System for determining a facility state of a facility for metal production and / or metal processing, use of a system of this type and method for determining a facility state

Abstract

The invention relates to a system (1) for determining a facility state of a facility (100) for metal production and / or metal processing, comprising a sound sensor (10) which is designed to receive an acoustic signal; a data processing device (20) which is data-connected to the sound sensor (10), wherein the data processing device (20) is designed to compare at least two data sets with one another, wherein a first data set has a reference signal and a second data set has the acoustic signal received by the sound sensor (10); and, on the basis of the comparison of the data sets, determining a facility state of the facility (100) for metal production and / or metal processing. The system (1) also has a control device (30), which is data-connected to the data processing device (20), wherein the control device (30) is designed to at least indirectly influence an operation of the facility (100) on the basis of the facility state determined by the data processing device (20). The invention also relates to the use of a system (1) of this type and to a method for determining a facility state.

Description

[0001] Page 1 of 74

[0002] Applicant: SMS group GmbH

[0003] Our reference number: P81102WO

[0004] System for determining the state of a plant for metal production and / or metal processing, use of such a system and method for determining the state of a plant

[0005] The invention relates to a system for determining the state of a plant for metal production and / or metalworking, a use of such a system and a method for determining the state of a plant for metal production and / or metalworking.

[0006] The condition of a system, particularly a metal production and / or metalworking plant, can be determined using a variety of different sensors. For example, temperature or pressure sensors are used to detect changes in the respective physical quantities in order to derive information about the system's condition. Depending on the system's condition determined from this information, it can be ascertained, for instance, whether there is a deviation from a previously defined state. The sensors typically used for physical quantities such as temperature, pressure, or acceleration have a limited range of detectable information, restricted to the specific measured change in the physical quantity.This limits the suitability of conventional sensors for determining differentiated deviations of a system state from a previously defined reference system state. Page 2 of 74.

[0007] P81102WO

[0008] The invention is based on the objective of providing an enrichment to the prior art.

[0009] The problem underlying the invention is solved by a system with the features of claim 1. Advantageous embodiments are described in the dependent claims.

[0010] More precisely, the problem underlying the invention, according to a first aspect of the invention, is solved by a system for determining the state of a metal production and / or metalworking plant, comprising a sound sensor configured to receive an acoustic signal, preferably an acoustic signal emanating from the metal production and / or metalworking plant; a data processing device connected to the sound sensor, wherein the data processing device is configured to compare at least two data sets, a first data set comprising a reference signal and a second data set comprising the acoustic signal received from the sound sensor; and determining the state of the metal production and / or metalworking plant based on the comparison of the data sets.The system further includes a control unit which is data-connected to the data processing unit, the control unit being configured to influence the operation of the plant, at least indirectly, based on the plant state determined by the data processing unit.

[0011] A system designed in this way has the advantage that even minor deviations of a system's condition from a reference condition can be detected. This allows maintenance measures to be planned with increased certainty and potential damage to a system to be prevented. For example, based on a determined system condition and the associated information (page 3 / 74)

[0012] P81102WO

[0013] A system designed to detect deviations from a reference state allows for preventive measures, such as maintenance, to be taken before the deviation leads to damage. Furthermore, such a system can increase the process stability of the metal production and / or metalworking plant, thereby improving the product quality of any product, particularly a metallic workpiece, produced and / or processed using such a plant. Finally, such a system offers the advantage of being easily integrated into existing metal production and / or metalworking plants, thus reducing investment costs.

[0014] In the present case, a metal may comprise one or more than one metal, including an ferrous metal, a non-ferrous metal, and / or an alloy. In particular, a metal may comprise a non-ferrous metal, a light metal, and / or a precious metal. A metal may, in particular, comprise steel, aluminum, and / or copper.

[0015] A plant for metal production and / or metalworking can be configured to produce and / or process a metallic workpiece. Preferably, a plant for metal production and / or metalworking is configured to produce and / or process a metallic workpiece, in particular a metallic workpiece with an endless length, at least temporarily continuously.

[0016] A metallic workpiece can be a semi-finished product containing at least one metal, or containing a metal content greater than or equal to 90 wt.%, preferably a metal content greater than or equal to 95 wt.%, and particularly preferably a metal content greater than or equal to 98 wt.%. Page 4 of 74

[0017] P81102WO

[0018] The metallic workpiece has a thickness, a width, and a length. The metallic workpiece can have an endless length. A metallic workpiece with an endless or significantly greater length compared to its width can also be called a metal strip.

[0019] A plant for metal production and / or metal processing can be configured to convey a metallic workpiece, preferably along its length, in a conveying direction. In other words, the conveying direction and the length of the metallic workpiece preferably run parallel to each other, at least in sections, preferably along a common central axis.

[0020] A metal production and / or metalworking plant may include a steel production and / or steelworking plant. A metal production and / or metalworking plant may have one or more than one unit. A metal production and / or metalworking plant may include a primary forming unit, in particular a casting unit, preferably a continuous casting unit, a permanent mold casting unit, and / or a die casting unit. A metal production and / or metalworking plant may include a forming unit, in particular a compression forming unit and / or a tensile forming unit, preferably a rolling mill. A metal production and / or metalworking plant may be equipped to process a metallic workpiece, preferably mechanically and / or thermally.In particular, a plant for metal production and / or metal processing may include a heating device, especially a furnace and / or an induction heating device. Page 5 / 74.

[0021] P81102WO

[0022] A pressure forming device is designed to deform a metallic workpiece by means of compressive forces. A pressure forming device can be a rolling mill or a forging device. A rolling mill can have at least one rolling stand. Preferably, a rolling mill has more than one rolling stand, for example, 2, 3, 4, or more rolling stands. Preferably, the rolling stands are arranged one behind the other with respect to the conveying direction. Advantageously, a metallic workpiece with a starting thickness of greater than or equal to 2 mm is pressure formed, in particular rolled, preferably with a starting thickness of greater than or equal to 6 mm, particularly preferably with a starting thickness of greater than or equal to 10 mm, and most preferably with a starting thickness of greater than or equal to 30 mm.

[0023] A tensile forming device is designed to deform a metallic workpiece by tensile forces. A tensile forming device can be a stretching device, in particular a stretching device for improving the flatness of the metallic workpiece. Advantageously, a metallic workpiece with a starting thickness of less than or equal to 12 mm is tensile formed, in particular stretched, preferably with a starting thickness of less than or equal to 10 mm, and most preferably with a starting thickness of less than or equal to 5 mm.

[0024] A plant for metal production and / or metalworking may include a cutting device. A cutting device may include a chip-forming cutting device, for example, a sawing device, a milling device, preferably a bar milling device, a drilling device, a grinding device, a honing device, a lapping device, and / or another chip-forming cutting device. A cutting device may include a shearing cutting device, for example, a punching device. A cutting device may further include a flame cutting device and / or an EDM device, preferably a page 6 / 74

[0025] P81102WO

[0026] A metal production and / or metalworking plant may include a spark erosion unit and / or a wire erosion unit. A metal production and / or metalworking plant may include a joining unit, preferably a welding unit, a soldering unit, an adhesive bonding unit, and / or a riveting unit. A metal production and / or metalworking plant may include a coating unit, in particular a painting unit, an electroplating unit, a powder coating unit, and / or a zinc plating unit.

[0027] An acoustic signal is the detectable, preferably temporal, progression of a sound by a sensor, preferably a sound sensor. A sound can consist of one or more sound waves. A sound wave can be understood as a mechanical vibration in an elastic medium. This mechanical vibration causes pressure and density fluctuations in the medium, which propagate through it. Therefore, a sound wave can be understood as a deformation propagating as a mechanical wave in an elastic medium.

[0028] A sound wave can originate from a sound source. A sound source is an object that superimposes an additional alternating pressure on the static pressure of an elastic medium, thereby generating one or more sound waves. For example, a body vibrating in an elastic medium is a sound source. According to one interpretation, the metal production and / or metalworking plant and / or a component of the plant, such as a rolling mill, can be a sound source.

[0029] A sound frequency is the number of oscillations or cycles that occur in a sound wave per second. In other words, the sound frequency describes how often the mechanical wave, i.e., the sound wave, repeats itself within one second. The sound frequency is typically measured in Hertz (Hz). Page 7 / 74

[0030] P81102WO is specified. Consequently, an acoustic signal can comprise a sound frequency detectable by a sound sensor. Typically, an acoustic signal comprises a variety of different sound frequencies.

[0031] The elastic medium can comprise a fluid, in particular a gas and / or a liquid, and / or a solid. Sound propagates at a speed of sound that is constant for a medium and its state. The state of the medium includes, in particular, its temperature and / or pressure.

[0032] A sound field is the region in an elastic medium through which sound waves propagate. A sound field can be described by one or more sound field quantities. These quantities include sound pressure, sound displacement, particle velocity, sound acceleration, and other alternating quantities used to describe a changing field, in particular hydrodynamic parameters such as fluctuations in the density and / or temperature of the elastic medium.

[0033] The sound pressure fluctuation refers to the pressure variations of a sound field in an elastic medium that occur during sound propagation. Consequently, the sound pressure fluctuation is the alternating pressure superimposed on a static pressure of an elastic medium. This alternating pressure can be detected by a sound sensor. Therefore, an acoustic signal can include the sound pressure fluctuation detectable by a sound sensor. A sound pressure level is a logarithmic quantity associated with the sound pressure fluctuation. The sound pressure level is defined as the base-10 logarithm of the square ratio between an effective value of the measured sound pressure fluctuation and a reference value. The reference value can be a commonly used value in acoustics of 20 pPa. According to one implementation, the reference value can also be a different value. Page 8 / 74

[0034] P81102WO is a significant value. The effective value of the measured sound pressure level is the root mean square of the sound pressure level.

[0035] Sound displacement refers to the instantaneous distance of a particle in an elastic medium within a sound field from its equilibrium position. Particle velocity refers to the rate at which particles in the elastic medium oscillate around their equilibrium position. Particle velocity is the time derivative of sound displacement. The time derivative of particle velocity is sound acceleration.

[0036] A sound field can also be described by one or more sound energy quantities. These sound energy quantities include sound energy, sound energy density, and other energy quantities used to describe a sound field.

[0037] Sound energy is the energy contained in a sound field and / or a sound event. Sound energy includes both kinetic and potential energy. In other words, the sound energy transmitted by a sound wave depends on the sound pressure level and the particle velocity of the sound wave. Sound energy density is a measure used to describe the sound energy present at a specific location within the sound field. Sound energy density is the sound energy per unit volume.

[0038] A sound event is a physical-acoustic process that is spatially and temporally determined by physical parameters and exists objectively, regardless of whether and / or how the sound event is perceived. A sound event can be described using sound field quantities. A sound event typically involves a sound source. Page 9 / 74

[0039] P81102WO

[0040] Sound power is the sound energy emitted by a sound source per unit of time. Sound intensity is the sound power transmitted per unit area through a surface. Sound intensity can be used to describe a sound field at any point. Furthermore, the sound power of a sound source can be determined using sound intensity. Sound intensity can be measured, for example, using the two-microphone technique known to those skilled in the art.

