System and method for identifying and distinguishing materials, method for identifying or distinguishing materials, and measuring device for recording material properties of materials

Abstract

A system for identifying or distinguishing materials, comprising at least one local apparatus and a central station. Each local apparatus comprises at least one measuring device for recording at least one actual signature for materials each and at least one local computer communicatively connected to the at least one measuring device, the at least one local computer having a local database for storing and/or processing the actual signature. The at least one central station comprises a server having a central database for storing and/or processing the actual signatures of the local apparatus. Furthermore, the system comprises a network, which communicatively connects the local computers of the local units via the server of the center. The invention further relates to a corresponding method for operating a system, to an analysis method for identifying or distinguishing the materials, and to a measuring device for recording material properties of the materials.

Claims

1. A method for identifying or discriminating materials with a system having at least one local unit, with each local unit having at least one measuring device for detecting at least one respective actual signature for the materials, the method comprising the steps of: acquiring the actual signature among a first quantity k of different measured values of a to be determined material by at least one user of a local unit of the system; providing a request of the at least one user to an operator of a center of the system, wherein with the request is the actual signature transmitted; storing the actual signature, provided by the at least one user, in non-volatile memory on a server of the center; creating a derived target signature from each of a plurality of correlations of at least one target material to the actual signature, wherein the plurality of correlations relate to an extended signature of the target material and wherein the extended signature consists of a second quantity n of different measured values and the second quantity n is greater than or equal to the first quantity k; calculating a deviation of each of the actual signatures for the material from the derived target signature for the at least one target material; and, reporting to the requesting user, for which target materials the respective deviation of the to be determined material, based on the at least one target material, is less than one for a target material defined tolerance.

2. The method of claim 1, wherein the actual signature is completed by metadata of the at least one user and/or to the operator.

3. The method of claim 2, wherein the metadata comprises a test report of at least the respective deviation and/or a certificate of the quality of the material and/or the measured actual signatures.

4. The method of claim 2, wherein for each user for at least one target material, the associated correlation and the associated derived target signature are optimized according to a combination of optimization criteria with associated weights, wherein the weights are modified based on metadata.

5. The method of claim 2, wherein the at least one user evaluates on the reporting of the operator the correct identification or discrimination of the material and reports a result of an evaluation back as metadata to the operator.

6. The method of claim 5, wherein the operator and/or a user on the basis of the metadata and/or the actual signatures: extends the actual signatures with measurement values detected by additional measurement methods and optimizes the derived target signatures and correlations for at least one target material; or, stores the to be determined material in a local database and/or in a central database as a new target material.

7. A method for identifying or discriminating of materials, comprising the steps of: selecting I measurement methods from m different measurement methods, wherein the number I is less or equal to m measurement methods; measuring an actual signature of a material by means of a measuring device for executing the I different measurement methods; generating a derived actual signature for each measured actual signature of the material from at least one associated correlation of at least one target material to the actual signature; determining a deviation per derived actual signatures of the material of a respective derived target signature of the at least one target material; comparing each deviation with a tolerance specified for the respective target material; and outputting for which target materials the respective deviation between the target material and material is less than the tolerance of the respective target material.

8. The method of claim 7, wherein the correlation associated with a target material enables the generation of the derived target signature from a target signature, wherein the target signature is generated by the selection from an extended signature.

9. The method of claim 8, wherein the actual signatures and the extended signatures consist of k and n measured values respectively, wherein the number n is greater than or equal to the number k.

10. The method of claim 9, wherein a target database for non-volatile storing of target data sets comprising the derived target signatures, the selections and the correlations for the at least one target material, is generated with the steps: applying of the I different measurement methods for generating a respective extended signature of the at least n measured values for at least one target material; creating a measurement database of extended signatures; optimizing each of a derived target signature, which is generated by the associated correlation from the associated target signature, which is generated by a selection of the measuring methods; applying of the n different measurement methods for generating a respective extended signature from the at least n measured values for at least one target material, which is suitable for product processing; creating a database of extended signatures; optimizing a derived target signature, which is generated by an associated correlation from an associated target signature, which is generated by a selection from the measurement methods; and, creating the target database from the derived target signatures, the selections and correlations.

11. The method of claim 10, wherein a computer-based algorithm carries out the optimizing step.

12. The method of claim 10, wherein the optimizing step takes place after a combination of optimization criteria, which are provided with associated weights according to their relevance for the processing of a material in a device for product processing.

13. The method of claim 7, wherein in a database a selected subset of reference data sets is stored non-volatile, wherein the reference data each consist of selection of measurement methods, a target signature, the correlation and/or the derived target signature of at least one target material, and/or wherein in the database, audit records are logged, which consist of the actual signatures, the derived actual signatures and/or the deviations.

