Research On Power Plant Energy Saving Based On Stochastic Distributed Control

With the demand for energy in human society is growing, The conflict between power generation and demand has become more prominent. To optimize the control strategy, improve the efficiency of power generation is a very effective way to solve the contradiction between supply and demand of electricity. Because China is currently still in hydropower and thermal power-based, so this paper to hydropower turbine governor systems and thermal power plant combustion control system as the background, optimizing its control were studied. Combustion control systems for thermal power plants, the establisliment of furnace temperature distribution model by B-spline is first, and then uses BP neural network as a controller, the difference between the B-spline weights of the system output temperature distribution and the B-spline weights of system settings distribution will be as the neural network input. The control target is to minimize the difference between two B-spline weights, then making the output furnace temperature distribution to be right.Finally, the simulation results were compared with the traditional control methods, speed and stability of the algorithm has been verified. The hydropower turbine governor control system is vulnerable to fluctuations in the bases position and the influent flow, and these are not necessarily Gaussian random disturbance. In order to minimize disturbance to the system, this paper studies for the turbine governor cascade control system introduced randomly distributed controller as the main controller, the corresponding performance index is the weighted sum of the information entropy, system error and control input, through minimize the performance index to calculate the system control law, the goal is to ensure that the shape of the probability density function of tracking error can be sharp and narrow.