[0041] A sound sensor can be designed as an electroacoustic transducer. An electroacoustic transducer is preferably configured to convert an acoustic signal into an audio signal. An audio signal is preferably an electrical signal, which in particular comprises an alternating current and / or an alternating voltage.

[0042] An audio signal preferably contains one or more characteristic features of an underlying acoustic signal. A characteristic feature of an acoustic signal is preferably a sound pressure level and / or a sound frequency. The sound pressure level of an acoustic signal can be represented by an amplitude of the audio signal. The sound frequency of an acoustic signal can be represented by a frequency of the audio signal.

[0043] A characteristic feature of an acoustic signal can further comprise a sound field quantity and / or a sound energy quantity and / or a sound power and / or a sound intensity of an acoustic signal. According to one implementation, a characteristic feature of an acoustic signal can be a distribution of sound energy and / or a sound energy density in a previously determined frequency band of an acoustic signal and / or an audio signal. Page 10 / 74

[0044] P81102WO A characteristic feature of an acoustic signal can be another measured and / or calculated quantity to describe the acoustic signal, a sound field and / or a sound source.

[0045] The data processing device is preferably an electronic component. The data processing device is preferably configured to receive, process, and / or transmit electrical signals, in particular audio signals. For this purpose, the data processing device may include a storage unit. A storage unit is preferably configured to store signals, preferably electrical signals, preferably in the form of a data set, and to make them at least temporarily accessible to the data processing device, in particular for processing. A storage unit may comprise a volatile storage unit and / or a non-volatile storage unit. A volatile storage unit may comprise RAM, DRAM, and / or other volatile storage. A non-volatile storage unit may comprise ROM, Flash, and / or other non-volatile storage.

[0046] The data processing device can be configured to receive an audio signal, in particular an acoustic signal converted into an audio signal by the sound sensor, and to store it in the second data record. Furthermore, the data processing device can be configured to receive an optical signal, in particular a signal received from an optical sensor and converted into an electrical signal representing the optical signal, and to store it in a third data record.

[0047] The data processing unit can be wirelessly connected to the sound sensor. According to a preferred embodiment, the data processing unit can be connected by means of a page 11 / 74

[0048] P81102WO

[0049] The sound sensor must be connected to the data network via a wire, in particular by means of a cable.

[0050] The reference signal is preferably an audio signal, in particular an acoustic reference signal converted into an audio signal by a sound sensor. This acoustic reference signal can have been received by a sound sensor at a previously determined time, preferably while the system is in a reference state, wherein the reference state is a known system state. In other words, the reference signal represents a known system state. The acoustic reference signal is preferably an acoustic signal as defined in the invention.

[0051] The reference signal and the underlying acoustic reference signal may have been received and / or converted at a previously determined time and stored in the first data set.

[0052] According to a preferred implementation, the reference signal can be a synthetic signal. A synthetic signal can be a signal generated by means of a simulation and / or a calculation. The simulation and / or calculation can generate an acoustic signal emanating from a metal production and / or metalworking plant and convert it into an audio signal.

[0053] According to one embodiment, the reference signal can comprise an audio signal and / or an electrical reference signal representing an optical reference signal. This electrical reference signal representing the optical reference signal can have been received by an optical sensor at a predetermined time, preferably while the system is in a reference state, wherein the reference state is a known system state. According to one embodiment, the optical reference signal can be a synthetic signal and page 12 / 74

[0054] P81102WO may be generated in particular by means of a simulation and / or a calculation, wherein the simulation and / or calculation can generate an optical signal emanating from a plant for metal production and / or metal processing and convert it into an electrical signal representing the optical signal.

[0055] Preferably, the first data set can comprise a plurality of reference signals. Each reference signal can be associated with a specific system state, wherein the specific system states are distinct from one another. In other words, each reference signal can represent a specific system state. In particular, each reference signal can be generated by an acoustic signal converted by a sound sensor. Each acoustic signal can be received by a sound sensor, preferably at a predetermined time. Furthermore, each reference signal can, in addition to the converted acoustic signal, comprise an optical signal converted into an electrical signal by an optical sensor, wherein each optical signal converted in this way is associated with a converted acoustic signal and preferably represents the same system state.In other words, the reference signal can comprise both a converted acoustic signal and a converted optical signal, so that an acoustic signal and an optical signal are associated with the same system state. This makes it possible to comprehensively describe a system state and, in particular, to determine it using the data processing device. Especially when the data processing device is configured to determine the system state using a machine learning method, combining an acoustic sensor and an optical sensor to increase the information density in the reference signal is advantageous for increasing the accuracy of determining the system state. Page 13 / 74.

[0056] P81102WO

[0057] In particular, any acoustic signal can be an acoustic signal emanating from the metal production and / or metalworking plant. In other words, the plant and / or a component of the plant can be a sound source of the acoustic signal. The plant can be in a specific, preferably known, state at any given time, such that each acoustic signal received by a sound sensor is associated with a different, preferably known, plant state. This makes it possible, by comparing the data sets using the data processing device, to determine differentiated deviations of the current plant state from the previously known plant states.Thus, after determining a preferably known plant state and / or a deviation from a known plant state, the system can take measures that are previously known to be advantageous with regard to the relevant plant state and / or a deviation from the relevant plant state and are assigned in the system to the relevant plant state and / or a deviation from the relevant plant state.

[0058] A data record, in particular the first data record and / or the second data record and / or the third data record, may be stored in the storage unit of the data processing device. In other words, a data record, in particular the first data record and / or the second data record and / or the third data record, may be stored in the storage unit and made accessible to the data processing device, at least temporarily.

[0059] The statement that a data set contains a signal can also be expressed as meaning that the signal is embedded in the data set. In other words, the data set includes a representation of the signal, preferably by means of data points over a time course and / or by means of data points over a frequency band. (See page 14 / 74)

[0060] P81102WO states that the feature that the third data set contains the optical signal received by the optical sensor can also be expressed as meaning that the third data set includes a representation of the optical signal received by the optical sensor. Furthermore, the feature that the second data set contains the acoustic signal received by the sound sensor can also be expressed as meaning that the second data set includes a representation of the acoustic signal received by the sound sensor. Finally, the feature that the first data set contains the reference signal can be expressed as meaning that the first data set includes a representation of the reference signal.

[0061] According to a preferred implementation, the first data record, and preferably also the second and / or third data record, can be stored in external storage, for example, in cloud storage. The external storage can be provided by an external service, such as a server service. The data processing device can be connected to the external storage, preferably wirelessly.

[0062] The data processing device can be configured to compare two data sets, preferably the first and second data sets, based on characteristics characteristic of the data sets. A characteristic feature of a data set can comprise a characteristic feature of a signal stored in the data set, preferably the stored audio signal. The data processing device can be configured to compare three data sets, preferably the first, second, and third data sets, based on characteristics characteristic of the data sets. A characteristic feature of a data set can be a characteristic feature. Page 15 / 74

[0063] P81102WO includes a signal stored in the data set, preferably the stored audio signal and / or the stored optical signal and / or the stored electrical signal representing the optical signal.

[0064] The data processing device can be configured to compare at least two data sets by determining a deviation between the received acoustic signal, preferably the acoustic signal converted into an audio signal, and the reference signal. Furthermore, the data processing device can be configured to compare the three data sets by determining a deviation between the received acoustic signal, preferably the acoustic signal converted into an audio signal, and the reference signal, as well as between the received optical signal, preferably the electrical signal representing the optical signal, and the reference signal.

[0065] According to one variant, the data processing device can be configured to determine an amplitude deviation and / or a frequency deviation between the acoustic signal converted into an audio signal and the reference signal, and / or between the optical signal converted into an electrical signal and the reference signal. For this purpose, the data processing device can be configured to compare the temporal profiles of the signals and to determine a deviation between the frequencies of the signals and / or a deviation between the amplitudes of the signals. According to one implementation, the data processing device can be configured to compare the temporal profiles of the signals over the same time range. Data processing device, page 16 / 74

[0066] The P81102WO can be configured to compare the amplitudes of signals based on their respective largest amplitudes, preferably within the same time range. A time range of a signal's time course is preferably a specific duration.

[0067] Furthermore, the data processing device can be configured to determine deviations from the effective values ​​of the signals. The effective value is the root mean square of a signal, preferably within one period of the signal, and preferably within a predetermined effective value time period. The effective value time period is preferably a previously determined time period.

[0068] According to another variant, the data processing device can be configured to determine an integral deviation between the acoustic signal converted into an audio signal and the reference signal, and / or between the optical signal converted into an electrical signal and the reference signal. For this purpose, the data processing device can be configured to determine an integral of the time course of the acoustic signal converted into an audio signal, an integral of the time course of the reference signal, and / or an integral of the time course of the electrical signal representing the optical signal, with the integrals being determined over the same time period. The data processing device can be configured to compare the previously determined integrals and determine an integral deviation.

[0069] According to another variant, the data processing device can be configured to detect a differential deviation between the acoustic signal converted into an audio signal and the reference signal and / or between the optical signal converted into an electrical signal and the reference signal (page 17 / 74).

[0070] P81102WO can be determined. For this purpose, the data processing device can be configured to determine a time derivative of the time course of the acoustic signal converted into an audio signal, a time derivative of the time course of the reference signal, and / or a time derivative of the time course of the optical signal converted into an electrical signal. The data processing device can be configured to compare the previously determined time derivatives of the signals and to determine any deviation, in particular an amplitude deviation of the signals.

[0071] According to a preferred embodiment, the data processing device can be configured to determine signal deviations by comparing the frequency spectra of the signals. Specifically, the data processing device can be configured to determine deviations in the amplitudes of a specific sound frequency. This allows for the frequency-specific determination of deviations in sound pressure, sound energy, sound energy density, sound power, and / or sound intensity. Furthermore, the data processing device can be configured to determine deviations in the amplitudes of a specific light frequency. This allows for the determination of deviations in the light components of specific wavelengths in the received optical signal and the reference signal.A specific sound frequency of an acoustic signal and / or a specific light frequency of an optical signal can be assigned to a specific component of a metal production and / or metalworking plant. Preferably, a sound frequency and / or a light frequency can be assigned to a rolling mill, more preferably a rolling stand and / or a roll of a rolling stand. A deviation in the amplitude of this specific sound frequency in the frequency spectrum of the acoustic signal and / or a deviation in the amplitude of this specific light frequency in the frequency spectrum of an optical signal can be used to identify the specific component. (Page 18 / 74)

[0072] P81102WO signals can be used to infer with increased accuracy a deviation from a reference state of a device of the plant for metal production and / or metalworking, in particular a rolling device, preferably a rolling stand and / or a roll of a rolling stand.

[0073] According to a preferred embodiment, the data processing device can be configured to determine signal deviations by comparing a multitude of amplitudes of specific frequencies with frequency spectra of the signals, whereby an amplitude of a specific frequency of the acoustic signal converted into an audio signal is compared with an amplitude of the same frequency of the reference signal, and / or an amplitude of a specific frequency of the optical signal converted into an electrical signal is compared with an amplitude of the same frequency of the reference signal. This makes it possible to determine differentiated deviations of even different components of the metal production and / or metalworking plant. Thus, the condition of a plant can be determined even more precisely.