14. The method of claim 13, wherein audit records of the database are transmitted automatically via a digital network to a further external database.

15. The method of claim 14, wherein the audit records are used in the database as a control parameter for a control circuit in the processing in a device for product processing.

16. The method of claim 13, wherein an external storage medium, in which the nominal data sets and/or audit records are stored, is readable connected with a container, a packaging and/or a charge of a desired target material, or of a material.

17. The method of claim 16, wherein the external storage medium is made up of a Radio-Frequency Identification (RFID) chip is, on which the data sets are stored in the form of RFID tags and/or wherein the storage medium is made up of a label on which the data sets are stored alphanumeric and/or as a barcode.

18. The method of claim 7, wherein a suitability of the at least one material for processing in a device for product processing is checked during the ongoing production process.

19. The method of claim 7, wherein application of a selection of the I measurement methods is at least ten times faster than the application of m measurement methods.

20. The method of claim 7, wherein a data processing means reads a database and controls a reduced set of u measuring devices for performing the selection of measurement methods on a material such that values contained in a database, which are to be measured for the comparison of the material with several target material are not redundantly collected.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Various embodiments are disclosed, by way of example only, with reference to the accompanying schematic drawings in which corresponding reference symbols indicate corresponding parts, in which:

(2) FIG. 1 is a schematic representation of the optimization method according to the invention;

(3) FIG. 2 is a schematic representation of the weighted optimization criteria and the optimization method;

(4) FIG. 3 is a schematic representation of the analysis steps according to the invention for identifying or distinguishing of a material from and against target materials, respectively;

(5) FIG. 4 is a schematic representation of the analysis method of the invention;

(6) FIG. 5 is a schematic representation of the processing system according to product;

(7) FIG. 6 is a schematic representation of a measuring device according to the invention for determining material properties of materials;

(8) FIG. 7A shows a sample of a material filled in a test specimen before pressurization by the plunger;

(9) FIG. 7B shows a sample of a material with filled in a test specimen after pressurization by the plunger;

(10) FIG. 7C a test specimen filled with a calibration device;

(11) FIG. 8 is a schematic representation of a system according to the invention for identifying or distinguishing of materials; and

(12) FIG. 9 is a schematic representation of the method for operating a system according to the invention for the identifying or distinguishing of materials.

DETAILED DESCRIPTION OF THE INVENTION

(13) At the outset, it should be appreciated that like drawing numbers on different drawing views identify identical, or functionally similar, structural elements of the invention. While the present invention is described with respect to what is presently considered to be the preferred aspects, it is to be understood that the invention as claimed is not limited to the disclosed aspect. The present invention is intended to include various modifications and equivalent arrangements within the spirit and scope of the appended claims.

(14) Furthermore, it is understood that this invention is not limited to the particular methodology, materials and modifications described and as such may, of course, vary. It is also understood that the terminology used herein is for the purpose of describing particular aspects only, and is not intended to limit the scope of the present invention, which is limited only by the appended claims.

(15) Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood to one of ordinary skill in the art to which this invention belongs. Although any methods, devices or materials similar or equivalent to those described herein can be used in the practice or testing of the invention, the preferred methods, devices, and materials are now described.

(16) FIG. 1 shows an embodiment of the inventive optimization method in a schematic representation. In step 11, one or more target materials M.sub.i are selected and measured with a set of m different measurement methods S1, . . . , Sm. Each measurement method S1, . . . , Sm appoints one or several different measurement values. Each measured value can be determined several times and combined statistically to a safe reading. For data analysis, the measured values are combined to form a feature vector. The m measurement methods S1, . . . , Sm are chosen such that the feature vector of target material M.sub.i represents a unique, characteristic extended signature 100i.

(17) In step 12, database 1 is created from extended signature 100i for each target material M.sub.i.

