorithm.
  • the machine learning model may be an artificial neural network (ANN).
  • ANNs are systems inspired by biological neural networks such as those found in a retina or brain.
  • ANNs comprise a plurality of interconnected nodes and a plurality of connections, called edges, between the nodes.
  • Each node can represent an artificial neuron.
  • Each edge can send information from one node to another.
  • the output of a node can be defined as a (non-linear) function of its inputs (e.g. the sum of its inputs).
  • a node's inputs can be used in the function based on a "weight" of the edge or node providing the input.
  • the weight of nodes and/or edges can be adjusted in the learning process.
  • the training of a Artificial neural network may include adjusting the weights of the nodes and/or edges of the artificial neural network, i.e. to achieve a desired output for a particular input.
  • the machine learning model can be a support vector machine, a random forest model, or a gradient boosting model.
  • Support Vector Machines ie support vector networks
  • Support Vector Machines are supervised learning models with associated learning algorithms that can be used to analyze data (e.g. in a classification or regression analysis).
  • Support Vector Machines can be trained by providing input with a plurality of training input values belonging to one of two categories. The Support Vector Machine can be trained to assign a new input value to either category.
  • the machine learning model can be a Bayesian network, which is a probabilistic directed acyclic graphical model. A Bayesian network can take a set of random variables and their conditional Represent dependencies using a directed acyclic graph.
  • the machine learning model can be based on a genetic algorithm, which is a search algorithm and heuristic technique that mimics the process of natural selection.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Quality & Reliability (AREA)
  • Computer Hardware Design (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Probability & Statistics with Applications (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Optics & Photonics (AREA)
  • Testing And Monitoring For Control Systems (AREA)

Abstract

L'invention concerne un module (100) pour un microscope comprenant au moins un actionneur (104) et/ou au moins un capteur (102), et une unité de commande et d'évaluation (106). L'unité de commande et d'évaluation (106) est conçue pour déterminer des valeurs de paramètre actuelles de paramètres du capteur (102) et/ou de l'actionneur (104) et pour analyser les valeurs de paramètre déterminées, à l'aide d'un modèle statistique (112) stocké dans l'unité de commande et d'évaluation (106). L'unité de commande et d'évaluation (106) est également conçue pour déterminer une valeur d'état de fonctionnement sur la base de cette analyse, laquelle valeur d'état de fonctionnement caractérise un état de fonctionnement du module et définit l'état de fonctionnement comme un état de fonctionnement anormal si la valeur d'état de fonctionnement satisfait une condition prédéfinie par rapport à une valeur de seuil.
EP21819726.7A 2020-12-09 2021-11-15 Prédiction de maintenance pour modules d'un microscope Pending EP4260119A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
DE102020132787.0A DE102020132787A1 (de) 2020-12-09 2020-12-09 Wartungsvorhersage für Baugruppen eines Mikroskops
PCT/EP2021/081658 WO2022122308A1 (fr) 2020-12-09 2021-11-15 Prédiction de maintenance pour modules d'un microscope

Publications (1)

Publication Number Publication Date
EP4260119A1 true EP4260119A1 (fr) 2023-10-18

Family

ID=78821943

Family Applications (1)

Application Number Title Priority Date Filing Date
EP21819726.7A Pending EP4260119A1 (fr) 2020-12-09 2021-11-15 Prédiction de maintenance pour modules d'un microscope

Country Status (5)

Country Link
US (1) US12585566B2 (fr)
EP (1) EP4260119A1 (fr)
CN (1) CN116615717A (fr)
DE (1) DE102020132787A1 (fr)
WO (1) WO2022122308A1 (fr)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20240403113A1 (en) * 2023-06-05 2024-12-05 ZipRecruiter, Inc. Machine learning based matching of entities using multiple queues
CN120805716A (zh) * 2025-07-21 2025-10-17 宏景科技股份有限公司 一种基于改进lstm模型的光学稳定性评估及预测方法