[0074] The control device is preferably an electronic component. The control device is preferably configured to receive, process, and / or transmit signals, preferably electrical signals. For this purpose, the control device may include a storage unit. The control device may preferably be wirelessly connected to the data processing device. The control device may also be connected to the data processing device by means of a wire, preferably a cable.

[0075] The control unit can be configured to receive a signal from the data processing unit, the signal containing information about the system state determined by the data processing unit. (Page 19 / 74)

[0076] P81102WO

[0077] The signal can also include information about the operating state of the metal production and / or metal processing plant. In particular, the information can include the current operating state of the metal production and / or metal processing plant. This allows the control system to better adapt its actions to the existing plant states and operating conditions.

[0078] The control unit can be data-connected to the metal production and / or metal processing plant, preferably wirelessly, alternatively or additionally by means of a wire, preferably by means of a cable.

[0079] The control unit can be configured to send a control signal to the metal production and / or metalworking plant. Preferably, the control unit is configured to directly or indirectly change at least one operating parameter of the metal production and / or metalworking plant by means of a control signal sent to the plant.

[0080] An operating parameter of a metal production and / or metalworking plant can be a power supply, preferably an electrical power supply to the plant. Furthermore, an operating parameter can, for example, be a rolling speed of a rolling mill. An operating parameter can also be a contact pressure of a roll of a rolling mill. An operating parameter can, for example, include a conveying speed of a metallic workpiece in the metal production and / or metalworking plant. An operating parameter can be a pressure, in particular a pressure in a subsystem of the metal production and / or metalworking plant. Preferably, an operating parameter can be a pressure in a hydraulic system of the plant. According to one embodiment, a page 20 / 74

[0081] P81102WO

[0082] Operating parameters may also include the activation, switching, and / or deactivation of a subsystem of the system. A subsystem of the system may comprise a hydraulic system, a lubrication system, a cooling system, an emulsion system, and / or a compressed air system. A hydraulic system is configured to hydraulically manipulate an actuator. For example, such an actuator may comprise a hydraulic cylinder. A hydraulic cylinder may, for example, be configured to apply, maintain, and / or release pressure and / or position, in particular contact pressure and / or stroke, of a roller, preferably a working roller, on a metallic workpiece located in the system.

[0083] A lubrication system is designed to supply moving mechanical components of the system with lubricant. For example, a lubrication system can be designed to supply a gearbox of the system with lubricant.

[0084] A cooling system of the plant is designed to supply the plant with cooling power, in particular one or more than one drive motor of a plant, for example a drive motor of a conveying device and / or a roller, preferably a working roller.

[0085] An emulsion system of the plant is designed to provide emulsion for a machining process that can be carried out using the plant. For example, an emulsion system is designed to apply rolling emulsion to a metallic workpiece conveyed by the plant and / or to a work roll, thus making it available for a rolling process fed by the plant.

[0086] Adding a subsystem can preferably be achieved by activating a previously inactive subsystem or page 21 / 74

[0087] P81102WO indicates that a previously inactive component of the system becomes active after being switched on. For example, a previously inactive emulsion system can become active after being switched on. For example, a previously inactive pump of an emulsion system can become active after being switched on. An active emulsion system provides emulsion for a machining process. In other words, an active emulsion system provides, for example, rolling emulsion on a metal workpiece to be rolled or on a work roll.

[0088] Switching off a subsystem can preferably be achieved by ensuring that a previously active subsystem or component of the plant is no longer active after shutdown. Finally, switching off a subsystem can preferably be achieved by switching an active subsystem in its operation. For example, an emulsion system can be switched with respect to the quantity of emulsion it provides, thus increasing or decreasing the quantity of emulsion supplied.

[0089] An operating parameter of a metal production and / or metalworking plant can be the positioning of a mechanical plant component, which is variable in terms of its location and setting. For example, a guide table or roller can be changed in position according to the control signal. For instance, the position of a guide table can be adjusted to change the vertical or lateral guidance of a metallic workpiece. The position of a metallic workpiece within the plant can also be monitored, particularly spatially and / or temporally, and the positions of plant components adjusted according to the control unit.

[0090] According to a preferred implementation form, an operating parameter of a plant for metal production and / or metal processing can be page 22 / 74

[0091] P81102WÖ includes a specific parameter regarding one or more than one facility selected from the following list:

[0092] A roller, a shaft, a coupling, a bearing, a roller, a motor shaft, an oil rotary feed, a joint block, a spray bar, a mechanical adjusting unit, a blow-off device, a traversing and / or adjusting valve, a pressure vessel, a storage container, in particular an oil tank, an emulsion tank and / or a pickling tank, a hydraulic adjusting unit, a pneumatic adjusting unit, a mechanical gate, in particular so-called looper gates, a drawing straightening machine, a straightening machine, a gearbox, in particular a main drive, a comb rolling gearbox, an adjusting gearbox, a switching gearbox, preferably a switching gearbox of a reel, a converter gearbox, a flange gearbox, a slab upsetting press, a tool slide, a pendulum lifting device, a roller guide, a side guide, a roller side guide, a bending and sliding system, a spindle bearing,a spindle lubrication device, in particular a rotary oil distributor, a descaling device, a nozzle, a hose, a spray bar, a water return, a sintering trough, a stand foot roller device, a wedge displacement, a slide rail, a displacement device, an inlet / outlet product, a water tank, an oscillation device, a shrinkage damping device, a spring, a geared motor, a skid, rail, and / or rail roller, a burner device, a laminar device, a flap valve, a guide table, a threading device, a spreading device, a spreading device, a spray device, a punching device of a chain drive, a shearing and cutting device, a drying device, a dusting device, a pickling tank product, a heater, a cleaning and / or rinsing device, a coiling device,in particular a reel spool, page 23 / 74,

[0093] P81102WQ

[0094] Welding device, cutting device, anti-coil-break device, strip guiding device, start-up and entry device, rotary and lifting device, in particular a lifting beam, traversing device, in particular a coil carriage, tying device, coil opening devices, crane and lifting device, transverse transport device, preferably an automatic, manual and / or semi-automatic transverse transport device, hydrostatic device, shifting device, in particular a roll changing platform, a roll changing carriage, a locking device, a clamping device, in particular a spindle head holder, a threading device, a crimping device, a positioning device, in particular a coil tail end positioning device, a strip tension device, in particular a looper, a rocker arm pinch roll, a strip tension measuring roller, a cross-cutting shear,a pendulum shear, a drum shear, a continuous shear, a longitudinal slitting shear, a sample shear, a trimming shear, a clipping shear, a punching device, a cutter head device, a knife changing device, a knife changing carriage, a scrap bucket device, a scrap chute, a scrap chute device, a hold-down device, a spreading and unspreading cylinder, a filter device, in particular an emulsion filter, a hydraulic filter and / or a plate filter, a paper feed device, a paper unwinding and / or rewinding device, a hydraulic power unit, a pump station, a heating station, a compressed air station, a high-pressure pump station, a lubricant station, in particular for oils and / or greases, a water treatment station, a scale treatment station, a water cooling device, a lubricant cooling device, a conveyor belt device, a drum drying station,A distribution device, in particular a distribution gearbox for a blast furnace, a strip guiding station and / or a tool grinding device, in particular for roll and / or roller grinding. Page 24 / 74

[0095] P81102WO

[0096] The operation of a plant can comprise one or more than one operating state. An operating state is preferably defined by a plurality of operating parameters. In a plant for metal production and / or metal processing, such an operating state can comprise a production state, a maintenance state, a standby state, and / or a fault state. A production state, for example, can be an operating state in which the plant produces and / or processes a metallic workpiece according to its specific purpose. A maintenance state can be an operating state in which scheduled and / or unscheduled maintenance and / or repair work is carried out on the plant. A production state and a maintenance state do not usually occur simultaneously. A standby state can be an operating state in which the plant is neither in production nor in maintenance mode.A metal production and / or metalworking plant can be switched directly from standby mode to production mode and / or maintenance mode. A fault mode can be an operating state in which the plant is not in one of the aforementioned regular operating states. The plant can be switched to fault mode manually by an operator. Alternatively or additionally, the plant can be switched to fault mode automatically, for example, by the control unit influencing the plant's operation. In a plant with a rolling mill, the plant can pressure-form a metallic workpiece in production mode, preferably by rolling, and thus achieve a specific thickness of the metallic workpiece. Furthermore, in production mode, a metallic workpiece can be threaded into a rolling mill.During maintenance of such a plant, a roller of a rolling mill stand can be replaced. Such a plant can be put back into service immediately after maintenance and before production begins (page 25 / 74).

[0097] P81102WO

[0098] Standby mode. In other words, no maintenance work is carried out in standby mode, nor is a metallic workpiece produced and / or processed, particularly rolled. In a fault operation of such a system, at least one operating parameter may lie outside a permissible range for that parameter. For example, a pressure sensor in the system may measure an impermissibly high pressure within a hydraulic system, such that the system's operating parameter associated with the measured pressure lies outside a permissible range for that pressure. The system may then report a fault and be capable of being transferred to a fault operation and / or enter a fault operation. From a fault operation, the system can only be transferred to production operation, maintenance operation, or standby mode after the reported fault has been rectified.

[0099] A plant state for metal production and / or metalworking can comprise a normal state, a transitional state, and / or a critical state. A plant state for metal production and / or metalworking preferably exists independently of the plant's operating state. Thus, while the plant for metal production and / or metalworking is in production mode, it can simultaneously be in a normal state. Furthermore, the plant for metal production and / or metalworking can be in maintenance or standby mode and simultaneously in a normal state. It is also conceivable that the plant for metal production and / or metalworking is in fault mode and simultaneously in a normal state.

[0100] In a normal operating condition, the acoustic signal received by the sound sensor and converted into an audio signal does not deviate from the reference signal, or deviates within a previously defined normal range. Furthermore, in a normal operating condition, see page 26 / 74.

[0101] P81102WO

[0102] The system does not deviate from the reference signal, or does so within a predetermined normal range, from the optical signal received by the optical sensor and converted into an electrical signal. In other words, the data processing device is preferably configured to determine a system state as the system's normal state, provided that the data records do not deviate from each other, or deviate within a predetermined normal range. The predetermined normal range can be a value of the deviation between a characteristic feature of an acoustic signal received by the sound sensor and converted into an audio signal, and the reference signal, and / or a value of the deviation between a characteristic feature of an optical signal received by the optical sensor and converted into an electrical signal, and the reference signal.For example, the previously determined normal measure can be a value for a deviation between a sound pressure level of the acoustic signal received by the sound sensor and converted into an audio signal and the reference signal, and / or a deviation between an intensity of the optical signal received by the optical sensor and converted into an electrical signal and the reference signal.

[0103] Accordingly, the normal state of the plant describes a state that is normal for the respective operating state of the plant. This is based on the fact that one or more reference plant states for the metal production and / or metalworking plant are stored in one or more reference signals for each operating state of the plant. Consequently, the metal production and / or metalworking plant can be in a normal state during a fault operation, provided that the acoustic signal detected by the sound sensor and converted into an audio signal corresponds to the reference signal assigned to this fault operation, and / or the optical signal detected by the optical sensor and converted into an electrical signal corresponds to the reference signal assigned to this fault operation. Page 27 / 74

[0104] The reference signal assigned to P81102WO does not deviate or deviates within the previously determined normal range.