(18) In step 13, a random or empirical based start selection of I of the m measurement methods S1, . . . , Sm is made, wherein I is less than or equal to m. This is iteratively optimized. In i-th optimization iteration 13, i-th selection 210ii of measurement methods is initially specified, preferably in matrix notation. Applying selection (matrix) 210ii to extended signature 100i generates i-th target signature 200ii for target material M.sub.1, with lower or equal dimension. Then, an algorithm generates i-th correlation 220ii, their application to the signature generates i-th signature 300ii derived from the same dimensionality as extended signature 100i. By combination 40 (see FIG. 2) of optimization criteria 41a, . . . , 41p, optimization method 50, like that of an ANN algorithm, optimizes iteratively i-th selection 210ii and correlation 220ii until the global or a local optimum is determined. Optimization 14, for example, can be terminated by reaching a specified convergence criterion or after a fixed number m of optimization iterations 13. An optimum has selection 210i of I measurement methods S1, . . . , Sl, target signature 200i, correlation 220i and derived target signature 300i for a respective target material M.sub.i. This selection 210i can be any combination of the m measurement methods S1, . . . , Sm. Similarly, the number I of selected measurement methods S1, . . . , Sl can be different at start selection and optimized selection 210i. In an embodiment, weights and cross connections are not included in selection 210i. These can equally be considered in correlation 220i. This means, that selection 210i may be represented as a diagonal matrix with diagonal elements of 1 or 0. In this case, selection 210i indicates directly and without computing time on which measurement value is to be collected with which measurement method S1, . . . , Sm.

(19) FIG. 2 illustrates combination 40 of optimization criteria 41a, . . . , 41p and how they influence optimization method 50. The selection of optimization criteria 41a, . . . , 41p depends, according to the invention, on a purpose. The purpose may be, for example, to coat a substrate in a plasma-assisted coating system with material M.sub.j, which exhibits specific characteristics of target material M.sub.i, and the rapid, accurate process-supporting quality control of each coating process. In addition, the optimization criteria are weighted with a respective weighting 42a, . . . , 42p, that optimization step 14 (see FIGS. 1 and 3) is better tuned to the purpose. Accordingly, special process relevant material properties are weighted high (by a factor close to 1) and process not relevant material properties are weighted with a factor close to zero. Furthermore, time-consuming or costly measurement methods S1, . . . , Sm can be weighted less, so it is less likely to be included in optimized selection 210i. In particular, the optimization refers to the minimization of distance from extended signature 100i to derived target signature 300i of target material M.sub.i. Distance indicates the tuning error with which one can reconstruct extended signature 100i from target signature 200i. The optimization result can thus be a compromise between selectivity, measurement cost or measurement speed.

(20) FIG. 3 is a schematic representation of the inventive analysis method steps for identifying or distinguishing a material from or against target materials. The off-line sequence of steps 10 includes steps 11, 12, 13 and 14, which have already been described with respect to FIG. 1 and FIG. 2. In addition, the optimization results and derived target signatures 300i for each target material M.sub.i, are stored in step 15 in a target database and are transmitted to external database 5. Contained thereon is, for example, all information about suitable target materials for processing in a production process. Furthermore, a procedural requirement is included, how material M.sub.j matches against the target materials.

(21) Then, in-line step sequence 20 is executed. The term in-line designates that step sequence 20, in contrast to the off-line step sequence, is provided for the integration into a production line.

(22) A particular advantage of the analytical method and system of the invention is that an automated process control is possible. In first step 21, selection 210i of I measurement methods S1, . . . , Sl is applied to material M.sub.j, M.sub.h to be worked on or to be tested for each of the relevant target material M.sub.i. Out of it, based on target material M.sub.i, an actual signature 200j and 200h are generated, respectively. In the subsequent step 22, correlation 220i is applied to actual signatures 200j and 200h, respectively, to generate derived actual signatures 300j and 300h. Alternatively, optimization step 14 can be performed also at each test step. In the final step 24 identifying and distinguishing of materials M.sub.j, M.sub.h of target material M.sub.i and the qualification examination for processing takes place respectively. For this, the vector norm of the difference between the derived target signature 300i with the respective derived actual signature is 300j or 300h preferentially formed. In this case, the derived target signature 300i and/or the correlation 220i are taken from the external database 3. This deviation ij is compared with tolerance . If deviation ij for material M.sub.j is less than tolerance , it is considered that material M.sub.j belongs to a material class defined by target material M.sub.i and suitable for a production process respectively. If deviation ij for material M.sub.h is greater than the tolerance , it is considered that material M.sub.h does not belong to a material class defined by target material M.sub.i and is not suitable for a production process respectively.

(23) FIG. 4 is a schematic overview of the analysis method according to the invention. Target data sets are drawn from the target database 2 or the external database 3 and in-line step sequence 20, described in FIG. 3, is performed. Step 24 will be described in more detail here. After comparing 23 of the actual signature 300j and the target signature 300i and norm-building therefrom, it follows the scalar capability 25 with the tolerance . The test result and/or the determined and derived signature 300j and actual signature 200j respectively are documented on the external database 3. These audit records can then be transmitted to another external database 5 or product database 6. Audit records from product database 6, for example, can be stored on a label or an RFID chip in an alphanumeric manner as a barcode or as digital data, which can be readably joined with a product 63, in order to definitely mark and identify the product 63. The further external database 5 is preferably in the domain of the operator or manufacturer of device 60 for product processing. The continuous, preferably encrypted communication of audit records to the other external database 5 provides the opportunity for automated, transparent and tamper-proof quality control. Tolerance can be, depending on the objective of the inventive analysis method, specified or dynamically adjusted. If it is dynamically adjusted by the operator of apparatus 60 for product processing, it is recommended to store their variation in time on external database 3 and to transmit it to the other external database 5 or product database 6.

