DE102018206521A1 - Method for handling measured data - Google Patents

Abstract

The invention relates to a method for processing measured data of machines (12, 14, 16) which are stored in a result file, wherein in a first step values of the measured data are identified which correspond to predetermined conditions, in a second step variable names are searched for and In a further step, positions of the values in the result file are determined and noted.

Description

The invention relates to a method for treating measured data and to a computing unit for handling measured data. The invention further relates to a computer program and a machine-readable storage medium for carrying out the method.

State of the art

In the production of components of a motor vehicle, such as in the production of a common rail injector, several machines are used. These provide measurement data that must be recorded and evaluated. When collecting and analyzing this data, software tools called tools are used. It is important to store the measurement data in tables, where it can be accessed quickly and reliably. For this purpose, the data must be treated in a special way, as described herein. In this case, programs are used which perform numerical calculations, for example with the aid of matrices. These programs allow the solution of mathematical problems and the graphical representation of the results.

The publication DE 10 2016 119 186 A1 describes a method for distributed performance monitoring and analysis of industrial plants. In this data from a plurality of machines are collected and transmitted to a central processing unit and evaluated there. In the method, a set of process variables may be generated using machine learning. New information or insights can be discovered, also an evaluation of data can be learned. The method may employ computer programs as previously explained.

From the publication DE 10 2009 024 102 A1 For example, a method of providing information about a component of a machine is known. In the method, an algorithm is provided in a memory of the machine, a replacement algorithm is received via an interface of the machine, a plurality of operating parameters of the machine are obtained, and information for wear of the machine is determined and output.

Disclosure of the invention

Against this background, a method for treating measurement data according to claim 1 and a computing unit for carrying out the method with the features of claim 7 are presented. Furthermore, a computer program according to claim 8 and a machine-readable storage medium according to claim 9 are presented. Embodiments result from the dependent claims and the description.

With the presented method, it is possible to comfortably and quickly access machine data, for example machine data of a CRI (Common Rail Injector) production, and to assign these one-to-one to the measured component. The challenge of doing this, taking into account the changeable name naming and the number of variable measures or labels, was countered by a self-learning algorithm. This enables complete acquisition of all measured variables per measurement. In a computer-implemented embodiment of the method, the corresponding computer program can run decentrally on the user's own computer. The evaluation of the recorded data can take place with a suitable program.

In the presented method, all measured variables can be reliably detected by a self-learning algorithm. Instead of searching for defined variable names and then reading out the associated value of the variable, the search algorithm presented herein recursively proceeds.

The individual steps in one embodiment are as follows: in a first step, all values that correspond to conditions are localized. Conditions are for example: Is the value a decimal number or is the value a string or a string? In a second step the variable names are searched for. In a further step, the exact positions of the values in the result file are determined and noted. Using the positions learned at a result file, further, similar result files can be very efficiently filtered for variable contents.

In particular, measurement data from all machines stored according to the so-called OpCon DAT file format (* .dat) can be recorded. A trial and a validation takes place on various machines of common rail production and locations.

Features of the program are:

• - Messdaten können direkt auf dem Datenkonzentrator bspw. über ein Dialogfenster ausgewählt werden. Ein zuvor durchgeführtes Kopieren der Daten ist nicht notwendig. Die Einlesereihenfolge ist alphabetisch auf Dateinamen.
• - Anlernen der Messgrößen bzw. Labels: Auf Basis der ersten Messung in Einlesereihenfolge. Lokalisieren aller Ergebniswerte, die bestimmte Bedingungen erfüllen. z. B. ungleich Null, Gleitkommazahl und Identifikation des zusammengesetzten Labels, z. B. Result-XXX-Vall-Value.
• - Analyse der erfassten Messdaten in Programmen zur Lösung mathematischer Probleme möglich. Es stehen eine Vielzahl von Werkzeugen, z. B. Statistik, mit graphischer Bedienoberfläche bereit.
• - Zusätzlich erfolgt eine Ausgabe der erfassten Daten in einem Tabellenkalkulationsprogramm. Weiterhin erfolgt eine Ablage der Ergebnisdatei im Arbeitsverzeichnis.