[0105] In a transitional state of the system, the acoustic signal received by the sound sensor and converted into an audio signal, and / or the optical signal received by the optical sensor and converted into an electrical signal, deviates from the reference signal within a previously defined transition range. In other words, the data processing device is preferably configured to determine a system state as a transitional state if the data records deviate from each other within a previously defined transition range. The control device can be configured to indirectly influence the operation of the system when a transitional state exists. According to an advantageous embodiment, the control device can be configured to output a warning signal, preferably by means of a warning device, when a transitional state exists and when the system is in production operation.The previously determined transition value can be a value representing the deviation between a characteristic feature of an acoustic signal received by the sound sensor and converted into an audio signal, and the reference signal, and / or a value representing the deviation between a characteristic feature of an optical signal received by the optical sensor and converted into an electrical signal, and the reference signal. For example, the previously determined transition value can be a value representing the deviation between the sound pressure level of the acoustic signal received by the sound sensor and the reference signal, and / or a value representing the deviation between the intensity of the optical signal received by the optical sensor and the reference signal.

[0106] In a critical state of the system, the acoustic signal received by the sound sensor and converted into an audio signal deviates (page 28 / 74).

[0107] P81102WO

[0108] The signal and / or the optical signal received by the optical sensor and converted into an electrical signal deviates from the reference signal by more than a predetermined limit. In other words, the data processing device is preferably configured to determine a plant state as a critical state if the data records deviate from each other by more than a predetermined limit. The control device can be configured to directly influence the operation of the plant when a critical state exists. According to an advantageous embodiment, the control device can be configured to perform an emergency shutdown of the plant when a critical state exists and the plant is in production operation.

[0109] The previously determined limit value can be a value representing a deviation between a characteristic feature of an acoustic signal received by the sound sensor and converted into an audio signal, and the reference signal, and / or a value representing a deviation between a characteristic feature of an optical signal received by the optical sensor and converted into an electrical signal, and the reference signal. For example, the previously determined limit value can be a value representing a deviation between the sound pressure level of the acoustic signal received by the sound sensor and the reference signal, and / or a value representing a deviation between the intensity of the optical signal received by the optical sensor and the reference signal.

[0110] According to one implementation form, a plant condition of the metal production and / or metal processing plant can comprise one or more than one plant event. A plant event can preferably occur independently of an operating condition of the plant. Furthermore, a plant event can occur independently of a normal operating condition. Page 29 / 74

[0111] P81102WO indicates that the system is in a transitional state and / or a critical state.

[0112] A plant event is preferably assigned a reference signal specific to this plant event, characterizing the plant event, wherein the reference signal assigned to the plant event may comprise an acoustic signal converted into an audio signal and / or an optical signal converted into an electrical signal.

[0113] According to one implementation, a plant event can include the threading process of a metallic workpiece into a rolling stand of the plant. During the threading process of a metallic workpiece into a rolling stand, a characteristic noise is expected as soon as the head of the metallic workpiece comes into direct physical contact with the working rolls of the rolling stand. If such a noise is not detected by the sound sensor, or if a noise deviates from the expected noise is detected, then the plant state deviates from the expected state and / or a reference state. Depending on the degree of deviation, the system can be configured to influence the operation of the plant indirectly or directly via the control device, preferably by initiating an emergency stop of the plant.This can prevent, in particular, the metallic workpiece from being incorrectly or not at all threaded onto the rolling mill stand, and thus from accumulating in front of the rolling mill stand, ultimately leading to rejects and potentially damage to the system.

[0114] A tracking function can be performed by monitoring a system event. The presence, absence, or termination of an expected acoustic signal can indicate the position of a metallic workpiece and be transmitted as information. Page 30 / 74

[0115] P81102WO

[0116] According to one implementation form, a system event can include the detection of a strip break and the localization of the origin of a strip break in a metallic workpiece conveyed by the system. This makes it possible to initiate an emergency stop of the system to avoid excessive scrap. Furthermore, one or more operating parameters can also be changed.

[0117] Furthermore, a system event can include the detection of a subsystem being switched on and / or off. The switching on and / or off of a subsystem is preferably accompanied by a characteristic acoustic and / or visual signal. For example, the switching on and / or off of a subsystem can include the switching on and / or off of a compressed air system and / or an emulsion system, in particular an emulsion dispensing system and / or an extraction system and / or a conveying device.

[0118] Finally, a plant event can be the detection of an unexpected noise and / or an unexpected visual event. In other words, a plant event can involve the detection of an unexpected acoustic and / or visual signal. Such an unexpected acoustic and / or visual signal could, for example, be caused by existing and / or impending wear of a roller in a rolling mill. Furthermore, such an unexpected acoustic and / or visual signal could be caused by bearing damage in a drive motor, such as the drive motor of a work roller.

[0119] According to a preferred variant, the control device can be configured to, based on the plant state determined by the data processing device and based on the operating state of the metal production and / or Me- plant, page 31 / 74

[0120] P81102WO machining at least indirectly, preferably directly, to influence the operation of the plant.

[0121] The system may include a facility for metal production and / or metal processing.

[0122] Preferably, the system is designed such that the acoustic sensor is arranged at a distance from the metal production and / or metalworking plant. Furthermore, the optical sensor can be arranged at a distance from the metal production and / or metalworking plant.

[0123] A system designed in this way has the advantage that it can be integrated into existing metal production and / or metalworking plants with even less effort. This further reduces investment costs for such a system. Advantageously, modifications to the plant itself can be avoided, thus enabling simple retrofitting.

[0124] The acoustic sensor and / or the optical sensor can be arranged at a distance from the system in at least one spatial direction, preferably in two or three spatial directions. The acoustic sensor and / or the optical sensor can be arranged above or below the metal production and / or metalworking system. According to one embodiment, the metal production and / or metalworking system can be located in a metal production and / or metalworking plant, preferably in a workshop of the plant. The acoustic sensor and / or the optical sensor can be arranged on the workshop, in particular on the roof of the workshop, preferably on a side of the roof facing the inside of the workshop, on a wall of the workshop, and / or on a supporting structure of the workshop. This allows the [page 32 / 74]

[0125] P81102WO

[0126] The system can be easily integrated into existing metal production and / or metalworking plants.

[0127] The sound sensor can further be arranged relative to the metal production and / or metalworking plant in such a way that the sound sensor is configured to receive the acoustic signal, wherein the acoustic signal preferably contains direct sound emanating from the plant and / or a section of the plant, and in particular contains exclusively direct sound. In other words, the sound sensor is preferably arranged relative to the plant in such a way that there is a direct line of sight between a sound source located on the plant and the sound sensor. The optical sensor can also be arranged relative to the plant in such a way that there is a direct line of sight between the plant and the optical sensor.

[0128] The acoustic sensor, or another acoustic sensor, and / or the optical sensor can be arranged on the system. For example, the acoustic sensor can be arranged on a section of the system and configured to receive an acoustic signal emanating from the same section and / or another section of the system. A system configured in this way has the advantage that the acoustic sensor receives a reduced amount of interfering signals, such as acoustic signals emanating from other systems, so that the system can receive the acoustic signal emanating from the system with increased accuracy and thus determine the system's state with greater accuracy. Similarly, the optical sensor can be arranged on a section of the system and configured to receive an optical signal emanating from the same section and / or another section of the system. Page 33 / 74

[0129] P81102WO

[0130] The acoustic sensor and / or the optical sensor can be connected, at least indirectly, to the metal production and / or metalworking plant, preferably directly. In particular, the acoustic sensor and / or the optical sensor can be physically connected, directly or indirectly, to the metal production and / or metalworking plant, for example, by positive locking, material locking, and / or force locking. A system designed in this way has the advantage that the acoustic sensor receives a further reduced amount of interference signals, so that the system can receive the acoustic signal emanating from the plant with even greater accuracy and thus determine the plant's state with even greater accuracy.

[0131] Preferably the system is designed such that the sound sensor and / or the optical sensor is / are arranged in a positionally variable or positionally fixed manner relative to the metal production and / or metalworking plant.

[0132] A sound sensor arranged in a position fixed relative to the metal production and / or metalworking plant preferably has a position that is constant with respect to the plant, and in particular, constant over time. In other words, a sound sensor arranged in a position fixed relative to the plant is fixed in relation to the plant. A system designed in this way has the advantage that it can be installed with reduced effort. Furthermore, due to the sound sensor's constant position relative to the plant, it can receive an acoustic signal emanating from the plant with increased robustness, in particular with increased repeatability.

[0133] A sound sensor arranged in a positionally variable manner relative to the plant for metal production and / or metal processing preferably has a position that is variable relative to the plant, in particular page 34 / 74

[0134] P81102WO time-varying position. In other words, a sound sensor arranged so that its position can change relative to the system is movable. Preferably, the sound sensor is mounted so that it can move relative to the system by means of an actuator. The system can include an actuator, wherein the sound sensor is at least indirectly, and preferably directly, physically connected to the actuator. The actuator can be configured to change the position of the sound sensor relative to the system. An actuator can include a crane, in particular an overhead crane. An actuator can include a manipulator, in particular a robot arm. A system designed in this way has the advantage that the sound sensor can be moved to a section of the system so that, for example, depending on the operating state of the system, a position of the sound sensor relative to the system that is favorable for this operating state is achieved.This increases the accuracy with which the sound sensor can receive an acoustic signal emanating from the system and reduces any interference signals, thus increasing the accuracy of determining the system's condition.

[0135] The data processing device can be configured to determine the position of the sound sensor with respect to one or more reference points. The reference point can be a previously determined point on the metal production and / or metalworking plant. According to another embodiment, the reference point can be a previously determined point in a factory building, preferably in a factory building in which the metal production and / or metalworking plant is located. The reference point can be the origin of a reference coordinate system. The data processing device can be configured to determine the position of the sound sensor with respect to the origin of the reference coordinate system. The data processing device can be configured to determine any [page 35 / 74]

[0136] P81102WO

[0137] to perform coordinate transformation and determine the position of the object with respect to this reference coordinate system or any other arbitrary coordinate system.

[0138] The data processing device can be configured to determine the position of the sound sensor relative to the metal production and / or metalworking plant. Preferably, the data processing device is configured to repeatedly, and in particular continuously, determine the position of the sound sensor relative to the plant. This makes it possible to determine and, in particular, to track changes in the position of the sound sensor relative to the plant over time, so that the current position of the sound sensor relative to the plant is always known.

[0139] Preferably the system is designed such that the sound sensor is designed as a microphone, preferably as an electrostatic microphone or as a piezoelectric microphone or as a dynamic microphone.

[0140] A system designed in this way has the advantage that, due to the use of a microphone as a standard component, the system can be manufactured and, in particular, operated at reduced costs.