(24) Parallel processing 62 of material M.sub.j takes place in device 60 for product processing in order to obtain product 63. Supply means 61 can control the supply of the material M.sub.j in response to the audit records in the external database.

(25) FIG. 5 shows a schematic representation of the inventive system 90 for product processing. For material M.sub.j, intended for processing, signature 200j is generated by a set of measuring units A1, . . . , Au. Data processing system 80 reads actual signature 200j of measuring units A1, . . . , Au and target records from external storage medium 81. Furthermore, a data processing device 83 of the data processing system 80 checks the material M.sub.j by performing in-line step sequence 20 of FIG. 2. The results are reported as audit records to external database 3 and are transmitted, preferably encrypted, via remote data link 84 to another external storage medium 85. In addition, the audit records are transmitted to control unit 64 which regulates supply means 61 of device 60 for product processing for material M.sub.j in dependency of this audit records. In case the examination results in, for example, material M.sub.j not being suitable for processing 62 in device 60 for product processing, the supply of material M.sub.j is stopped, so that it will not be processed into product 63.

(26) FIG. 6 is a schematic representation of an embodiment of measuring device 400 for the detection of material properties of materials M.sub.j. Measuring device 400 may be received in externally accessible housing 402. In particular, measuring device 400 is suitable for compressible materials M.sub.j, such as powders or gases, and in particular for electrically conductive powders. However, little compressible fluids can be measured in general. Suspensions of solid particles in liquids can, according to the invention, be dried at least partially under controlled and reproducible conditions, so that they are available in powder form.

(27) Measuring device 400 comprises an electrically insulating test specimen 410 with a lower electrically conductive shutter 413, upper opening 411 and cavity 412 for receiving sample 414 of material M.sub.j. Test specimen 410 is preferably of cylindrical form. In order to obtain comparable and reproducible results, the initial conditions are set during the filling of sample 414 in test specimen 410 in a defined manner. For example, sample 414 has a defined weight of powder. The powder can be weighed by means of a balance. Preferably, it is filled by a powder metering apparatus (not shown), which makes not only the weight but also rheological relevant variables such as the compacting of the powder or measurement-related variables such as the shape of surface 415 of the powder or air pockets in specimens 410 controllable. The validity of the measurements increases, for example, if the same or a similar powder metering apparatus charges device 60 for product processing. Test specimen 410 is sitting in test specimen holder 418 during the measurement. Test specimen 410 and test specimen holder 418 can have corresponding positioning elements (not shown) in order to obtain a definite positioning of test specimen 410 in test specimen holder 418.

(28) The measuring device 400 further comprises a plurality of measuring units A1, . . . , Au associated to test specimen 410 for the execution of I different measuring methods S1, . . . , Sl on sample 414 in order to detect actual signature 220j of material M.sub.j of at least I time-dependent measurement values.

(29) One of the measuring units A1, . . . , Au is composed of an electrically conductive plunger 420 and pressure gauge 430 for detecting the pressure P acted upon sample 414 by plunger 420. Plunger 420 is formfitting with upper opening 411 of test specimen 410 and is received along axis A of measuring device 400 into test specimen 410 in order to pressurize sample 414 in cavity 412 of test specimen 410. Plunger 420 is coupled via rod 422 with pneumatic motor 421. The drive pressure of pneumatic motor 421 is a measure of the pressure P, exerted by plunger 420 on sample 414. Due to the form fit and the surface equality of plunger 420 and opening 411 of test specimen 410, the pressure P is in turn a direct measure for the force acting on sample 414. In a typical measurement, a start pressure, for example 0 bar, is acted upon, which is increased at appropriate intervals, for example 100 mbar, up to an end pressure, for example 5 or 6 bar. The pressure P is registered as a function of the time.

(30) Another one of measuring devices A1, . . . , Au is distance meter 440 for detecting distance W traveled by plunger 420 along axis A. For example, for this purpose, an inductive odometer that detects distance W, rod 422 travels relative to frame 403, which rigidly supports pneumatic motor 421 and test specimen holder 418. Consequently, this distance W (shown in FIG. 6) corresponds to the distance W of plunger 420 along axis A (as is shown in FIG. 7). Distance W is detected during the pressurization of plunger 420 as a function of time.