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7096153B2 (en) 2003-12-31 2006-08-22 Honeywell International Inc. Principal component analysis based fault classification
US7478014B2 (en) * 2006-09-29 2009-01-13 Tokyo Electron Limited Method and system for facilitating preventive maintenance of an optical inspection tool
DE102009022394A1 (de) 2009-05-22 2010-11-25 Leica Microsystems Cms Gmbh System und Verfahren zum computergestützten Durchführen mindestens eines Tests bei einem Scanmikroskop
DE102018209108A1 (de) 2018-03-05 2019-09-05 Robert Bosch Gmbh Schnelle Fehleranalyse für technische Vorrichtungen mit maschinellem Lernen
DE102018123436A1 (de) 2018-09-24 2020-03-26 Endress+Hauser Conducta Gmbh+Co. Kg Verfahren zum Überwachen einer Anlage der Automatisierungstechnik
DE102018217903A1 (de) * 2018-10-18 2020-04-23 Leica Microsystems Cms Gmbh Inferenz Mikroskopie
CA3113084A1 (fr) * 2018-10-31 2020-05-07 Alcon Inc. Systeme et procede d'utilisation de donnees de systemes medicaux
DE102018133196A1 (de) 2018-12-20 2020-06-25 Carl Zeiss Microscopy Gmbh Bildbasierte wartungsvorhersage und detektion von fehlbedienungen
US11307570B2 (en) * 2019-05-31 2022-04-19 Panasonic Intellectual Property Management Co., Ltd. Machine learning based predictive maintenance of equipment

Also Published As

Publication number Publication date
WO2022122308A1 (fr) 2022-06-16
US20240095145A1 (en) 2024-03-21
DE102020132787A1 (de) 2022-06-09
CN116615717A (zh) 2023-08-18
US12585566B2 (en) 2026-03-24

Similar Documents

Publication Publication Date Title
DE102021124253A1 (de) Maschinelles Lernverfahren zur Anomaliedetektion in einem elektrischen System
EP3671632B1 (fr) Prévision d'entretien à base d'image et détection de fausses man uvres
EP3767403B1 (fr) Mesure de forme et de surface assistée par apprentissage automatique destinée à la surveillance de production
DE102019217613A1 (de) Verfahren zur diagnose eines motorzustands und diagnostisches modellierungsverfahren dafür
WO2021121695A1 (fr) Procédé, appareil et système de détection d'états de fonctionnement anormaux d'un dispositif
DE102021124254A1 (de) Maschinelles Lernverfahren zur Leckageerkennung in einem pneumatischen System
EP4193227A1 (fr) Dispositif et procédé de détection d'anomalies dans une installation industrielle pour l'exécution d'un processus de production
EP2971768B1 (fr) Développement d'un modèle supérieur pour contrôler et/ou surveiller un système de compresseurs
EP4260119A1 (fr) Prédiction de maintenance pour modules d'un microscope
DE102022131925A1 (de) Selbstlernendes verfahren für einen fertigungs- und/oder montageprozess
WO2024046780A1 (fr) Commande de systèmes d'équipements techniques du bâtiment
WO2024027868A1 (fr) Procédé de détermination précise d'un couple de sortie et robot collaboratif
WO2025219164A1 (fr) Procédé de génération d'un programme de commande pour un contrôleur logique programmable et système d'automatisation
WO2024008232A1 (fr) Procédé de détection dynamique d'encliquetage dans un robot collaboratif par intelligence artificielle et compensation dynamique des trajectoires
DE102019123447B4 (de) Verfahren zum Betrieb von Fahrzeugkomponenten
DE112024001713T5 (de) Gebäudemanagementsystem mit Ressourcenidentifizierung und -verbesserung
DE102021134294B4 (de) Hybrid-Ansatz zur Schätzung und Überwachung von Reibungs- und Masseparametern in einem elektrischen Antriebssystem
EP4268162A1 (fr) Procédé et système permettant de transformer une situation d'objet de démarrage en une situation d'objet cible (découverte de solution de tacite intuitive)
WO2018197550A1 (fr) Procédé d'analyse informatisée du fonctionnement d'un système de production
EP3907574A1 (fr) Procédé de génération d'une explication d'une décision prise d'un système de commande de fabrication
EP4100180A1 (fr) Appareil pour réguler et/ou surveiller une installation technique
DE102023210500B3 (de) Roboterüberwachung
DE102022119202A1 (de) Verfahren und System zur Koordination kollaborativer Roboter mit Nutzlast-/Ladungs-/Prioritätsbewusstsein sowie kollaborativer Roboter
EP4302163B1 (fr) Procédé et système d'évaluation d'une mesure de maintenance nécessaire pour une machine, plus particulièrement pour une pompe
DE102023123439A1 (de) Verfahren zum Analysieren einer technischen Anlage sowie System

Legal Events

Date Code Title Description
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: UNKNOWN

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE

PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE

17P Request for examination filed

Effective date: 20230424

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

P01 Opt-out of the competence of the unified patent court (upc) registered

Effective date: 20231221

DAV Request for validation of the european patent (deleted)
DAX Request for extension of the european patent (deleted)
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: EXAMINATION IS IN PROGRESS

17Q First examination report despatched

Effective date: 20251021