[0141] A microphone is an electroacoustic transducer that converts mechanical vibrations in a gas, especially air, into an audio signal. A microphone can therefore also be called a gas transducer, and more specifically, an air transducer. According to a typical microphone design, an elastically mounted diaphragm can follow the pressure fluctuations of the acoustic signal caused by the mechanical vibration and, through its movement, replicate the temporal distribution of the alternating pressure. A transducer that is mechanically and / or electrically connected to page 36 / 74

[0142] The P81102WO, which is coupled to the diaphragm, can generate an audio signal from the diaphragm movement.

[0143] The microphone can be designed as a dynamic microphone. A sound sensor designed in this way has the advantage of increased robustness and a reduced amount of background noise, so that a system with such a sound sensor exhibits overall increased robustness.

[0144] The microphone can be designed as an electrostatic microphone. A sound sensor designed in this way has the advantage of increased sensitivity, enabling it to receive even subtle acoustic signals, thus resulting in an overall increased sensitivity for a system with such a sound sensor.

[0145] The microphone can be designed as a piezoelectric microphone. Such a sound sensor has the advantage of increased mechanical robustness and comparatively low manufacturing costs, so that a system with such a sound sensor offers overall increased mechanical robustness and low manufacturing costs.

[0146] According to a preferred embodiment, the sound sensor can be designed as a pickup. A pickup is an electroacoustic transducer that converts mechanical vibrations in solids into an audio signal. A pickup can therefore also be called a structure-borne sound sensor. This allows an acoustic signal emanating from the system to be received with significantly increased accuracy and, in particular, with reduced interference, especially by means of a pickup physically connected to the system. Page 37 / 74

[0147] P81102WO

[0148] According to a preferred embodiment, the sound sensor can be configured as a hydrophone. A hydrophone is an electroacoustic transducer that converts mechanical vibrations in liquids into an audio signal. A hydrophone can therefore also be referred to as a liquid sound sensor, and in particular as an underwater sound sensor.

[0149] Preferably, the system is designed such that the sound sensor has an electret.

[0150] A sound sensor designed in this way has the advantage of being even more robust while simultaneously having even lower manufacturing costs, so that a system with such a sound sensor has overall increased robustness with simultaneously lower manufacturing costs.

[0151] The electret is preferably designed as an electret foil. An electret is an electrically insulating material that has quasi-permanently stored electric charge or quasi-permanently aligned electric dipoles and can thus generate a quasi-permanent electric field in its surroundings or within its interior.

[0152] Preferably the system is designed such that the sound sensor is configured to receive acoustic signals with a sound frequency of < 2 GHz, preferably < 1.6 GHz and particularly preferably < 20 kHz.

[0153] A system designed in this way has the advantage that it can receive acoustic signals of different sound frequencies, thus enabling the determination of a system's condition with increased accuracy. If an acoustic signal is generated by the system and / or a component of the system as a sound source, this signal can contain a plurality of sound frequencies. Page 38 / 74

[0154] The P81102WO sequences are included. Information about different components of the plant can be extracted from different sound frequencies. Because the sound sensor can receive acoustic signals with a plurality of sound frequencies, the data processing unit has more information about the plant available from the different sound frequencies.

[0155] The sound sensor can be configured to receive acoustic signals with a sound frequency of < 1.8 GHz, preferably < 1.3 GHz, more preferably < 1 GHz, and most preferably < 500 MHz. The sound sensor can be configured to receive acoustic signals with a sound frequency of < 250 MHz, more preferably < 100 MHz, more preferably < 1 MHz, and most preferably < 500 kHz. The sound sensor can be configured to receive acoustic signals with a sound frequency of < 250 kHz, more preferably < 100 kHz, more preferably < 50 kHz, and most preferably < 5 kHz.

[0156] Preferably the system is designed such that the sound sensor is configured to receive acoustic signals with a sound frequency of > 1 Hz, preferably of > 15 Hz and particularly preferably of > 10 kHz.

[0157] A system designed in this way has the advantage that acoustic signals of different sound frequencies can be received, so that the system status can be determined with increased accuracy.

[0158] The sound sensor can be configured to receive acoustic signals with a sound frequency of > 500 Hz, preferably > 1 kHz, more preferably > 50 kHz, and particularly preferably > 150 kHz. The sound sensor can be configured to receive acoustic signals with a sound frequency of > 300 kHz, preferably > 600 kHz, more preferably > 1 MHz, and particularly preferably > 300 MHz. The sound sensor can be configured to receive acoustic signals with a sound frequency of > 300 kHz, preferably > 600 kHz, more preferably > 1 MHz, and particularly preferably > 300 MHz. (See page 39 / 74.)

[0159] P81102WO is configured to receive acoustic signals with a sound frequency of > 600 MHz, preferably > 800 MHz, preferably > 1.1 GHz and particularly preferably > 1.4 GHz.

[0160] Preferably, the system is designed such that the data processing device is configured to perform a time-frequency analysis of the acoustic signal received by the sound sensor, preferably before a system state of the metal production and / or metalworking plant is determined, and to store a result of the time-frequency analysis in the second data set.

[0161] A system designed in this way has the advantage that deviations between the signals stored in the data sets can be determined with increased accuracy in the frequency spectrum, so that the system condition can be determined with increased accuracy.

[0162] A time-frequency analysis can include an integral transformation, in particular a Fourier transformation, preferably a continuous Fourier transformation or a discrete Fourier transformation. In other words, the acoustic signal and / or audio signal received by the sound sensor can be decomposed into a frequency spectrum using the data processing device and stored in the second data set.

[0163] The reference signal stored in the first data set can be decomposed into its frequency spectrum and stored in the first data set.

[0164] Preferably, the system is designed such that the data processing device is configured to extract one or more than one signal component with a sound frequency of < 1 Hz from the acoustic signal received by the sound sensor (page 40 / 74).

[0165] Filter P81102WO, preferably before determining the condition of the metal production and / or metalworking plant.

[0166] A system designed in this way has the advantage that signal components which do not contain information relevant for a comparison of the two data sets can be filtered out of the signals, so that a system condition can be determined with increased accuracy.

[0167] The data processing device may include a filtering device or be data-connected to one. The filtering device may include a low-pass filter, a high-pass filter, and / or a band-pass filter.

[0168] The data processing device can be configured to filter one or more than one signal component with a sound frequency of < 100 Hz from the acoustic signal received by the sound sensor, preferably with a sound frequency of < 500 Hz, preferably with a sound frequency of < 1 kHz and particularly preferably with a sound frequency of < 10 kHz.

[0169] Preferably, the system is designed such that the data processing device is configured to filter one or more than one signal component with a sound frequency of > 2 GHz from the acoustic signal received by the sound sensor, preferably before a system state of the plant for metal production and / or metal processing is determined.

[0170] A system designed in this way has the advantage that signal components which do not contain information relevant for comparing the two data sets can be filtered out of the signals, thus enabling a more accurate determination of the system's condition. Page 41 / 74

[0171] P81102WO The data processing device can be configured to filter one or more than one signal component with a sound frequency of > 100 kHz from the acoustic signal received by the sound sensor, preferably with a sound frequency of > 500 kHz, preferably with a sound frequency of > 1 MHz and particularly preferably with a sound frequency of > 100 MHz.

[0172] Preferably the system is designed such that the sound sensor has a substitute noise level of < 30 dB, preferably < 25 dB, preferably < 15 dB and particularly preferably < 10 dB.

[0173] A system designed in this way has the advantage that even slight deviations of a plant's condition from a reference condition can be detected with increased accuracy. Furthermore, the acoustic sensor can be positioned at a greater distance from the metal production and / or metalworking plant. The lower the equivalent noise level of an acoustic sensor, the lower the background noise in an audio signal converted by this sensor.

[0174] The sound sensor can have a substitute noise level of < 50 dB, preferably < 12 dB, preferably < 8 dB and particularly preferably < 5 dB.

[0175] The equivalent noise level of a sound sensor, especially a microphone, is a measure of the sensor's inherent noise. The lower the equivalent noise level, the lower the inherent noise. For example, an equivalent noise level of 15 dB means that the inherent noise of the sound sensor, especially the microphone, is as loud as a sound with a sound pressure level of 15 dB. Page 42 / 74

[0176] P81102WO

[0177] The equivalent noise level can generally be determined using two different measurement methods. A so-called A-weighting according to the DIN IEC 651 standard typically results in equivalent noise level values ​​that are approximately 10 dB lower than those obtained using a more critical measurement method according to the ITU-R 468-3 standard (ITU-R stands for International Telecommunication Union - Radiocommunication Sector). The values ​​given above primarily refer to the equivalent noise level measurement method according to the ITU-R 468-3 standard. Alternatively, the values ​​given above for the equivalent noise level can also refer to the measurement method according to DIN IEC 651.

[0178] Preferably, the system is designed such that the sound sensor is configured to generate an acoustic identification signal, wherein a reflection of the acoustic identification signal generated by the sound sensor at the metal production and / or metalworking plant is the acoustic signal received by the sound sensor.

[0179] A system designed in this way has the advantage that the acoustic detection signal, particularly its amplitude and frequency, is known. Consequently, a reflection of the acoustic detection signal is also known. This allows an acoustic signal received by the sound sensor to be identified with increased accuracy as a reflection of the acoustic detection signal.

[0180] The sound sensor can therefore also include a sound source and / or be designed as such. For this purpose, the sound sensor can be configured to convert an electrical signal, preferably an audio signal, into an acoustic signal. The sound sensor can include a loudspeaker and / or be designed as such. Page 43 / 74

[0181] P81102WO

[0182] The acoustic identification signal is preferably an acoustic signal. In particular, it comprises one or more than one sound wave.

[0183] The sound sensor can be configured to generate an acoustic detection signal more than once, in particular regularly, preferably at a predetermined frequency. With a system designed in this way, changes in the state of a system over time can be determined with increased accuracy.

[0184] Preferably, the system is designed such that the control device is configured to indirectly influence the operation of the system by issuing a warning signal, wherein the warning signal comprises an optical signal and / or an acoustic signal and / or a haptic signal.

[0185] A system designed in this way has the advantage that the system's status can be checked again before the control unit intervenes in its operation. This prevents the system from being affected by an incorrectly determined status. Unnecessary downtime and thus unnecessary production losses can therefore be avoided.

[0186] The indicator signal can be designed as a warning signal, advantageously being associated with a transition state and / or a critical state of the system. This allows the operator to be notified of the occurrence of a transition state and / or a critical state of the system. The warning signal can also indicate that a transition state and / or a critical state of the system is being reached. Consequently, the warning signal can be triggered even before the system enters a transition state and / or a critical state. (Page 44 / 74)

[0187] P81102WO is given so that a plant operator can potentially intervene in the operation of the plant before the plant enters, for example by changing one or more than one operating parameter of the plant.

[0188] The indicator signal can include information about the condition of a metallic workpiece being processed by the metal production and / or metalworking plant. Information about the condition of a metallic workpiece can, in particular, include information about product quality. In other words, information about the condition of the metallic workpiece contains information about the presence of defects and / or flaws in the metallic workpiece. For example, an indicator signal can be designed as an optical indicator signal, where the optical indicator signal represents a portion of the surface of a metallic workpiece being processed by the plant, and where flaws on the surface of the metallic workpiece are indicated.This makes it possible, for example, to provide a plant operator with information about current product quality via the warning signal, so that the plant operator can influence the operation of the plant.

