The Industrial Internet of Things has grown significantly in recent years. While implementing industrial digitalization, automation, and intelligence introduced a slew of cyber risks, the complex and varied industrial Internet of Things environment provided a new attack surface for network attackers. As a result, conventional intrusion detection technology cannot satisfy the network threat discovery requirements in today’s Industrial Internet of Things environment. An intrusion detection system (IDS) is a critical component of network security protection because it enables the system to detect network intrusions efficiently. However, in recent years, as the operating environment and structure of the Industrial Internet of Things have changed, traditional intrusion detection models (such as intrusion detection models based on simple machine learning) have been unable to provide adaptive detection, response, and defence against complex network attacks. Machine learning nowadays is a developing topic; its applications are wide. We can forecast the future through machine learning and classify the right class. Unsupervised solutions do reduce computational complexities and manual support for labeling data but current unsupervised solutions do not consider spatio-temporal correlations in traffic data. To address this, in the existing basic convolutional autoencoder methods are presented. However, the existing autoencoders have lot of issues, such as Insufficient training data, training the wrong use case, too lossy, imperfect decoding, misunderstanding important variables, better alternatives, algorithms become too specialized, bottleneck layer is too narrow. So, to overcome these drawbacks, this work presented the deep learning convolutional network-based intrusion detection framework. The simulations will conduct on UNSW-NB15 dataset, which contains attack and normal classes of data. Initially, the dataset preprocessing operation will perform to remove the missing symbols, unknown characters. Then, the deep learning model applied to perform the training of dataset, which also predicts the normal and attack class from test data.