(31) Another one of measuring devices A1 , . . . , Au is electrical measuring instrument 450. Plunger 420 is electrically conductive. If plunger 420 is in contact with sample 414, a current path forms along axis A of plunger 420 through sample 414 into the electrically conductive shutter 413 of test specimen 410. At plunger 420 and closure 413 is electrical measuring instrument 450 is connected in order to measure the end-to-end resistance R relative to the longitudinal axis A with a 2-point measurement configuration or, preferably, a current less 4-point measurement configuration, as illustrated in FIG. 6. Beneath closure 413 electrical contact 417.

(32) Also conceivable is the measurement of an end-to-end resistance R with respect to longitudinal axis A or the additional determination of the temperature T of the sample by one or more thermometers 480, which for example is designed as a Peltier element. Thereby, the temperature change of sample 414, caused by compression, is detected at different locations of sample 414, which allows conclusions to be drawn about the heat capacity and thermal conductivity of material M.sub.j. Electrical measuring instrument 450 includes, for example, voltage source 451 and ohm meter 452, preferably with plural series resistors (not shown) and a relay (not shown) in order to wire series resistors for preferably automated switching between different measurement ranges for high impedance and low impedance samples. The end-to-end resistance of sample 414 is detected during pressurization of plunger 420 as a function of time. An initially high impedance powder can become successively low impedance during pressurization, so it is convenient if the measuring range can be changed one or more times during a typical measurement.

(33) Another one of the measuring devices A1, . . . , Au is optical measurement instrument 460. Optical measurement instrument 460 can be a CCD camera. Likewise, the CCD camera can be equipped with a lighting device, in order to obtain an image of a portion of surface 415 of sample 414 under environmentally independent illumination conditions. It is also conceivable to arrange a calibration sample (not shown) in the optical path between sample 414 and the camera. The calibration sample may comprise a hole through which surface 415 is photographed, as well a color calibration mark, of at least three color segments spanning a color space (for example RGB), a white mark for calibrating the illumination conditions and the brightness distribution and/or a black and white mark for focusing the camera and for calibration of the edge sharpness or the contrast. The calibrations can be carried out computerized prior each measurement. The optical image of sample 414 can be captured during the pressurization of plunger 420 as a function of time. Alternatively, it is recorded only once or several times, preferably at defined time intervals, for example, when sample 414 is subjected to the initial pressure and the final pressure because then the principle pressure-induced morphology of the powder in sample 414 is in a defined and reproducible state. The optical measurement can be done along each spectral range, for example via the visual spectral range, an infrared spectral range, particularly for temperature detection, or a X-ray spectral range, for detecting structures of sample 414 below its surface 415. In particular, optical measuring instrument 460 may be mounted in test specimen holder 418 and directed to test specimen 410. Test specimen 410 is then made at least in part of the surface and/or transparent in the spectral regions to be measured.

(34) Another of the measuring units A1, . . . , Au may be an active or passive acoustic measuring instrument 470 for detecting the acoustic eigenspectrum of sample 414. In particular, acoustic measuring instrument 470 may be held in test specimen holder 418 and directed to test specimen 410 and is acoustically coupled with the powder via the test specimen 410. For this, test specimen 410 is formed such that it is resonant in the acoustic frequency ranges to be measured. Acoustic measuring instruments 470 can provide information about conditions of still or moving liquids, powders or solids, for example the filling level in test specimen 410, flow velocity of the powder or the presence or absence of impurities or foreign bodies. Acoustic transducer can generate electronic signals from acoustic oscillations.

(35) In passive acoustic sensors, a noise that the observed process itself generates (for example the noise of the powder during the compression by plunger 420) is evaluated. Active acoustic sensors, for example, generate an ultrasonic field in order to excite the powder and to measure its sound frequency and/or time-resolved response. From these sound reflections, for example positions of boundaries can be determined by sound diffraction at edges or frequency changes (for example by means of the acoustic Doppler effect) velocities are determined.