[0189] Preferably, the system is designed such that the system has a notification device connected to the control unit, preferably an operator interface, wherein the control unit is configured to output the notification signal by means of the notification device.

[0190] A system designed in this way has the advantage that a recheck of the system's condition can be carried out regardless of the proximity of a system operator to the system.

[0191] The warning device can be part of the control device. In other words, the control device can display the warning message. Page 45 / 74

[0192] The P81102WO device can be equipped with a warning device. The warning device can be wirelessly connected to the control unit.

[0193] The warning device can have one or more warning devices. A warning device can be, for example, a loudspeaker, such as a siren. A warning device can also be a screen, such as an operator's screen.

[0194] Finally, the notification device may include an end device, for example a mobile device, preferably a mobile phone, a smartphone and / or a tablet.

[0195] The control unit is preferably wirelessly connected to the indicator device. Alternatively or additionally, the control unit is connected to the indicator device via a wire, preferably a cable.

[0196] The visual signal can be, for example, a visual indication on a screen, preferably an operator screen. According to another embodiment, the visual signal can be a light signal, for example, a light signal emitted by means of a lighting device. The acoustic signal can be, for example, a signal tone, in particular a signal tone emitted by means of a loudspeaker or a siren. The haptic signal can be a vibration, in particular a vibration of a mobile device.

[0197] Preferably, the system is designed such that the control unit is configured to directly influence the operation of the plant based on the plant state determined by the data processing unit, preferably by changing at least one operating parameter of the plant for metal production and / or metal processing. Page 46 / 74

[0198] P81102WO

[0199] A system designed in this way has the advantage of being able to react more quickly to an undesirable system condition. For example, in the event of an existing or imminent malfunction, the control unit can immediately intervene in the operation of the system and thus prevent further damage to the system or further damage, in particular quality defects, to a product produced and / or processed by the system, especially a metallic workpiece.

[0200] An operating parameter of a system can be, for example, a power supply, preferably an electrical power supply. For example, the operation of a system can be influenced by changing the electrical power supply, preferably by interrupting the electrical power supply.

[0201] Furthermore, an operating parameter can be, for example, the rolling speed of a rolling mill. An operating parameter can also be the contact pressure of a roller in a rolling mill. An operating parameter can, for example, include the conveying speed of a metallic workpiece in a metal production and / or metalworking plant.

[0202] An operating parameter can further include a roll gap and / or a strip tension. A roll gap is formed between two working rolls of a rolling mill in a metal production and / or metalworking plant. A roll gap therefore defines the extent between two working rolls perpendicular to the conveying direction of a metallic workpiece conveyed by a metal production and / or metalworking plant.

[0203] A belt tension of a plant for metal production and / or metal processing is preferably a tensile force applied by the plant to a metallic workpiece located in the plant, page 47 / 74

[0204] P81102WO preferably on a metal belt, wherein the applied tensile force is directed in the direction or against the direction of a conveying direction of the metallic workpiece.

[0205] According to one implementation form, the control device can be configured to directly influence the operation of the plant based on the plant state determined by the data processing device by changing, in particular increasing or decreasing, a rolling speed, and / or a conveying speed, and / or a contact pressure, and / or a rolling gap, and / or a strip tension of the plant for metal production and / or metalworking.

[0206] The control unit can be configured to directly influence the operation of the plant by changing a number of operating parameters of the metal production and / or metal processing plant, based on the plant state determined by the data processing unit.

[0207] The control unit can be configured to directly influence the operation of the system by changing at least one operating parameter, depending on the system's operating state. According to a first variant, the system can be in production mode and simultaneously a critical state can be determined by the data processing unit. In this case, the control unit can be configured to perform an emergency shutdown of the system. According to a second variant, the system can be in standby mode and simultaneously a critical state can be determined by the data processing unit. In this case, the control unit can be configured to prevent the system from starting production mode, in particular until the system is no longer in a critical state. Page 48 / 74

[0208] P81102WO

[0209] Preferably, the system is designed such that comparing the two data sets includes comparing at least one amplitude of the reference signal with at least one amplitude of the received acoustic signal.

[0210] A system designed in this way has the advantage that even differentiated changes in the condition of the plant for metal production and / or metal processing can be detected.

[0211] The received acoustic signal can be converted into an audio signal before comparison with the reference signal. Preferably, the data processing device is configured to determine the operating state of the metal production and / or metalworking plant based on an amplitude deviation of the acoustic signal compared to the reference signal. An amplitude deviation of the acoustic signal can be a ratio of the amplitude of the acoustic signal, preferably the acoustic signal converted into an audio signal, to the amplitude of the reference signal. An amplitude deviation can have a value greater than or equal to 1, preferably greater than or equal to 1.01, more preferably greater than or equal to 1.1. An amplitude deviation can have a value less than 1, preferably less than or equal to 0.95, more preferably less than or equal to 0.9.

[0212] Preferably, the system is designed such that the comparison of the two data sets includes a comparison of one or more than one amplitude of a previously determined frequency band of the reference signal with one or more than one amplitude of the same frequency band of the received acoustic signal.

[0213] Such a system has the advantage that even subtle changes in the condition of the metal production and / or metalworking plant can be detected. Page 49 / 74

[0214] P81102WO

[0215] The frequency band can be a fixed frequency band. For example, the frequency band can comprise a frequency range of greater than or equal to 1 Hz and less than or equal to 2 GHz, preferably a frequency range of greater than or equal to 15 Hz and less than or equal to 100 kHz, preferably of greater than or equal to 25 Hz and less than or equal to 50 kHz.

[0216] The system may include additional sensors, such as pressure sensors, temperature sensors, and / or accelerometers. Alternatively, the data processing unit may be data-connected to additional sensors, such as pressure sensors, temperature sensors, and / or accelerometers. The system may also include one or more optical sensors data-connected to the data processing unit. An optical sensor may include an infrared sensor, an ultraviolet sensor (UV sensor), a laser sensor, an LED-based sensor, and / or a hyperspectral sensor. Furthermore, an optical sensor may be configured according to any measurement principle.In particular, an optical sensor may include a reflection sensor and / or a transmitted light sensor and / or a fork light barrier sensor and / or a triangulation sensor and / or a phase comparison sensor and / or an interferometric sensor and / or an imaging sensor.

[0217] The optical sensor can be particularly advantageous if it includes an imaging sensor, especially a camera. In particular, if the data processing device is configured to determine the plant condition using a machine learning method, an optical sensor including an imaging sensor is advantageous because the available information density in an image and / or video of a plant and / or a section of a plant provided by an imaging sensor is significantly higher compared to other optical sensors. (Page 50 / 74)

[0218] P81102WO has been increased so that the plant condition can be determined with even greater accuracy using a machine learning method.

[0219] The data processing device can be wirelessly connected to the optical sensor. According to a preferred embodiment, the data processing device can be connected to the optical sensor by means of a wire, in particular a cable.

[0220] The data processing unit can be configured to store signals received from additional sensors in a third data set. The data processing unit can also be configured to determine a system status taking this third data set into account.

[0221] According to an advantageous embodiment, the system comprises an optical sensor connected to the data processing unit, which is configured to receive an optical signal, preferably an optical signal emanating from the metal production and / or metalworking plant. The data processing unit is further configured to determine the state of the metal production and / or metalworking plant by taking into account a third data set, wherein the third data set comprises the optical signal received from the optical sensor. With a system configured in this way, a plant state can be determined based on information obtained by means of at least two different sensor types.Because the sensor types operate independently and thus provide independent sources of information, the determination of the system's condition can be made more robust against measurement errors. Page 51 / 74.

[0222] P81102WO

[0223] The data processing device is particularly advantageously configured to compare at least three data sets, wherein a first data set contains a reference signal, a second data set contains the acoustic signal received by the sound sensor, and a third data set contains the optical signal received by the optical sensor; and to determine a system state of the metal production and / or metalworking plant based on the comparison of the data sets.

[0224] An optical signal is preferably the temporal profile of a signal detectable by an optical sensor. Depending on the optical sensor, the optical signal can have different characteristics. In the case of an imaging optical sensor, particularly a camera, the optical signal can be the temporal profile of the light intensities incident on a sensor of the camera, especially a CMOS sensor and / or a CCD sensor. In the case of a reflection sensor, for example, a reflection sensor operating by means of laser radiation, the optical signal can be the temporal profile of a reflection detectable by the optical sensor of a measurement signal previously emitted by the optical sensor, whereby, for example, a temporal profile of the time differences is determined.

[0225] An optical sensor is preferably configured to convert an optical signal into an electrical signal representing the optical signal, which in particular comprises an alternating current and / or an alternating voltage. The electrical signal representing the optical signal preferably contains one or more characteristic features of the optical signal. A characteristic feature of an optical signal is preferably an optical power and / or an intensity and / or a wavelength and / or a frequency and / or a spectral bandwidth. Page 52 / 74

[0226] P81102WO

[0227] Preferably, the system is designed such that the data processing device is configured to determine the plant status of the metal production and / or metal processing plant using a machine learning method.

[0228] A system designed in this way has the advantage that the condition of the plant can be determined with increasing accuracy as the plant continues to operate.

[0229] The machine learning method can include linear regression and / or logistic regression and / or a k-means algorithm and / or a support vector machine and / or a decision tree-based method, for example a random forest and / or a neural network.

[0230] According to an advantageous embodiment, the system is designed such that the data processing device is configured to determine the plant status of the metal production and / or metalworking plant using an AI / ML model.

[0231] An AI / ML model is preferably a computer-based system designed, based on algorithmic procedures, to transform input data into output data, relying on patterns, relationships, and / or decision structures learned previously through training and / or rules. Such models can utilize machine learning, deep learning, symbolic AI, and / or hybrid methods to perform tasks such as classification, regression, prediction, decision-making, natural language processing, and / or image recognition. A typical AI / ML model usually consists of a defined architecture, for example, neural networks with specific layers and activation functions, and weighting parameters that are typically adjusted during a training process (page 53 / 74).

[0232] P81102WO comprises a training mechanism, such as an optimization method like gradient boosting, and an execution environment, usually hardware and / or software, that the model uses when applied to new data. Typically, an AI / ML model is trained on a dataset, iteratively adjusting its internal parameters to optimize a specific target function or performance measure. After training, the model can analyze new, unknown input data and make a prediction and / or decision based on the learned relationship. The quality of the model's performance depends on factors including the data source, the model architecture, the training method, and the application configuration.

[0233] The system can be expediently trained and continuously improve itself over an ongoing operational period if the AI / ML model is trained with training data, where the training data includes the first and / or second and / or third data set and the plant state determined based on the first and second and / or third data set. With such a trained system, the accuracy of determining a plant state based on signals stored in the data sets can be continuously improved.

[0234] Preferably, the system is designed such that the system has a plurality of sound sensors, in particular two or more, or three or more, or four or more, wherein the data processing device is connected to the sound sensors and wherein the sound sensors are each arranged at a distance from each other.

[0235] A system designed in this way has the advantage that the acoustic signal, preferably the acoustic signal emanating from the system, can be received with improved quality. In particular, several acoustic signals can be received simultaneously with improved quality. (Page 54 / 74)

[0236] P81102WO can be received and used to determine the plant status.

