SFLA-SVM-based digital water island online agent adding control method

Abstract

The invention discloses an SFLA-SVM-based digital water island online agent adding control method. The method comprises the steps of obtaining an inflow monitoring index and determining a target function and a decision variable of an SFLA-SVM agent adding model; building an SVM model of the agent adding amount and the monitoring index; applying a chaos SFLA to the SVM model to obtain an optimal solution of each parameter; using the optimal solutions for learning training of the SVM model, and building the SFLA-SVM agent adding model; and inputting the detected inflow monitoring index into the SFLA-SVM agent adding model, and determining the optimal agent adding amount of a current time. The method overcomes bad effect on running process caused by time-varying characteristics of the water quality of inflow; the optimal agent adding amount is updated in real time, so that online control of agent adding process is achieved, waste of agent and labor and running cost are reduced, and the quality standard of outflow is satisfied; the calculation speed and the precision of the agent adding model are improved, so that local optical problem is avoided, non-linear problems are solved, and explanation and prediction capabilities of the model are enhanced.

Description

technical field [0001] The invention relates to the technical field of water treatment in thermal power plants, in particular to an SFLA-SVM-based digital water island online dosing control method. Background technique [0002] In modern thermal power plants above 300MW, more than 90% have realized the DCS automatic control system, but the water treatment system of the power plant has basically realized the DCS automatic control system, but the input rate of the intelligent optimization analysis of the system is very low. Ammonia treatment, ammonia addition to feed water, hydrazine addition treatment, chlorine addition treatment, phosphate addition treatment in bubble furnace, etc., due to the large difference in the water volume and water quality components of incoming water, and various reactions between different components in the water, especially The neutralization process of strong acid and strong base has complex nonlinear characteristics. At present, the water treat...

AI Technical Summary

Problems solved by technology

[0006] In order to solve the above problems, the object of the present invention is to provide an SFLA-SVM-based online dosing control method for digital water islands, which solves the problems that the traditional dosing methods of water islands in thermal power plants are difficult to accurately determine the quantity, and the online update model cannot be realized. On-line control of digital water island dosing process

Images