WO2024025503A1 - Machine learning technic and big data backup - Google Patents

Machine learning technic and big data backup Download PDF

Info

Publication number
WO2024025503A1
WO2024025503A1 PCT/TR2023/050729 TR2023050729W WO2024025503A1 WO 2024025503 A1 WO2024025503 A1 WO 2024025503A1 TR 2023050729 W TR2023050729 W TR 2023050729W WO 2024025503 A1 WO2024025503 A1 WO 2024025503A1
Authority
WO
WIPO (PCT)
Prior art keywords
bending
data
big data
machine
backup
Prior art date
Application number
PCT/TR2023/050729
Other languages
French (fr)
Inventor
Özlem KAYA TOPÇUOĞLU
Original Assignee
Kaya Topcuoglu Oezlem
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Kaya Topcuoglu Oezlem filed Critical Kaya Topcuoglu Oezlem
Publication of WO2024025503A1 publication Critical patent/WO2024025503A1/en

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/18—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form
    • G05B19/4097—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form characterised by using design data to control NC machines, e.g. CAD/CAM
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B21—MECHANICAL METAL-WORKING WITHOUT ESSENTIALLY REMOVING MATERIAL; PUNCHING METAL
    • B21D—WORKING OR PROCESSING OF SHEET METAL OR METAL TUBES, RODS OR PROFILES WITHOUT ESSENTIALLY REMOVING MATERIAL; PUNCHING METAL
    • B21D5/00—Bending sheet metal along straight lines, e.g. to form simple curves
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/36—Nc in input of data, input key till input tape
    • G05B2219/36203—Bending of workpiece, also for long slender workpiece
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/37—Measurements
    • G05B2219/37403—Bending, springback angle
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/45—Nc applications
    • G05B2219/45143—Press-brake, bending machine

Definitions

  • the invention is directed to “the machine learning technic and big data backup” which allows getting bending results much faster and faultlessly by communicating machines with each other that are used for bending with high precision without operator and regardless of the type or strength value of material, especially at plate, profile, and metal forming machines.
  • the existing invention is directed to the machine learning technic and big data backup which supplies all the requirements above, eliminates all disadvantages, and provides additional advantages.
  • the main purpose of the invention is; to make bending faster by structuring that eliminates operator’s skill and provides everybody able to use easily and faultlessly by machine learning method and to learn by benefits from its previous bendings either made by their own or from other machines done and to provide to upgrade bending times.
  • Another purpose of the invention is; to able to learn through the other machines’ knowledge by connectimg to the big data backup and also to provide other machines benefit by recording their own knowledge.
  • Another purpose of the invention is; to upgrade the knowledge it learned from its own bendings within the own not only from big data and to provide machine learning without big data connection.
  • FIG. 1 there is a schematic view of the invention subject product.
  • FIG. 1 there is a schematic view of the invention subject product.
  • First bending material is placed on the machine (20) by choosing from the menus according to the specifications. Necessary adjustments are made and bending diameter is defined. After that step, if the machine (20) is connected to the big data backup (10), it continues bending by listing the values after searching according to the bending diameter and material dimensions which are indicated either at its own bending backups or at the big data backup. Bending is done by the machine until reached the bending diameter geometry and when it reached the bending diameter, the material spring back value is calculated by cnc attachment on the machine. It interprets by matching the results at its own backup or at the big data backup according to the ingenerate spring back values.
  • the big data backup that is used in the system is a kind of structuring like a cloud or it can be worked like a server, web-based backup.
  • the system can produce machine learning without being connected to the big data, and because of being it learns just from the bendings that it is done by its own, upgrading the time of the bending process takes longer.
  • the system can produce learning from other machines’ bendings which are inside of a closed cycle by a big data backup that will be built in a plant with a closed cycle.

Abstract

Innovation is directed to; the machine learning technic which includes one big data backup (10), one machine (20), and one internet provider (30), and big data backup structuring.

