The process industry is the pillar of the national economy and a key force in promoting social development. The complexity of its production processes and the diversity of working conditions lead to problems such as variable coupling, dynamic time variation, and data distribution mismatches. Especially when faced with massive amounts of data, it is crucial to effectively learn new knowledge while retaining existing knowledge. This helps avoid the dilemma of catastrophic forgetting. To address the above problems, this paper proposes a dynamic domain adaptation regression method for multiple working conditions based on continual learning. Construct a latent variable space to describe the dynamic regression relationship between process variables and quality variables. Introduce domain adaptation regularization terms for latent variable reconstruction and a dynamic continual learning regularization term. It is demonstrated on industrial data sets that this method can maintain high prediction accuracy and low forgetting degrees under multiple working conditions.