In current article an improved salp swarm algorithm (SSA) has been suggested. Likewise, the adjustable scaling-component is locked in to manage salp’s situation in time of the hunt system to limit arbitrary development salps. The upgraded salp swarm method combines random objection-based learning, multiple management, and simulated hardening in swarm brilliant-based metaheuristic salp swarm method, which rises the exploration and exploitation of the primary salp swarm method. To exhibit viability of the improved SSA, a bunch of multichannel test capabilities are locked in. A partially decentralized combined fuzzy proportional integral derivative (PD-CFPID) regulator is planned for integrated grid. A partially decentralized robust control plan is introduced to deal with a class of multi-inputs multi outputs (MIMO) non-square, linear framework different systems. Partially decentralized command is a control composition that lies between a completely decentralized structure and a wholly centralized one, and has the primacy of achieving comparable performance as an entirely centralized regulator but with simpler shape. The proposed flexible partially decoupled command scheme works in a straightforward and systematic way. Simulations show that the proposed partially decentralized control performs well for the nonlinear system, nearly equivalent to the multivariable control, and they achieve better performance indexes. Simulation outcomes appear that the stated method is simple and can reach better performance. The results of the PD-CFPID regulator are compare among CFPID and PID as many circumstances of the presented control outlook.