Virtual Reality (VR) offers the possibility of creating engaging and intensive training programs to enhance sports performance. Football, a widely popular sport worldwide, attracts millions of spectators. However, the physical demands and risks inherent in the game necessitate the creation of a training environment immune to external influences, minimizing sports-related injuries while enhancing player immersion. Existing training methods are affected by external factors like, weather, injuries, and space limitations, hindering effectiveness and skill development. To overcome these issues, in this research, enhanced soccer training simulation using Progressive Wasserstein GAN and Termite Life Cycle Optimization in Virtual Reality (PWGAN-TLCOA-ST-VRT) are proposed. This study begins by analyzing the fundamental features of computer-generated soccer training techniques and introduces the implementation of VRT in football training via PWGAN technology. The goal is to develop virtual systems resilient to external factors and provide theoretical support for the expansion of virtual football sports software. By overcoming constraints like, weather, player injuries, space limitations, and financial constraints, the proposed approach aims to enhance training effectiveness, teaching techniques, and football skill development. Software tools like Poser 8.0, 3Ds MAX and EON Studio are introduced for system development. Evaluation of the proposed PWGAN-TLCOA-ST-VRT method is examined using performance metrics including, precision, accuracy, F1-score, recall, specificity, error rate, computation time, and Receiver Operating Characteristic (ROC) analysis. The proposed PWGAN-TLCOA-ST-VRT technique achieves significant improvements compared to existing approaches like, Analysis of the application of virtual reality technology in football training (AI-ST-VRT), Cortex VR: Immersive analysis and training of cognitive executive functions of soccer players using VR and machine learning, Predict the value of football players using FIFA video game data and machine learning techniques (MLR-ST-VRT)