(36) Measurements with acoustic sensors are (for most materials M.sub.j) non-invasive, suitably gentle to the particular requirements of material M.sub.j, and flexibly adapted, for example by modulation of the excitation sound frequency. Acoustic measuring instrument 470 may comprise, for example a piezoelectric actuator or sensor. Piezoelectric devices are robust, reliable, easy to use, compact configurable and flexible to use. A piezoelectric material converts applied electrical voltage into mechanical or acoustic oscillations (operated as actuator) or vice versa (sensor mode). A piezoelectric crystal and piezoelectric actuator respectively coupled acoustically to sample 414 can therefore absorb the mechanical vibrations and convert them directly into an electrical measurement signal. The frequency spectrum of this electric signal depends definitely from the frequency spectrum of the mechanical vibrations detected by the piezo. The information to be measured is extractable from the frequency spectrum, for example its amplitude spectrum and phase spectrum and consists for example of material M.sub.j typical resonant frequencies. Preferably, the acoustic measurements are recorded during the pressurization of plunger 420 as a function of time.

(37) It is also conceivable that optical instruments 460, acoustic measuring instruments 470 and/or thermometers 480 are integrated in plunger 420 and/or shutter 413 of test specimen 410.

(38) Measuring device 400 has data acquisition device 460 communicatively connected to measuring units A1, . . . , Au via data lines 401, so that it can detect k time-dependent measurement values of actual signature 200j. In particular, data acquisition device 460 is designed for synchronized recording of these k time-dependent measurement values. In an example embodiment, data acquisition device 460 is an analog-digital converter which converts the measurement values output, usually analog measured, from measuring units A1, . . . , Au values into digital values. By synchronizing the time t can act as a correlation value between measured values by different measuring units A1, . . . , Au. For example, pressure gauge 430, the pressure P=P (t) and distance meter 440 measures distance W=W (t) synchronized with one another as a function of the time t, so that the pressure P and distance W, for example, are correlated to a measured curve P(W) as a function of pressure P and distance W.

(39) The information content of the correlated measured curve P(W) is higher than that of temporally uncorrelated measured curves of the pressure P=P(t) and the distance W=W (t). The additional information content is simply determined by fitting of the measured curve, for example with a sigmoid function in the form of fit parameters. Analogously, higher dimensional measured curves can be created and determined, for example the pressure P=P (W, R, T) in dependence of distance W, the end-to-end resistance R and a temperature T of sample 414. Actual signatures 200j or extended target signatures 100i can comprise both temporally correlated and uncorrelated values. An example of a time uncorrelated value with the pressure P=P (W, R, T) are values which have been extracted from a single captured image of the sample 414.

(40) In derived actual signatures 300j and derived target signatures 300i are contained additional hidden correlations, which are evaluated by an algorithm according to the invention. Data acquisition device 460 is linked via standardized interface 461, for example a USB port, with local computer 541. Local computer 541 can execute the temporal correlation of the time-dependent measurement values P =P (t), W =W (t), R =R (t), etc., determine the fit parameters of correlated waveforms such as P(W), compile actual signatures 200j, determine derived actual signatures 300j by applying correlation 220i of target materials M.sub.1 to actual signatures 200j and possibly perform the algorithm for determining selections 210i, correlations 220i and derived target signatures 300i for target materials M.sub.i.

(41) FIGS. 7A and 7B show test specimen 410 with filled sample 414 of material M.sub.j before and after application of a pressure P by plunger 420. Plunger 420 pulls back distance W. On this path, plunger 420 compresses sample 414 by a compression distance L. The beginning of the compression distance L results, for example, from the closing of the circuit between conductive plunger 420 and sample 414, conductive shutter 413, possibly test specimen holder 418 and electrical measuring instrument 450 or from the correlation of pressure P and distance W.

(42) It is also conceivable that measuring device 400 is regularly calibrated with calibration device 419. Calibration device 419 is used, for example, for calibration of the measurements of pressure P, distance W and possibly resistance R and, for example, a conductive spring with defined electrical resistance and defined spring constant (both uncompressed and/or charged with a certain pressure P). As shown in FIG. 7C, calibration device 419 is, analog to sample 414, enclosed in a form-fitting manner in test specimen 410 of measuring device 400 and measured. It is conceivable that user 511, 521, 531 of measuring device 400 regularly measures calibration device 419 instead of sample 414. Alternatively, calibration device 419 can be included in a second test specimen 410 in test specimen holder 418 of measuring device 400, so that calibration device 419 can be measured simultaneously with sample 414.

(43) FIG. 8 shows an exemplary embodiment of system 500 according to the invention for identifying or distinguishing of materials M.sub.j in a schematic representation. System 500 consists of at least one local unit 510, 520, 530, which is associated with a respective user 511, 521, 531. Furthermore, center 550 of operator 551 is at least associated with system 500.

(44) Each local unit 510, 520, 530 includes at least one measuring device 400 for detecting at least one respective actual signature 220j for materials M.sub.j. A preferred embodiment of such a measuring apparatus 400 is described with reference to the preceding FIGS. 6 through 7C. Measuring device 400 is communicatively coupled to local computer 541 of local units 510, 520, 530, respectively. Local computer 541 includes local database 4 and serves to store and/or process actual signature 220j.

