Con esta herramienta te facilitamos un acceso a todas las ofertas y demandas de tecnología europeas y a búsquedas de socios para participar en propuestas europeas de I+D publicadas en la red Enterprise Europe Network, pudiendo filtrar los resultados para facilitar las búsquedas más acordes con tus necesidades.

¿Quieres recibir estos listados de oportunidades de colaboración en tu correo de forma periódica y personalizada? Date de alta en nuestro Boletín

Para optimizar los resultados de la búsqueda, se recomienda utilizar términos en inglés.

Nuevo software basado en redes neuronales para calcular el coeficiente de fricción de ensambles prensados

Resumen

Tipo:
Oferta Tecnológica
Referencia:
TORS20161013006
Publicado:
31/10/2016
Caducidad:
31/10/2017
Resumen:
Una empresa serbia especializada en el área de fabricación y análisis de ensambles de máquinas ha desarrollado un software basado en redes neuronales para calcular el coeficiente de fricción de ensambles prensados. En comparación con otras soluciones, el error que aparece en el cálculo del coeficiente de fricción es un 30% menor. Otras ventajas son la reducción de las dimensiones de los componentes, la disminución de los costes y el uso de materiales más económicos. La empresa busca socios en el campo de fabricación de ensambles mecánicos prensados con el fin de cooperar en el desarrollo e implementación del software en sistemas existentes.

Details

Tittle:
Innovative Neural network-based software for calculating the coefficient of friction in pressed assemblies
Summary:
Innovative company from Serbia, specialised in the area of production and testing of machine assemblies, has developed
network based software for the coefficient of friction in pressed assemblies. In comparison to other solution an error that appears in the determination of the coefficient of friction is 30% lower. The authors are looking for partners in the field of pressed mechanical assemblies manufacturing for cooperation on development and software implementation for existing systems.
Description:
Company has five year of experience research and development of neural network based software. These systems have been successfully applied in many private and public companies in Serbia and Bosnia -Y- Herzegovina.

The process of pressed assembly design is in a causal relation with the evaluation of friction coefficient. The evaluation of this parameter can be carried out by using various methods. The algorithm for friction coefficient determination is created. Analyses are based on roughness and hardness of tactile surfaces, with the implementation of a hybrid system based on neural networks. In this case, the ratio between the minimum value and the maximum value is 1:17, and it depends on the designer´s decision.

A pressed assembly should provide the transfer of a given load on the direct contact of joined surfaces. There should be
a sufficient pressure between these surfaces. A load can be a force or torque. The shape of the pressed assembly is generally tube-like or cone-like. The coefficient of friction is the most important factor for the determination of minimal force of pressed assemblies. An incorrect coefficient value can produce serious problems. The minimal force of pressed assemblies is determined accordingly. In other words, the calculation of pressed assemblies is, in fact, the solution to the problem of friction coefficient calculation.

The friction coefficient value depends on different factors, such as: existence or nonexistence of a boundary layer of lubricant during the assembly, external normal load on the friction area, value of contact pressure, overleaf value of two joint parts, mechanical characteristics of two joint parts, roughness of contact surfaces, hardness of contact surfaces, cleanliness of contact surfaces, speed of assembling the parts.

A valid neural network implemented in software is being used to predict the output values for a particular set of input data, with two assumptions:
1.The input values must be different from those used for the training or the validation of the model,
2.The input values must be in the range of training and validation data.
Advantages and Innovations:
The major innovation of network based software for the coefficient of friction in pressed assemblies is implementation of
neural networks in the determination of friction coefficient.

The major advantages of network based software for the coefficient of friction in pressed assemblies is based on the
lowest known level of error. Error in the range of 8%, and that is 30 percent lower than the quality of similar products
on the market. Secondary advantages are:
·reduce bad influence of subjective evaluation of the coefficient of friction,
·apply good engineering experience in the process of calculation,
·increase the reliability of calculated assembly,
·reduce the dimensions of the components,
·get low costs with exact calculation,
·use less expensive materials.
Stage of Development:
Already on the market
IPs:
Secret Know-how

Partner sought

Type and Role of Partner Sought:
The preferred partners are industrial SMEs involved in the manufacturing of electronic circuits, components and equipment. The Serbian SME offers commercial agreement with technical assistance, as well as joint venture agreement. For joint venture the cooperation could be focused on development and software implementation for the coefficient of friction in pressed assemblies.

Client

Type and Size of Client:
Industry SME <= 10
Already Engaged in Trans-National Cooperation:
Si
Languages Spoken:
English

Keywords

Technology Keywords:
01003020 Building Automation Software
01003006 Computer Software