Predictive control algorithm based on improved distributed model

Abstract

The invention relates to the technical field of distributed calculation models, and discloses a predictive control algorithm based on an improved distributed model, and the method specifically comprises the following steps: S1, decomposing a whole system and a control problem into a plurality of small sub-problems, enabling the sub-problems to be correlated with each other, and enabling each controller to exchange information with other controllers; s2, structurally splitting the linear dynamic system into a plurality of subsystems, and splitting the linear dynamic system into a plurality of subsystems with coupling input; s3, when a coupling relationship with clear physical significance exists between the subsystems, performing system decomposition according to the coupling relationship; s4, when the centralized model of the system is unknown or indefinite, directly modeling the subsystems according to experience, and establishing a correlation model between the subsystems; the controllers share resources, each controller can know the action of the adjacent controller and set the characteristics of automatic learning, the next step of the controller can be predicted, the adaptability of a cooperation protocol can be improved, automatic learning can be applied to parameter identification, and the problem solving and decision-making capacity is improved.

Description

technical field [0001] The invention relates to the technical field of distributed computing models, in particular to a predictive control algorithm based on an improved distributed model. Background technique [0002] Distributed computing mode refers to a whole composed of multiple interconnected computers, which cooperate to perform a common or different task in the environment of a set of system software (distributed operating system or middleware), with the least dependence on Centralized control of processes, data and hardware. Distributed computing patterns are special in that they have the opposite characteristics of client / server streaming, or the pattern is a mix of peer-to-peer and client / server streaming. In this mode, traffic occurs primarily between the task manager and its computing devices (similar to the client / server model), or between computing devices (similar to the peer-to-peer mode). The behavioral characteristics of a specific mode depend on how the...

AI Technical Summary

Problems solved by technology

[0003] The distributed model is still in the research stage, and there are still some key or difficult problems to be solved, for example, the traditional allocation problem is inefficient, and whether the controllers can coordinate to deal with the problem