Predicting a Biomechanical Model Using Neural Networks for the Serving Skill of Iraqi National Volleyball Players

Abstract

The biomechanical-physiological model, structured according to its variables, involves the collection and analysis of components using advanced technologies. This study utilises cutting-edge equipment, including the Electromyography (EMG) device, which measures muscle electrical activity, and the Biosyn system, which provides data on body height and angular positioning through its integrated sensors. Among the most significant artificial intelligence techniques are artificial neural networks, which serve multiple purposes, including the recognition of individuals, scenarios, speech, images, fingerprints, handwriting, and patterns. Additionally, they facilitate system simulation and predictive modelling of performance. The application of artificial neural networks in learning offers valuable insights for athletes, coaches, and sports specialists, enabling them to assess performance levels, evaluate technical and skill-based execution, obtain results, and generalise findings within the field. Moreover, this model allows for future reference and refinement. This study aimed to develop a model based on artificial neural networks to enhance the analysis of volleyball serving skills, integrating physiological and biomechanical compatibility outcomes. It sought to predict the most critical physiological and biomechanical variables influencing the serve performance of the Iraqi national volleyball team. The research population comprised 18 national team players, from whom a sample of 8 players (44.44% of the total) was selected. Each participant was given 10 attempts in the serve test, resulting in a total of 80 observations. The findings indicate that volleyball players can be categorised in terms of serving skills based on physiological and biomechanical variables.