Evaluated measurement data can be filtered by product names: 20 21- Type Wanted = 0445110814 '; 22- tic 23 %% Identify relevant file names

• - Measurement data can be selected directly on the data concentrator eg via a dialog box. A previously performed copying of the data is not necessary. The reading order is alphabetically on file names.
• - Teaching the measured variables or labels: Based on the first measurement in read-in order. Locate all result values that meet specific conditions. z. B. non-zero, floating-point number and identification of the composite label, z. B. Result-XXX-Vall-Value.
• - Analysis of acquired measurement data in programs for solving mathematical problems possible. There are a variety of tools, such. B. statistics, with graphical user interface ready.
• - In addition, the collected data is output in a spreadsheet program. Furthermore, the result file is stored in the working directory.

It should be noted that the focus of the presented method is on the acquisition of the data and not on the evaluation of the data.

Distinction to conventional data acquisition in CRI production:

The reference for the evaluation of the machine data is the quality control computer of the TEF3.4, version V3.9.0.44G. The development of a separate tool or tool for data acquisition was necessary, because again and again new measures or labels were added at short notice. These have to be manually retraced in a QLR database and therefore were not available until days or weeks later. For immediate validation of these innovations, these delays are unacceptable.

Further advantages and embodiments of the invention will become apparent from the description and the accompanying drawings.

It is understood that the features mentioned above and those yet to be explained below can be used not only in the particular combination indicated, but also in other combinations or in isolation, without departing from the scope of the present invention.

list of figures

• 1 shows a possible arrangement for carrying out the method.
• 2 shows in a flow chart an embodiment of the presented method.
• 3 shows in a flow chart a further embodiment of the presented method.

Embodiments of the invention

The invention is schematically illustrated by means of embodiments in the drawings and will be described in detail below with reference to the drawings.

1 shows an embodiment of an arrangement, in total with the reference numeral 10 is designated and with which the presented method can be performed. The illustration shows a first machine 12 , a second machine 14 , a third machine 16 , a data concentrator 18 , a computing unit 20 for carrying out the method and a direct connection 22 ,

Measurement data from all machines can be recorded 12 . 14 . 16 which are stored according to the OpCon DAT file format (* .dat). Access is either via the data concentrator 18 or directly via the direct connection 22 ,

2 shows an embodiment of the described method in a flowchart. In a first step 50 the procedure begins. In a subsequent step 52 a result file is scanned to teach or adapt the algorithm. Subsequently, in one step 54 all values found.

After that, in one step 56 all values checked for conditions. If an entry in the result file does not satisfy the conditions, this entry is ignored by the algorithm (step 58 ). If the conditions are met, the associated label is searched in the structure of the result file and both, value of the variables and labels, are searched within the matrix (step 60 ). Subsequently, all remaining result files are addressed sequentially (step 62 ) and extract therefrom the values of the noted variables at the learned positions (step 64 ).

Thereupon, in one step 66 filtered files from all files connected in a matrix in the selected file format output. With step 68 the procedure ends.

The program is started from a start script. Subsequently, additional secondary or sub-functions are addressed. The course of the program corresponds to the simplified scheme of the following scheme, which is described in 3 is shown in a flow chart.

In a first step 150 starts the procedure. In an optional further step 152 a filter type part number is applied. Connecting will be done in one step 154 run a script. Thereupon, in one step 156 in a dialogue Measurement data folder selected. Subsequently, in step 158 Subfunctions executed. Then in step 160 Variables set in the workspace (workspace). In a subsequent step 162 * .xlsx data is exported. With step 164 the procedure ends.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 102016119186 A1 [0003]
• DE 102009024102 A1 [0004]

Claims (9)

Method for handling measurement data from machines (12, 14, 16) stored in a result file, wherein in a first step values of the measurement data are localized, which correspond to predetermined conditions, in a second step variable names are searched and in a further step Positions of the values in the result file are determined and noted. Method according to Claim 1 , in which new measured variables are detected on the basis of the previously determined position. Method according to Claim 1 or 2 in which access to the measurement data via a data concentrator (18). Method according to Claim 1 or 2 in which access to the measurement data via a direct connection (22). Method according to one of Claims 1 to 4 in which measurement data to be evaluated are filtered via injector type part number. Method according to one of Claims 1 to 5 , in which an output of the collected data takes place in a spreadsheet program. Arithmetic unit for handling measurement data, which is used to carry out a method according to one of the Claims 1 to 6 is set up. Computer program with program code means adapted to perform a method according to any one of Claims 1 to 6 execute, if the computer program on a computing unit (20), in particular a computing unit (20) according to Claim 7 , is performed. Machine-readable storage medium with a computer program stored thereon Claim 8 ,

Images