Description

DESCRIPTION
MACHINE LEARNING TECHNIC AND BIG DATA BACKUP
TECHNICAL FIELD
The invention is directed to “the machine learning technic and big data backup” which allows getting bending results much faster and faultlessly by communicating machines with each other that are used for bending with high precision without operator and regardless of the type or strength value of material, especially at plate, profile, and metal forming machines.
STATE OF THE ART
Because of being the profile or plate forming process is done by operators at today’s workshops, industry, etc., and the spring back value that generates while forming the profile or plate is changed according to the material hardness, there is a serious waste of time and salvage costs.
Also, most of these bendings are similar to each other within definite measurements, the same bending diameters can’t be provided even if the materials have the same hardness value. The waste of salvage and time is changed according to the operator, the material’s hardness value varies and it increases the waste of salvage.
The machine CNC transformation attachment that we design to solve that issue with patent number 2022/003965 solves this negation partly. However, because the bending process time will be longer, this method is suitable only for precision works, it can’t upgrade the bending process which he has done with machine learning. In these circumstances, to get more precision and much faster bending works the structuring has to be programmed with machine learning and has to get the old bending process much faster and perfect by taking decisions toward measurements that come after geometrically bending by making benefits from either a piece of pieces of knowledge at big data backup or from the machine’s own memory. PURPOSE OF THE INVENTION
The existing invention is directed to the machine learning technic and big data backup which supplies all the requirements above, eliminates all disadvantages, and provides additional advantages.
The main purpose of the invention is; to make bending faster by structuring that eliminates operator’s skill and provides everybody able to use easily and faultlessly by machine learning method and to learn by benefits from its previous bendings either made by their own or from other machines done and to provide to upgrade bending times.
Another purpose of the invention is; to able to learn through the other machines’ knowledge by connectimg to the big data backup and also to provide other machines benefit by recording their own knowledge.
Another purpose of the invention is; to upgrade the knowledge it learned from its own bendings within the own not only from big data and to provide machine learning without big data connection.
FIGURES
Figure 1 , there is a schematic view of the invention subject product.
USED REFERENCES
10. Big data backup
20. Machine
30. Internet provider DETAILED DESCRIPTION OF INVENTION
In this detailed description, the expression of the invention subject machine learning technic and big data backup is explained by means not to generate any limiting effects and intended to be only understood well of the subject. Figure 1 , there is a schematic view of the invention subject product.
On the system, there are one or more machines (20) that are connected to the big data backup (10) with an internet provider (30).
The working principle of the system is like this:
First bending material is placed on the machine (20) by choosing from the menus according to the specifications. Necessary adjustments are made and bending diameter is defined. After that step, if the machine (20) is connected to the big data backup (10), it continues bending by listing the values after searching according to the bending diameter and material dimensions which are indicated either at its own bending backups or at the big data backup. Bending is done by the machine until reached the bending diameter geometry and when it reached the bending diameter, the material spring back value is calculated by cnc attachment on the machine. It interprets by matching the results at its own backup or at the big data backup according to the ingenerate spring back values. It shortens the finish steps for improving the reaching time to the potential diameter and next spring back value by taking a bending which is done with the nearest values as a reference and it aims to make truer bendings in a faster time by trying to shorten the time and by trying to upgrade the bending at the backup one level more by increasing the pressure value to reach final bending value at 5 times that he reached at 6 times before.
The big data backup that is used in the system, is a kind of structuring like a cloud or it can be worked like a server, web-based backup.
Also, the system can produce machine learning without being connected to the big data, and because of being it learns just from the bendings that it is done by its own, upgrading the time of the bending process takes longer.
Also, the system can produce learning from other machines’ bendings which are inside of a closed cycle by a big data backup that will be built in a plant with a closed cycle.

Claims

CLAIMS 1. The innovation is directed to “the machine learning technic and big data backup” which allows for continuous upgrade of bending processes and getting faultless bending results in a shorter time by communicating machines with each other that are used for bending metal materials much easier and precisely through the backup point and includes;
Machine (20), which allows faultless and fast bending by choosing the best suitable program within definite tolerances and upgrading an old piece of knowledge with that program by checking the spring back measurement that the machine makes for verifying the final diameter after finishing the bending geometrically against the data that come after filtering data by matching data that the machine sensors generate and user data that obtained from old pieces of knowledge by means of internet provider (30) while bending data according to the geometry, and one big data backup (10) that makes backup again for the purpose of using again the data that generated.
PCT/TR2023/050729 2022-07-26 2023-07-25 Machine learning technic and big data backup WO2024025503A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
TR2022/011854 2022-07-26
TR2022/011854A TR2022011854A2 (en) 2022-07-26 2022-07-26 MACHINE LEARNING TECHNIQUE AND BIG DATA BACKUP