(45) According to the inventive concept, different types or classes of users 511, 521, 531 are integrated in system 500. A first class of users 511, 521, 531 uses their local unit 510, 520, 530 for quality control of materials M.sub.j, M.sub.h like powders for plasma coating system or powdered food. They are interested in a fast, accompanying production, cost-effective, reliable, documentable and possibly certifiable powder analysis. Accordingly, measuring devices 400 of this first class of users 511, 521, 531 include a selection of measuring units A1, . . . , Au from all measuring units A1, . . . , Au, . . . , Aw, . . . , Av which can be used effectively for the analysis on materials M.sub.j, M.sub.h. Which u concrete measuring devices A1, . . . , Au, depends on the application and the technical specifications of user 511, 521, 531. These specifications are transmitted in the form of metadata to center 550 and are evaluated there, where they are considered in combination 40 of optimization criteria 41a, . . . 41p and their weighting 42a, . . . , 42p. For example, the color composition of a powdered dye for an auto paint shop is high and is low weighted for the manufacturer of non-visible primary color in the final product.

(46) A second class of users are for example, analysis provider or university laboratories with measuring devices 400, which have a set w of measuring units A1, . . . , Aw. The w measuring devices A1, . . . , Aw can have u measuring devices A1, . . . , Au above the first class of users 511, 521, 531 or additional measuring devices that can collect meaningful measurements in time-intensive and/or cost-intensive manner, for example a synchrotron or elaborate (wet) chemical processes. This second class of users 511, 521, 531 can verify actual signatures 200j, collect additional measurement values if the identification or discrimination of material M.sub.j, M.sub.h is not definite enough, or record extended target signatures 100i for target materials M.sub.i, which are not yet stored in central database 7. Furthermore, they may transmit test conditions, test certificates as metadata. Also, users 511, 521, 531 of the first class can mutually check the measurements contained in actual signatures 200j. This peer review can be arranged through center 550 and advantageously be anonymous among users 511, 521, 531.

(47) The at least one central 550 includes server 552 with central database 7 for storing and/or processing of actual signatures 220j of local units 510, 520, 530. Also in center 550, one or more further measuring devices 404 may be provided, which all A1, . . . , Au, . . . , Aw, . . . , Av, are useful dedicated to the analysis to each for user 511, 521, 531 relevant material. Materials M.sub.j, M.sub.h are considered to be relevant if they are intended for the processing or use, or impurities or toxins that are excluded during processing or use.

(48) Network 560 communicatively connects local computers 541 of local units 510, 520, 530 and the server. This communication link can exist in particular via the internet or an intranet of a company. This cloud-based system 500 enables the exchange of information among users 511, 521, 531. Preferably, local computers 541 are connected via server 552 of center 550, so that all actual signatures 200j of materials M.sub.j, and (optionally) extended target signatures 100i of material M.sub.j, are collected by users 511, 521, 531 in central database 7 and are able to be analyzed with server 552, for example in the manner as described with reference to FIGS. 1 to 4.

(49) The maintenance and the training of central database 7 steadily improves the data situation about target material M.sub.i, and thus identifying or distinguishing materials M.sub.j, M.sub.h becomes more selective. At the same time, identifying or distinguishing becomes more efficient, because the sufficient accuracy for a particular user 511, 521, 531 requires, due to the improved data situation in central database 7, fewer measuring units A1, . . . , Au and measurement methods S1, . . . , Sl in measuring device 400 of local units 510, 520, 530.

(50) An algorithm determines which measuring units A1, . . . , Au are dispensable. In the clustered recording and evaluation of the information of the whole system 500 on server 552 of center 550, an additional benefit exists for users 511, 521, 531 of local units 510, 520, 530. The additional benefit consists particularly in a statistically broader database of the collective of users 511, 521, 531 participating in system 500.

(51) The basic idea of the invention is to improve the identifying or distinguishing of complex materials M.sub.j by a network-based analysis method and a good statistical basis, without the requirement of an analytical understanding of possibly highly complex composite material M.sub.j. Another benefit is that each user 511, 521, 531 can also rely on the heuristic experience of other users 511, 521, 531. This is accessed indirectly via operator 551 of center 550 of system 500. Actual signatures 200j and metadata, provided by users 511, 521, 531 are in principle sensitive, because they allow to draw conclusions to trade secrets and production secrets. The anonymized grouping of these data with respect of third parties by a trusted operator 551 of center 550 allows all users 511, 521, 531 to benefit from the entire contents of central database 7, without revealing sensitive information. It is also conceivable that users 511, 521, 531 themselves maintain and train their local databases 4.

