The relevance of the study is due to the importance and necessity of unifying the construction and use of intelligent decision support systems for managing complex industrial facilities and systems. The aim of the study is to substantiate the unified approach to managing knowledge bases of various configurations and develop unified mathematical models of operations on ontology elements. The method for managing the evolution of ontologies of professional fields, based on the unification of the structural-logical model of metaknowledge representation, is proposed. The method of unification of the structural-logical model of evolution of ontology incorporation is developed. Formal linguistic models are developed, the similarity of forms of knowledge representation and evolutionary inheritance within the general ontology incorporation are proved. For the synthesis of the model of incorporation of the evolutionary inheritance of ontologies, the subtasks of the development of models of the evolutionary inheritance of concepts, graphs and ontologies of KB levels are solved. The model provides an opportunity of a single approach to the interpretation of the interaction structures of concepts for all KB levels. The generalized model of the signal graph of the KB structure levels is developed. The model includes the atomic concept, signal, node potential, node activity, threshold of node sensitivity to the input signal. A set of formal models of basic operations on the signal graph of the KB necessary for the interpretation and computing of knowledge forms is developed. The metarule syntax and the formal-linguistic basis are developed. Formalisms of the labeling parameter and labeling function of the KB signal graph are introduced. Labeling models are introduced into the general model of the signal graph of the KB. Possibilities of applying the developed models of the signal graph of the knowledge base to various professional areas are investigated. It is shown that the proposed metaknowledge models do not depend on forms of representation and formalisms of professional ontologies. This allows the use of a single knowledge management mechanism in any intelligent decision support systems. The method of effective dynamic management of the structure of all KB levels and inference process depending on the input parameters of the intelligent system is proposed