Publications (1)

Publication Number Publication Date
WO2024025503A1 true WO2024025503A1 (en) 2024-02-01

Family

ID=85161947

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/TR2023/050729 WO2024025503A1 (en) 2022-07-26 2023-07-25 Machine learning technic and big data backup

Country Status (2)

Country Link
TR (1) TR2022011854A2 (en)
WO (1) WO2024025503A1 (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110802142A (en) * 2019-09-30 2020-02-18 南京航空航天大学 Numerical control flexible roll bending machine for three-dimensional variable-curvature sheet metal
JP2021183354A (en) * 2020-05-22 2021-12-02 Ｊｆｅスチール株式会社 Method for predicting bend of shaped steel, method for manufacturing shaped steel, method for generating learned machine learning model and device for predicting curvature of shaped steel
TR2022003965A2 (en) * 2022-03-16 2022-04-21 Oezlem Kaya Topcuoglu MACHINE CNC CONVERSION APPARATUS CONSTRUCTION

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110802142A (en) * 2019-09-30 2020-02-18 南京航空航天大学 Numerical control flexible roll bending machine for three-dimensional variable-curvature sheet metal
JP2021183354A (en) * 2020-05-22 2021-12-02 Ｊｆｅスチール株式会社 Method for predicting bend of shaped steel, method for manufacturing shaped steel, method for generating learned machine learning model and device for predicting curvature of shaped steel
TR2022003965A2 (en) * 2022-03-16 2022-04-21 Oezlem Kaya Topcuoglu MACHINE CNC CONVERSION APPARATUS CONSTRUCTION

Also Published As

Publication number Publication date
TR2022011854A2 (en) 2022-10-21

Similar Documents

Publication Publication Date Title
JP6832327B2 (en) Data-driven method for automatic detection of anomalous workpieces during the production process
WO2017124211A1 (en) Method for measuring and evaluating gear precision
CN109472057A (en) Based on product processing quality prediction meanss and method across the implicit parameters memorizing of work step
Kleiner et al. Combined methods for the prediction of dynamic instabilities in sheet metal spinning
US10983501B2 (en) Tool-life prediction system and method thereof
Gauder et al. Development of an adaptive quality control loop in micro-production using machine learning, analytical gear simulation, and inline focus variation metrology for zero defect manufacturing
WO2024025503A1 (en) Machine learning technic and big data backup
US8346383B2 (en) Method for determining the machining quality of components, particularly for metal cutting by NC machines
CN108089457B (en) Process quality control method based on-line finite element simulation
Vorkov et al. Data-driven prediction of air bending
CN111805301B (en) Measuring device and measuring method
Heingärtner et al. Process control of forming processes to compensate temperature induced friction changes
WO2021197935A1 (en) Reduction of friction within a machine tool
Barbato et al. Method for automatic alignment recovery of a spur gear
CN104050277A (en) 3MAD-MMMD gross error detection method based on clustering analysis
EP3951523A1 (en) Method and apparatus for configuring processing parameters of production equipment, and computer-readable medium
CN100493751C (en) On-line testing apparatus for roller non-circularity and detecting method thereof
Weldgbrel et al. Experimental Investigation and Optimization of Cutting Parameters of Dry Turning EN-8 Steel for Enhanced Surface Finishing
Dallinger et al. Adaptive process control strategy for a two-step bending process
Shim Measurement of shape error for the optimal blank design of stamped part with 3 dimensional contour lines
EP4062245B1 (en) Methods and systems for workpiece quality control
EP4151330A1 (en) Steel pipe roundness prediction method, steel pipe roundness control method, steel pipe production method, method for generating steel pipe roundness prediction model, and steel pipe roundness prediction device
Campos et al. In-Situ Measurement and Dimensional Error Modeling: A Case Study in a Machine Tool Manufacturer
Wehmeyer et al. Investigation of the Process Limits for the Design of a Parameter-Based CAD Forming Tool Model
Hübner et al. Investigation of the Process Limits for the Design of a Parameter-Based CAD Forming Tool Model

Legal Events

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

Ref document number: 23847106

Country of ref document: EP

Kind code of ref document: A1