The Industrial Internet of Things (IIoT) has fundamentally changed and revolutionized industry 4.0 and global production by creating a richer ecosystem of intelligent, networked devices and opening up new avenues for digital innovation. Conversely, IIoT is a remarkable and possible target for cyber attackers due to its widely distributed nature, Industrial 5G, underlying IoT sensing devices, IT/OT convergence, Edge Computing, and Time Sensitive Networking. Sophisticated and multi variant bot attacks are deemed disastrous for IIoT connections. Furthermore, botnet attack detection is a highly intricate and precise process. Therefore, it is imperative that IIoT botnets be detected quickly and effectively. Our suggestion is a hybrid intelligent Deep Learning (DL) enabled system to protect IIoT infrastructure against highly skilled and deadly multi-variant botnet attacks. Using the most recent dataset available, standard and extended performance evaluation measures, and the most recent deep learning benchmark methods, the suggested mechanism has been thoroughly examined. Additionally, we cross-validate our findings to provide a comprehensive picture of overall performance. With a 99.94% detection rate, the suggested approaches outperform in precisely recognizing multi-variant complex bot attacks. Furthermore, the time of 0.066(ms) achieved by our suggested method demonstrates encouraging outcomes in terms of speed efficiency.