[0237] Each of the sound sensors can be designed like the sound sensor previously described in the first aspect of the invention.

[0238] The data processing device can be wirelessly connected to the sound sensors. Alternatively or additionally, the data processing device can be connected to the sound sensors via a wire, preferably a cable.

[0239] Preferably, at least two, more preferably three, four, or more, or the majority of the sound sensors are connected to the data processing device in a time-synchronized manner. With sound sensors connected in a time-synchronized manner, the data processing device can determine a common start time and a common end time for the acoustic signals received by the sound sensors. In other words, the acoustic signals received by the at least two sound sensors exhibit a temporal deviation from each other of less than or equal to a previously determined maximum temporal deviation value. The maximum temporal deviation value between the acoustic signals received by the sound sensors can be less than or equal to 25 ps, preferably less than or equal to 100 ns, and more preferably less than or equal to 50 ns.

[0240] A time-synchronized connection between multiple sensors can be achieved using a regular trigger signal, for example, from an external reference. Time synchronization can be performed according to the RTF (Precision Time Protocol) as defined in IEEE 1588. Alternatively or additionally, time synchronization can be achieved using GPS. Page 55 / 74

[0241] P81102WO

[0242] Preferably, the system is designed such that the sound sensor and another sound sensor are arranged in a position-changing or position-fixed manner relative to each other.

[0243] Two sound sensors arranged in a position that remains fixed relative to each other preferably have a constant, and in particular a constant, relative position to each other over time. In other words, two sound sensors arranged in a position that remains fixed relative to each other are fixed in position relative to each other.

[0244] Two sound sensors arranged so that their positions can change relative to each other preferably have a variable, and in particular, time-varying relative position to each other. In other words, two sound sensors arranged so that their positions can change relative to each other are movably mounted relative to each other. Preferably, the sound sensors are movably mounted relative to each other by means of one or more actuators. The system can have one actuator, wherein a first sound sensor is at least indirectly, preferably directly, physically connected to a first actuator. A second sound sensor can be at least indirectly, preferably directly, physically connected to the first actuator and / or to a second actuator. The first actuator and / or the second actuator can be configured to change the position of the first sound sensor relative to the position of the second sound sensor.

[0245] According to a preferred embodiment, more than one acoustic sensor can be arranged in a common housing. This allows the system to be designed compactly and integrated into existing metal production and / or metalworking plants with even less effort.

[0246] The data processing device can be configured to determine a relative position between a first sound sensor and a page 56 / 74

[0247] P81102WO to determine a second sound sensor. In other words, the data processing device can be configured to determine a relative position of a first sound sensor with respect to a second sound sensor and / or a relative position of a second sound sensor with respect to a first sound sensor. Preferably, the data processing device is configured to repeatedly, and in particular continuously, determine a relative position between a first sound sensor and a second sound sensor, preferably continuously over time. This makes it possible to determine and, in particular, to track changes in the relative position between a first sound sensor and a second sound sensor over time, so that a current relative position between the first sound sensor and the second sound sensor is always known.

[0248] Preferably, the system is designed such that the sound sensors are equidistant from each other.

[0249] A system designed in this way has the advantage that the acoustic signal, preferably the acoustic signal emanating from the system, can be received with improved quality. In particular, this allows several acoustic signals to be received simultaneously with improved quality and used to determine the system status.

[0250] The sound sensors can be spaced apart from each other in one, two, or three spatial directions. For example, the sound sensors are spaced apart from each other along a line, preferably equidistant from each other. The line can be straight. Alternatively, the line can have one or more curves. For example, the line can comprise a segment of a circular arc, preferably a complete circular arc. The sound sensors can, for example, be arranged spaced apart from each other along such a circular arc, preferably equidistant from each other. Page 57 / 74

[0251] P81102WO

[0252] Preferably, the system is designed such that the sound sensors are arranged in a regular pattern relative to each other, in particular in a rectangular pattern, a square pattern, a circular pattern or an annular pattern.

[0253] A system designed in this way has the advantage that the acoustic signal, preferably the acoustic signal emanating from the system, can be received with improved quality. In particular, this allows several acoustic signals to be received simultaneously with improved quality and used to determine the system status.

[0254] The sound sensors can be arranged in a plane and spaced apart from each other within that plane, particularly in a regular pattern. The sound sensors can be arranged in a cross-shaped pattern or in a spiral pattern.

[0255] The sound sensors can, for example, be arranged along a number of parallel lines. Preferably, the sound sensors arranged in a line, preferably a straight line, are equidistant from each other. Preferably, the parallel lines are equidistant from each other. In other words, the sound sensors can be arranged in the form of a regularly spaced grid.

[0256] Preferably, the system is designed such that at least a first sound sensor has a distance to a second sound sensor that differs from a distance between the first sound sensor and a third sound sensor.

[0257] Such a system has the advantage that, despite the use of a large number of sound sources, the occurrence of page 58 / 74

[0258] P81102WO avoids false sound sources and at the same time an acoustic signal, preferably an acoustic signal emanating from the system, can be received better.

[0259] The sound sensors can be arranged in a quasi-random distribution, especially in a plane.

[0260] According to a second aspect of the invention, the problem underlying the invention is solved by using a system according to the first aspect of the invention to at least indirectly influence the operation of a plant for metal production and / or metalworking.

[0261] Features of the first aspect of the invention can also be combined with the second aspect of the invention, both individually and cumulatively. Advantages realized for the first aspect of the invention accordingly also apply to the second aspect of the invention.

[0262] According to a third aspect of the invention, the problem underlying the invention is solved by a method for determining a plant state, preferably using a system according to the first aspect of the invention, wherein the method comprises the following steps:

[0263] Receiving an acoustic signal emanating from a metal production and / or metalworking plant; comparing at least two data sets, wherein a first data set contains a reference signal and a second data set contains the received acoustic signal;

[0264] Determining the condition of a metal production and / or metalworking plant based on a comparison of data sets; and at least indirectly influencing the operation of the metal production and / or metalworking plant based on page 59 / 74

[0265] P81102WO on the previously determined plant condition of the metal production and / or metalworking plant.

[0266] Such a sophisticated method offers the advantage of detecting even minor deviations of a system's condition from a reference state. This allows for more reliable pre-planning of maintenance measures and the prevention of potential system failures. For example, based on a detected system condition and its associated deviation from a reference state, preventive measures, such as maintenance, can be implemented before the deviation leads to damage. Furthermore, such a method can improve the system's process stability and, consequently, the product quality of any product manufactured or processed using such a system, particularly metallic workpieces.

[0267] Features of the first aspect of the invention can also be combined with the third aspect of the invention, both individually and cumulatively. Advantages realized for the first aspect of the invention accordingly also apply to the third aspect of the invention. Features of the second aspect of the invention can also be combined with the third aspect of the invention, both individually and cumulatively. Advantages realized for the second aspect of the invention accordingly also apply to the third aspect of the invention.

[0268] As a non-limiting example, the invention will be explained in more detail using a first application example. A plant for metal production and / or metalworking can have at least one rolling device and one or more than one cutting device. The cutting device is preferably configured to cut a metallic workpiece, preferably one shown on page 60 / 74.

[0269] P81102WO is used to cut a conveyed metallic workpiece to a predetermined dimension along its width. This process produces metal strips. The system further comprises at least one, preferably more than one, edge-cutting device, which is configured to cut the resulting metal strips into smaller pieces. The system also comprises one or more edge channels, which are configured to transport the metal strips cut into smaller pieces by the edge-cutting device. The edge channel can have one or more conveying devices for transporting the cut metal strips, for example, a conveyor belt.

[0270] Both the cutting of the metallic workpiece to a previously determined dimension, the cutting of the metal strips into smaller pieces, and the transport of the smaller pieces each generate one or more than one characteristic acoustic signal, in particular an acoustic signal pattern, during operation, in particular during production operation of the plant, which are each stored as reference signals in the second data set.

[0271] During operation of the plant, particularly during production, the sealing channel and the conveying mechanism of the sealing channel can become clogged. These malfunctions lead to operational interruptions and are consequently associated with production interruptions.

[0272] Preferably, the sound sensor of the system is configured to receive a first acoustic signal, wherein the first acoustic signal comprises an acoustic signal emanating from the cutting device. The first acoustic signal may further comprise an acoustic signal emanating from the hem cutting device. Finally, the first acoustic signal may comprise an acoustic signal emanating from the conveying device, in particular from the conveyor belt. Page 61 / 74

[0273] P81102WO includes an acoustic signal. According to an advantageous embodiment, the system may include a second acoustic sensor, wherein the second acoustic sensor is configured to receive a second acoustic signal, the second acoustic signal emanating from the hem cutting device during the cutting of the metal strips into smaller pieces. Furthermore, the system may include a third acoustic sensor, wherein the third acoustic sensor is configured to receive a third acoustic signal, the third acoustic signal emanating from the conveying device, in particular from the conveyor belt, during the removal of the smaller pieces.

[0274] The first sound sensor can be arranged near one or more cutting devices. Preferably, the first sound sensor is physically connected to at least one cutting device, more preferably indirectly or directly.

[0275] The second acoustic sensor can be arranged near one or more hem cutting devices. Preferably, the second acoustic sensor is physically connected to at least one hem cutting device, preferably indirectly or directly.

[0276] The third sound sensor can be arranged near the seam channel. Preferably, the third sound sensor is connected to the seam channel, in particular to a wall of the seam channel, preferably directly or indirectly.

[0277] The second data set can contain the first acoustic signal and / or the second acoustic signal and / or the third acoustic signal.

[0278] The data processing unit may be configured to process the first acoustic signal and / or the second acoustic signal. Page 62 / 74

[0279] P81102WO to store the gnal and / or the third acoustic signal in the second data set.

[0280] A system designed in this way has the advantage that blockages of the seam channel or blockages of a conveying device can be prevented by early measures, such as maintenance measures, in order to avoid a malfunction and thus a production standstill.

[0281] As a non-limiting example, the invention will be explained in more detail using a second application example. A plant for metal production and / or metalworking can have at least one rolling mill with at least one rolling stand, wherein the rolling stand has one, preferably more than one, roll, in particular a working roll. During operation of a rolling mill, the working rolls are subject to wear, in particular mechanical and / or thermally induced wear of the roll surface, which can impair the product quality of the products to be manufactured with increasing wear. Wear, in particular increasing wear of the roll surface, can be determined by so-called pitch noises. These noises arise, for example, because the rolls rotate faster or slower on a metallic workpiece, i.e.,A relative movement occurs at the contact point between the roller surface and the surface of the metallic workpiece. Such pitch noises, and especially an increase in such pitch noises, can therefore indicate increasing wear.

[0282] The system preferably includes a sound sensor, the sound sensor being arranged such that an acoustic signal emanating from the rollers of the rolling mill's frame can be received. The acoustic signal can contain one or more characteristic signal components. (Page 63 / 74)

[0283] P81102WO indicates that pitch noises and / or changes in pitch noises are present.