(52) FIG. 9 shows a schematic representation of the method for operating system 500 for identifying or distinguishing of materials M.sub.j, in particular of system 500 as illustrated in FIG. 8.

(53) In a first step of the process, acquiring 610 takes place of actual signature 220j of k different measured values of specific material M.sub.j by at least one user 511, 521, 531 of local unit 510, 520, 530 of system 500. The k various measurement values are collected with I different measurement methods S1, . . . , Sl with u measuring units and A1, . . . , Au. The I is greater than or equal to u because a measurement method S1, . . . , SI can collect only one or more measurement values, and likewise measuring unit A1, . . . , Au only performs one or several measurement methods S1, . . . , Sl. The same is true for n different measurement values, which are detected by the m different measurement methods S1, . . . , Sk, . . . , Sm by means of v measuring units A1, . . . , Av. It follows that k, l, and u are less or equal to n, m, and v, respectively.

(54) Acquiring step 610 is followed by requesting 620 of the at least one user 511, 521, 531 to operator 551 of center 550 of system 500. With the request, at least actual signature 220j is transmitted. In an example embodiment, metadata related to user 511, 521, 531 is transmitted with actual signature 220j. Metadata cover, for example, the identity of user 511, 521, 531, his specifications of material M.sub.j, his comments or the device ID of its measuring units A1, . . . , Au.

(55) Forwarded by the at least one of the users, actual signatures 220j (and, when applicable, the metadata) are directly stored in non-volatile memory or buffered, evaluated and non-volatile stored in evaluated form on server 552 of center 550 of system 500. From actual signatures 220j (and, when applicable, from the metadata), derived target signature 300j is generated. As described with reference to FIGS. 1 to 3, this is done by applying correlation 220i per actual signature 220j for at least one target material M.sub.i. Correlation 220i is based on extended signature 100i of target material M.sub.i. Each of the extended signatures 100i consists of n different measurement values. On server 552 of center 550 of system 500, deviation ij is derived from actual signature 300j of material M.sub.j, in relation to the respective extended signature 100i of the at least one target material M.sub.i.

(56) Then, reporting 630 is carried out by server 552 and operator 551 of center 550 of system 500, respectively, to the requesting user 511, 521, 531, if the respective deviation ij of material M.sub.j with respect to target material M.sub.i is less than tolerance for target material M.sub.i. In an example embodiment, this result can be returned as a list, which notes a probability in percent, which depends from the difference of deviation ij and tolerance and, if applicable, taking the metadata into account with which material M.sub.j coincides with target material M.sub.i. If the result is ambiguous because several target materials M.sub.i coincide with high probabilities with material M.sub.j, there are two options that can take place in principle.

(57) The first option includes the requesting user 511, 521, 531 evaluating the result and rejecting any implausible matches based on, for example, uncommunicated metadata or his experience. For example, a milk powder manufacturer would be able to reject a match of his milk powder with a powdery car paint, even if the metrological analysis indicates a high level of agreement. According to the invention, this evaluation 640 of the requesting user 511, 521, 531 is reported as feedback to operator 551 of system 500.

(58) In the second option, the measured values of the actual signature 200j are collected by users 511, 521, 531 or operator 551 again, or are completed by additional readings of previously unrelated measurement methods S11 . . . , Sm. In particular, evaluation 640 of requesting user 511, 521, 531 serves for further training of central database 7 and/or as a basis for decisions such as which measurement values need to be measured in addition, or whether a new target material M.sub.i needs to be defined. Also, empirical values can be saved as metadata and assigned to certain matches (for example, the above implausible coincidence between milk powder and car paint).

(59) The entire invention is subject to the basic idea to improve the analysis of complex materials with networked and standardized measuring units and methods by recording and evaluating a wide statistical database without the need of an accurate analytical understanding of the interrelationships of various properties of complex materials. Therefore, features that apply with respect to one aspect of the invention, as defined in the independent claims, are also applicable to their other aspects. One skilled in the art will also understand that the disclosed features, removed from the context of illustrating the invention, merely have exemplary context, are combined with the features of the independent claims.

(60) Although modifications and changes to the invention should be readily apparent to those having ordinary skill in the art, such modifications are intended to be within the spirit and scope of the invention as claimed. It also is understood that the foregoing description is illustrative of the present invention and should not be considered as limiting. Therefore, other embodiments of the present invention are possible without departing from the spirit and scope of the present invention as claimed.