[0284] The data processing device is preferably configured to store the acoustic signal received by the sound sensor in a second data record. A first data record can contain a reference signal, preferably more than one reference signal, wherein the reference signal can comprise an acoustic signal previously received by means of a sound sensor and emanating from one or more rollers of a rolling mill frame. The system state associated with this acoustic signal, in particular the state of the rollers, is known, and preferably a state in which the roller has not exceeded a previously determined wear limit of the roller surface.

[0285] The control unit is preferably configured to indirectly influence the operation of the plant based on the plant state determined by the data processing unit, in particular by outputting a warning signal via a warning device. For example, the control unit is configured to output a visual warning signal on a warning device designed as an operator screen.

[0286] It should be explicitly mentioned that features of the first and second application examples can each be combined with another application example, and / or with each other, both individually and cumulatively.

[0287] Further advantages, details and features of the invention will become apparent from the illustrated examples below.

[0288] Specifically, see pages 64 / 74.

[0289] P81102WO

[0290] Figure 1: a schematic view of a system according to a first implementation form; and

[0291] Figure 2: a schematic view of a system according to a second implementation form.

[0292] In the following description, identical reference symbols denote identical components or identical features, so that a description given for one component in relation to one figure also applies to the other figures, thus avoiding repetitive descriptions. Furthermore, individual features described in connection with one embodiment can also be used separately in other embodiments.

[0293] Figure 1 shows a schematic view of a system 1 for determining the operating state of a metal production and / or metalworking plant 100 according to a first embodiment, comprising a sound sensor 10 configured to receive an acoustic signal, preferably an acoustic signal emanating from the metal production and / or metalworking plant 100; a data processing device 20 connected to the sound sensor 10, wherein the data processing device 20 is configured to compare at least two data sets, a first data set comprising a reference signal and a second data set comprising the acoustic signal received from the sound sensor 10; and to determine the operating state of the metal production and / or metalworking plant 100 based on the comparison of the data sets.System 1 further comprises a control unit 30 which is data-connected to the data processing unit 20, wherein the control unit 30 is configured to influence the operation of the plant 100, at least indirectly, based on the plant state determined by the data processing unit 20. Page 65 / 74.

[0294] P81102WO

[0295] Plant 100 comprises a continuous casting unit 110 for producing a metallic workpiece 2. Furthermore, plant 100 comprises a rolling unit 120 following the continuous casting unit 110 in a conveying direction RI, wherein the rolling unit 120 has a first rolling stand 121 and a second rolling stand 122 following it in the conveying direction RI. Finally, plant 100 comprises a winding unit 150.

[0296] Figure 2 shows a schematic view of a system 1 for determining the operating state of a metal production and / or metalworking plant 100 according to a second embodiment, comprising a first acoustic sensor 11, a second acoustic sensor 12, and a third acoustic sensor 13. The first acoustic sensor 11 is configured to receive an acoustic signal emanating from a roller of the first rolling stand 121 of the rolling mill 120. The second acoustic sensor 12 is configured to receive an acoustic signal emanating from a roller of the second rolling stand 122 of the rolling mill 120. The third acoustic sensor 13 is configured to receive an acoustic signal emanating from a cutting device 130 and / or from a hemming device 140 of the plant 100.The cutting device 130 of the plant 100 is arranged in the conveying direction RI behind the rolling device 120 and the hem cutting device 140 is arranged in the conveying direction RI behind the cutting device 130.

[0297] The data processing unit 20 is data-connected to the first, second, and third sound sensors 11, 12, 13 and is configured to compare two data sets, wherein a first data set contains more than one reference signal and a second data set contains the acoustic signals received by the sound sensors 11, 12, 13; and based on the comparison of the data sets, to determine the plant status of the plant 100 for metal production and / or metal processing. (Page 66 / 74)

[0298] P81102WO agree. System 1 also has a warning device 40 which is data-connected to the control unit 30. The control unit 30, which is data-connected to the data processing unit 20, is configured to indirectly influence the operation of the system 100 by issuing a warning signal via the warning device 40, based on the system state determined by the data processing unit 20.

[0299] Page 67 / 74

[0300] P81102WO

[0301] Reference symbol list

[0302] 1 system

[0303] 2. Metallic workpiece

[0304] 10 sound sensor

[0305] 11 First sound sensor

[0306] 12 Second sound sensor

[0307] 13 Third sound sensor

[0308] 20 Data processing equipment

[0309] 30 Control unit

[0310] 40 Information device

[0311] 100 Plant for metal production and / or metal processing

[0312] 110 Continuous casting equipment

[0313] 120 rolling mill

[0314] 121 First rolling mill frame

[0315] 122 Second rolling mill

[0316] 130 cutting device

[0317] 140 Hem cutting device

[0318] 150 winding device

[0319] RI Conveyor Direction

Claims

Page 68 / 74 Applicant: SMS group GmbH Our reference number: P81102WO Patent claims 1. System (1) for determining the operating condition of a plant (100) for metal production and / or metalworking, comprising: a sound sensor (10) configured to receive an acoustic signal, preferably an acoustic signal emanating from the plant (100) for metal production and / or metalworking; a data processing device (20) connected to the sound sensor (10) via data transmission, wherein the data processing device (20) is configured to: compare at least two data sets, wherein a first data set contains a reference signal and a second data set contains the acoustic signal received from the sound sensor (10); and determine the operating condition of the plant (100) for metal production and / or metalworking based on the comparison of the data sets;a control device (30) which is data-connected to the data processing device (20), wherein the control device (30) is configured to influence the operation of the plant (100) at least indirectly, based on the plant state determined by the data processing device (20).

2. System (1) according to claim 1, characterized in that the sound sensor (10) is arranged at a distance from the system (100) for metal production and / or metal processing.

3. System (1) according to claim 1 or 2, characterized in that the sound sensor (10) is connected to the system (100) for metal product- Page 69 / 74 P81102WO is arranged in a positionally variable or positionally fixed manner for metalworking and / or metal processing.

4. System (1) according to one of the preceding claims, characterized in that the sound sensor (10) is designed as a microphone, preferably as an electrostatic microphone or as a piezoelectric microphone or as a dynamic microphone.

5. System (1) according to one of the preceding claims, characterized in that the sound sensor (10) has an electret.

6. System (1) according to any one of the preceding claims, characterized in that the sound sensor (10) is configured to receive acoustic signals with a sound frequency of < 2 GHz, preferably < 1.6 GHz and particularly preferably < 20 kHz.

7. System (1) according to any one of the preceding claims, characterized in that the sound sensor (10) is configured to receive acoustic signals with a sound frequency of > 1 Hz, preferably > 15 Hz and particularly preferably > 10 kHz.

8. System (1) according to one of the preceding claims, characterized in that the data processing device (20) is configured to: perform a time-frequency analysis of the acoustic signal received by the sound sensor (10), preferably before a system state of the plant (100) for metal production and / or metal processing is determined, and to store a result of the time-frequency analysis in the second data set. Page 70 / 74 P81102WO 9. System (1) according to one of the preceding claims, characterized in that the data processing device (20) is configured to filter one or more than one signal component with a sound frequency of < 1 Hz from the acoustic signal received by the sound sensor (10), preferably before a system state of the plant (100) for metal production and / or metal processing is determined.

10. System (1) according to one of the preceding claims, characterized in that the data processing device (20) is configured to filter one or more than one signal component with a sound frequency of > 2 GHz from the acoustic signal received by the sound sensor (10), preferably before a system state of the plant (100) for metal production and / or metal processing is determined.

11. System (1) according to one of the preceding claims, characterized in that the sound sensor (10) has a substitute noise level of < 30 dB, preferably of < 25 dB, preferably of < 15 dB and particularly preferably of < 10 dB.

12. System (1) according to one of the preceding claims, characterized in that the sound sensor (10) is configured to generate an acoustic identification signal, wherein a reflection of the acoustic identification signal generated by the sound sensor (10) at the plant (100) for metal production and / or metal processing is the acoustic signal received by the sound sensor.

13. System (1) according to one of the preceding claims, characterized in that the control device (30) is configured to indirectly influence the operation of the system (100) by issuing a warning signal, wherein the warning signal comprises an optical signal and / or an acoustic signal and / or a haptic signal. - 70- Page 71 / 74 P81102WO 14. System (1) according to claim 13, characterized in that the system has a notification device connected to the control unit (30) via data, preferably an operator interface, wherein the control unit (30) is configured to output the notification signal by means of the notification device.

15. System (1) according to one of the preceding claims, characterized in that the control device (30) is configured to directly influence the operation of the plant (100) based on the plant state determined by the data processing device (20), preferably by changing at least one operating parameter of the plant (100) for metal production and / or metal processing.

16. System (1) according to one of the preceding claims, characterized in that the comparison of the two data sets comprises a comparison of at least one amplitude of the reference signal with at least one amplitude of the received acoustic signal.

17. System (1) according to one of the preceding claims, characterized in that the comparison of the two data sets comprises a comparison of one or more than one amplitude of a previously determined frequency band of the reference signal with one or more than one amplitude of the same frequency band of the received acoustic signal.

18. System (1) according to one of the preceding claims, characterized in that the data processing device (20) is configured to determine the plant status of the plant (100) for metal production and / or metal processing by means of a machine learning method. - 71 - Page 72 / 74 P81102WO 19. System (1) according to one of the preceding claims, characterized in that the system has a plurality of sound sensors (11, 12, 13), in particular two or more, or three or more, or four or more, wherein the data processing device (20) is data-connected to the sound sensors (11, 12, 13) and wherein the sound sensors (11, 12, 13) are each spaced apart from each other.

20. System (1) according to claim 19, characterized in that the sound sensors (11, 12, 13) are equidistant from each other.

21. System (1) according to claim 19 or 20, characterized in that the sound sensors (11, 12, 13) are arranged in a regular pattern relative to each other, in particular in a rectangular pattern, in a square pattern, in a circular pattern or in an annular pattern.

22. System (1) according to claim 19, characterized in that at least a first sound sensor (11) has a distance to a second sound sensor (12) which differs from a distance between the first sound sensor and a third sound sensor (13).

23. System (1) according to one of the preceding claims, characterized in that the system (1) has an optical sensor connected to the data processing unit (20) and configured to receive an optical signal, preferably an optical signal emanating from the metal production and / or metalworking plant (100); wherein the data processing unit (20) is configured to additionally monitor the plant status of the metal production and / or metalworking plant (110) To determine the consideration of a third data set, -72- Page 73 / 74 P81102WO where the third data set contains the signal received from the optical sensor.

24. Use of a system ( 1 ) according to one of the preceding claims for at least indirectly influencing the operation of a plant ( 100 ) for metal production and / or metal processing .

25. Method for determining a plant condition, preferably using a system ( 1 ) according to any one of claims 1 - 23, wherein the method comprises the following steps: Receiving an acoustic signal emanating from a plant (100) for metal production and / or metalworking; comparing at least two data sets, wherein a first data set contains a reference signal and a second data set contains the received acoustic signal; Determining the state of the plant ( 100 ) for metal production and / or metalworking based on the comparison of the data sets; and at least indirectly influencing the operation of the plant ( 100 ) for metal production and / or metalworking based on the previously determined state of the plant ( 100 ) for metal production and / or metalworking . - 73 -

Images